Portland State UniversityPDXScholar
Dissertations and Theses Dissertations and Theses
1996
A Methodology for Regional Seismic Damage Assessment andRetrofit Planning for Existing BuildingsThomas C. McCormackPortland State University
Let us know how access to this document benefits you.Follow this and additional works at: http://pdxscholar.library.pdx.edu/open_access_etds
This Dissertation is brought to you for free and open access. It has been accepted for inclusion in Dissertations and Theses by an authorizedadministrator of PDXScholar. For more information, please contact [email protected].
Recommended CitationMcCormack, Thomas C., "A Methodology for Regional Seismic Damage Assessment and Retrofit Planning for Existing Buildings"(1996). Dissertations and Theses. Paper 1239.
10.15760/etd.1238
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A methodology for regional seismic damage assessment and retrofit planning for existing buildingsMcCormack, Thomas CraigProQuest Dissertations and Theses; 1996; ProQuestpg. n/a
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in e<illal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 3Bn6I-4 100 800/S21-06oo
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in e<illal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313n61-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A METHODOLOGY FOR REGIONAL SEISMIC DAMAGE ASSESSMENT
AND RETROFIT PLANNING FOR EXISTING BUILDINGS
by
THOMAS McCORMACK
A dissertation submitted in partial £ul£illment o£ the requirements £or the degree o£
DOCTOR OF PHILOSOPHY in
SYSTEMS SCIENCE:CIVIL ENGINEERING
Portland State University
1996
A METHODOLOGY FOR REGIONAL SEISMIC DAMAGE ASSESSMENT
AND RETROFIT PLANNING FOR EXISTING BUILDINGS
by
THOMAS McCORMACK
A dissertation submitted in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY in
SYSTEMS SCIENCE:CIVIL ENGINEERING
Portland State University
1996
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 9635653
UMI Microform 9635653 Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103
UMI Number: 9635653
UMI Microform 9635653 Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DISSERTATION APPROVAL
The abstract and dissertation of Thomas McCormack for the
Doctor of Philosophy in Systems Science:Civil Engineering
were presented on June 14, 1996, and accepted by the
dissertation committee
COMMITTEE APPROVALS:
Wendelin H. Mueller
Martin Zwick
Scott F. Burns Representative of the Office of Graduate Studies
DOCTORAL PROGRAM APPROVAL: Beatrice T. Oshika, Director Systems Science Ph.D. Program
* * * * * * * * * * * * * * * * * * * * * * * * * * * * *
ACCEPTED FOR PORTLAND STATE UNIVERSITY BY THE LIBRARY
by
DISSERTATION APPROVAL
The abstract and dissertation of Thomas McCormack for the
Doctor of Philosophy in Systems Science:Civil Engineering
were presented on June 14, 1996, and accepted by the
dissertation committee
COMMITTEE APPROVALS:
Wendelin H. Mueller
Martin Zwick
Scott F. Burns Representative of the Office of Graduate Studies
DOCTORAL PROGRAM APPROVAL: Beatrice T. Oshika, Director Systems Science Ph.D. Program
* * * * * * * * * * * * * * * * * * * * * * * * * * * * *
ACCEPTED FOR PORTLAND STATE UNIVERSITY BY THE LIBRARY
by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
An abstract of the dissertation of Thomas McCormack for
the Doctor of Philosophy in Systems Science:Civil
Engineering presented June 14, 1996.
Title: A Methodology for Regional Seismic Damage
Assessment and Retrofit Planning for
Existing Buildings
Recent geologic research has shown that earthquakes
more destructive than formerly expected are likely to occur
in the Pacific Northwest. To mitigate catastrophic loss,
planners are gathering information to make decisions on
implementing regional seismic retrofit programs.
This research develops a model to estimate regional
earthquake losses for existing buildings, and determine
optimal retrofit priorities and budgets.
Fragility curves are developed to provide earthquake
damage estimates for a range of seismic intensities. The
published earthquake damage estimates of a large group of
prominent earthquake engineering experts are extended to
include the combined effect of structure type, earthquake-
sensitive variations in building design, site-specific soil
conditions, and local seismic design practice.
ABSTRACT
An abstract of the dissertation of Thomas McCormaCK for
the Doctor of Philosophy in Systems Science:Civil
Engineering presented June 14, 1996.
Title: A Methodology for Regional Seismic Damage
Assessment and Retrofit Planning for
Existing Buildings
Recent geologic research has shown that earthquakes
more destructive than formerly expected are likely to occur
in the Pacific Northwest. To mitigate catastrophic loss,
planners are gathering information to make decisions on
implementing regional seismic retrofit programs.
This research develops a model to estimate regional
earthquake losses for existing buildings, and determine
optimal retrofit priorities and budgets.
Fragility curves are developed to provide earthquake
damage estimates for a range of seismic intensities. The
published earthquake damage estimates of a large group of
prominent earthquake engineering experts are extended to
include the combined effect of structure type, earthquake-
sensitive variations in building deSign, site-specific soil
conditions, and local seismic design practice.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
Building inventory data from a rapid visual screening
survey of individual buildings form the basis for modeling
structural variations. EarthquaKe Hazard Haps are the basis
of modeling the effect on building damage of ground motion
amplification, soil liquefaction, and slope instability.
Published retrofit effectiveness estimates and
retrofit cost data are used to estimate post-retrofit
damage avoided, lives saved, and retrofit cost. A Building
Classification System is formulated to aggregate
buildings with similar retrofit benefit magnitudes.
A cost-benefit analysis is used as the basis for a
retrofit prioritization and efficiency analysis, to
establish the cut-off point for an optimal retrofit
program. Results from an Expected Value and a Scenario
EarthquaKe Event are compared.
Regional EarthquaKe Loss and Retrofit Analysis
Program (REAL-RAP) software was developed, and used to maKe
a loss estimate for more than 7, 500 buildings inventoried
in the 1993 Portland Seismic Hazards Survey. One hundred
percent of the loss of life is attributed to only
10-percent of the buildings.
A retrofit analysis is made for a Design Basis
EarthquaKe. Twelve-percent of the building inventory was
identified for the optimal retrofit program, wherein
98-percent of the loss of life is avoided at less than
2
Building inventory data from a rapid visual screening
survey of individual buildings form the basis for modeling
structural variations. EarthquaKe Hazard Haps are the basis
of modeling the effect on building damage of ground motion
amp 1 ification, soil liquefaction, and slope insta.l:>il i ty.
Published retrofit effectiveness estimates and
retrofit cost data are used to estimate post-retrofit
damage avoided, lives saved, and retrofit cost. A Building
Classification System is formulated to aggregate
buildings with similar retrofit benefit magnitudes.
A cost-benefit analysis is used as the basis for a
retrofit prioritization and efficiency analysis, to
esta.l:>lish the cut-off point for an optimal retrofit
program. Results from an Expected Value and a Scenario
EarthquaKe Event are compared.
Regional EarthquaKe Loss and Retrofit Analysis
Program (REAL-RAP) software was developed, and used to maKe
a loss estimate for more than 7,500 buildings inventoried
in the 1993 Portland Seismic Hazards Survey. One hundred
percent of the loss of life is attributed to only
la-percent of the buildings.
A retrofit analysis is made for a Design Basis
EarthquaKe. Twelve-percent of the building inventory was
identified for the optimal retrofit program, wherein
98-percent of the loss of life is avoided at less than
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
one-quarter the cost of retrofitting all the buildings.
An alternate optimal retrofit program was determined
using an Expected Value Analysis. Most of the buildings in
the Design Basis Earthquake optimal retrofit program are
also contained in the alternate program.
3
one-quarter the cost of retrofitting all the buildings.
An alternate optimal retrofit program was determined
using an Expected Value Analysis. Most of the buildings in
the Design Basis Earthquake optimal retrofit program are
also contained in the alternate program.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
This dissertation is dedicated to my wife,
susan K. McCormacK, for her love, unselfish support, and
faith in me, without which the project could not have been
completed.
DEDICATION
This dissertation is dedicated to my wife,
Susan K. McCormacK, for her love, unselfish support, and
faith in me, without which the project could not have been
compi eted.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
Dr. Franz H. Rad, Chair of Civil Engineering,
Portland State University, originally conceptualized the
integration of the ATC-13 damage data and the ATC-21
inventory scoring system. The present research fleshes
out this concept, and Dr. Rad deserves special thanks for
inspiring and supervising this project.
Thanks also to Dr. Gerry Uba of METRO, who
administered the Portland Seismic Hazards Survey, which
provided the building inventory data for this research.
Thanks to the members of my committee, for providing
helpful review and commentary on the dissertation:
Dr. Trevor Smith, Dr. Wendelin Mueller, Dr. Martin Zwick,
and Dr. Scott Burns.
Thanks also to Richard McKenzie, PSU Civil
Engineering student, for his assistance on the regressions
and the graphics presentations.
ACKNOWLEDGEMENTS
Dr. Franz N. Rad, Chair of Civil Engineering,
Portland State University, originally conceptualized the
integration of the ATC-13 damage data and the ATC-21
inventory scoring system. The present research fleshes
out this concept, and Dr. Rad deserves speCial thanks for
inspiring and supervising this project.
Thanks also to Dr. Gerry Uba of METRO, who
administered the Portland Seismic Hazards Survey, which
provided the building inventory data for this research.
Thanks to the members of my committee, for providing
helpful review and commentary on the dissertation:
Dr. Trevor Smith, Dr. Wendelin Mueller, Dr. Martin Zwick,
and Dr. Scott Burns.
Thanks also to Richard McKenzie, PSU Civil
Engineering student, for his assistance on the regressions
and the graphics presentations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
PAGE
ACKNOWLEDGEMENTS iii
LIST OF TABLES . Xi
LIST OF FIGURES xviii
CHAPTER
I
II
INTRODUCTION . . . . .
Problem Definition
General Nature of the Problem
Systems Approach System Defined Multi-Disciplinary Aspects Complexity Uncertainty Decision Theoretic Considerations
LITERATURE REVIEW
ATC-13: Earthquake Damage Evaluation
1
1
14
Data for California . . . . . . . 14
Earthquake ShaKing Characterization Facility Classifications Physical Damage Caused by
Ground Shaking Death and Injury Estimates Significance and Limitations
ATC-21: Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook. . . . .
Rapid Screening Procedure Seismicity Scoring System Basis of the Scoring System Significance and Limitations
26
TABLE OF CONTENTS
PAGE
ACKNOWLEDGEMENTS iii
LIST OF TABLES . xi
LIST OF FIGURES xviii
CHAPTER
I
II
INTRODUCTION . . . . .
Problem Definition
General Nature of the Problem
Systems Approach System Defined Multi-Disciplinary Aspects Complexity Uncertainty Decision Theoretic Considerations
LITERATURE REVIEW
ATC-13: Earthquake Damage Evaluation
1
1
Data for California . . . . . . . 14
Earthquake Shaking Characterization Facility Classifications Physical Damage Caused by
Ground Shaking Death and Injury Estimates Significance and Limitations
ATC-21: Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook . . . .
Rapid Screening Procedure Seismicity Scoring System Basis of the Scoring System Significance and Limitations
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
III
Portland Seismic Hazards Survey
Building Inventory Database Range of Estimated Humber
of People Post-BenchmarK Year Significance and Limitations
State of Oregon Seismic Design Mapping . . . . . . . .
EarthquaKe Sources Seismic Hazard Analysis Limitations and Significance
EarthquaKe Hazard Maps of the Portland Quadrangle . . . .
Liquefaction Susceptibility Map Lateral Spread Displacement Map Ground Motion Amplification Map Dynamic Slope Instability Map Relative EarthquaKe Hazard Map Significance and Limitations
FEMA-227: A Benefit-Cost Model for Seismic Rehabilitation of Buildings . . . . .
Expected Het Present Value Model without the Value of Life
Expected Het Present Value Model with the Value of Life
Benefit/Cost Ratios Significance and Limitations
FEMA-156: Typical Costs for Seismic Rehabilitation of Existing
v
36
41
58
Buildings, Second Edition 68
Significance and Limitations
FOCUS OF RESEARCH
Limitations of Current Practice
EarthquaKe Loss Modeling Retrofit Decision Analysis
Outline of the Research
74
74
79
III
Portland Seismic Hazards Survey
Building Inventory Database Range of Estimated Humber
of People Post-BenchmarK Year Significance and Limitations
State of Oregon Seismic Design Happing ....... .
EarthquaKe Sources Seismic Hazard Analysis Limitations and Significance
EarthquaKe Hazard Haps of the Portland Quadrangle . . . .
Liquefaction Susceptibility Hap Lateral Spread Displacement Hap Ground Hotion Amplification Hap Dynamic Slope Instability Hap Relative EarthquaKe Hazard Hap Significance and Limitations
FEMA-227: A Benefit-Cost Hodel for Seismic Rehabilitation of Buildings . . . . .
Expected Ret Present Value Hodel without the Value of Life
Expected Ret Present Value Hodel with the Value of Life
Benefit/Cost Ratios Significance and Limitations
FEMA-156: Typical Costs for Seismic Rehabilitation of Existing
v
36
41
58
Buildings, Second Edition 68
Significance and Limitations
FOCUS OF RESEARCH
Limitations of Current Practice
EarthquaKe Loss Hodeling Retrofit Decision Analysis
Outline of the Research
74
74
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IV DEVELOPMENT OF AN EARTHQUAKE LOSS MODEL FOR EXISTING BUILDINGS
Formats for Expressing Seismic Vulnerability ....
Ground Motion Characterization Damage Function Format Fragility Curve Format Significance and Limitations
The Basis of the ATC-21 Scoring System . . . . . . . . . .
The Basic Structural Hazard Score Extension to Non-California Bldgs Performance Modification Factors
Development of a Technique to Derive Damage Factor from Structural Score . . .
Development of Fragility Curves Based
vi
84
85
89
100
on Performance Modifiers . . . . . . 110
Modified Basic Structural Hazard Score
Mean Damage Factor vs. E(PMF) Fragility Curves Based on E(PMF) Application of Fragility Curves
to NEHRP Areas 3,4
Development of a Soil Effects Model . . . . . . . .
The ATC-21 Soil Profile Modifer Amplification Factor Liquefaction Factor Slope Factor Soil Modification Factor
Comparison to Goettel and Horner Study . . . . . . . .
Earthquake Loss Estimation .
Dollar Value of Building Damage Loss of Life and Serious Injuries
131
138
142
IV DEVELOPMENT OF AN EARTHQUAKE LOSS MODEL FOR EXISTING BUILDINGS
Formats for Expressing Seismic Vulnerability . . . .
Ground Motion Characterization Damage Function Format Fragility Curve Format Significance and Limitations
The Basis of the ATC-21 Scoring System.. ....... .
The Basic Structural Hazard Score Extension to Non-California Bldgs Performance Modification Factors
Development of a Technique to Derive Damage Factor from Structural Score . . .
Development of Fragility Curves Based
vi
84
85
89
100
on Performance Modifiers. . . . .. 110
Modified Basic Structural Hazard Score
Mean Damage Factor vs. E(PMF) Fragility Curves Based on E(PMF) Application of Fragility Curves
to NEHRP Areas 3,4
Development of a Soil Effects Model . . . . . . . .
The ATC-21 Soil Profile Modifer Amplification Factor Liquefaction Factor Slope Factor Soil Modification Factor
Comparison to Goettel and Horner Study ........ .
Earthquake Loss Estimation .
Dollar Value of Building Damage Loss of Life and Serious Injuries
131
138
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
v
VI
VII
DEVELOPMENT OF A RETROFIT DIRECT BENEFIT AND COST MODEL . .
Retrofit Effectiveness Model
Retrofit Direct Benefit Model
Retrofit Cost Model
DEVELOPMENT OF A REGIONAL RETROFIT PLANHIHG METHODOLOGY . . . . . . .
Building Classification System
Retrofit Cost-Benefit Analysis
Dollar Value of Life-Safety Benefits Scenario Earthquake Event Analysis Expected Value Analysis
Retrofit Prioritization and System Optimization
DEVELOPMENT OF REGIONAL EARTHQUAKE LOSS AHD RETROFIT ANALYSIS SOFTWARE
vii
1.1!-8
1.1!-9
159
160
16.1!-
16.1!-
167
171
173
Earthquake Loss Estimation Modules . 173
DAM-FACT Module DAMAGE-$ Module LIFELOSS Module DAM-SUM Module REGIOH-L Module
Retrofit Direct Benefit and Cost Modules . . . . . . .
RETRO-DF Module RETRO-$D Module RETRO-LL Module RFBEH-u Module
Regional Retrofit Classification, Prioritization, and Optimization
180
Modules . . . . . . . . . . . 185
BCS-RBEH Module RETRO-BC Module EXP-VAL Module SORT-BC Module SYST-EFF Module
V
VI
VII
DEVELOPMENT OF A RETROFIT DIRECT BENEFIT AND COST MODEL . .
Retrofit Effectiveness Model
Retrofit Direct Benefit Model
Retrofit Cost Model
DEVELOPMENT OF A REGIONAL RETROFIT PLANHIHG METHODOLOGY. . . . . . .
Building Classification System
Retrofit Cost-Benefit Analysis
Dollar Value of Life-Safety Benefits Scenario Earthquake Event Analysis Expected Value Analysis
Retrofit Prioritization and System Optimization
DEVELOPMEHT OF REGIONAL EARTHQUAKE LOSS AHD RETROFIT ANALYSIS SOFTWARE
vii
1.1!-8
1.1!-9
159
160
16.1!-
16.1!-
167
171
173
Earthquake Loss Estimation Modules . 173
DAM-FACT Module DAMAGE-$ Module LIFELOSS Module DAM-SUM Module REGIOH-L Module
Retrofit Direct Benefit and Cost Modules . . . . . . .
RETRO-DF Module RETRO-$D Module RETRO-LL Module RFBEH-II* Module
Regional Retrofit Classification, Prioritization, and Optimization
180
Modules ........... 185
BCS-RBEH Module RETRO-BC Module EXP-VAL Module SORT-BC Module SYST-EFF Module
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VIII
IX
EARTHQU.AX:E LOSS ESTIMATE OF THE PORTLAND, OREGON QUADRANGLE
Non-Residential Building Stock Data ...
Building Type, Area, and Structural Score
Soil Modification Factor Occupancy Type and Humber
of Occupants
Loss Estimates for Four Scenario Earthquake Events . . . . .
Loss Estimates for the Design Basis Earthquake . . . .
Damage and Loss of Life by Structure Type
Total Dollar Value of Loss Loss of Life by Structural Score Loss of Life by Soil Modification
Factor Loss of Life by Humber of People Damage State by Building Type Comparison to Kobe City Earthquake
of 1994.-Losses Attributable to the
Performance Modifiers
RETROFIT ANALYSIS OF THE PORTLAND, OREGON QUADRANGLE
Post-Retrofit Earthquake Loss Estimates for Four Scenario Earthquake Events Following
viii
189
189
193
198
206
Full Retrofit . . . . . . . 207
Development of an Optimal Retrofit Program for the Design Basis Earthquake . . . . . . . . . 212
Total Value of Retrofit Benefits Retrofit Prioritization by
Building Class System Efficiency Parameters of the Optimal
Retrofit Program
VIII
IX
EARTHQU.AX:E LOSS ESTIMATE OF THE PORTLAND, OREGON QUADRANGLE
Non-Residential Building Stock Data ...
Building Type, Area, and Structural Score
Soil Modification Factor Occupancy Type and Number
of Occupants
Loss Estimates for Four Scenario Earthquake Events . . . . .
Loss Estimates for the Design Basis Earthquake ....
Damage and Loss of Life by Structure Type
Total Dollar Value of Loss Loss of Life by Structural Score Loss of Life by Soil Modification
Factor Loss of Life by Number of People Damage State by Building Type Comparison to Kobe City Earthquake
of 1994.-Losses Attributable to the
Performance Modifiers
RETROFIT ANALYSIS OF THE PORTLAND, OREGON QUADRANGLE
Post-Retrofit Earthquake Loss Estimates for Four Scenario Earthquake Events Following
viii
189
189
193
198
206
Full Retrofit . . . . . . . 207
Development of an Optimal Retrofit Program for the Design Basis Earthquake. . . . . . . .. 212
Total Value of Retrofit Benefits Retrofit Prioritization by
Building Class System Efficiency Parameters of the Optimal
Retrofit Program
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X
Comparison of Optimal Retrofit to Full Retrofit
Building Class Indices in the Optimal Retrofit
Sensitivity Analysis
Development of an Optimal Retrofit Program Based on an Expected
ix
Value Analysis . . . . . . . . . 228
Present Value of Full Retrofit Retrofit Prioritization by
Building Class System Efficiency Parameters o£ the Optimal
Retrofit Program Building Class Indices in the
Optimal Retrofit Comparison of Expected Value
and Design Basis Retrofits
SUMMARY AND COHCLUSIOHS
Earthquake Loss and Retrofit Modeling . . . . . ...
Regional Retrofit Planning Model
Computer Software Development
Earthquake Loss Estimate of the Portland Quadrangle
Retrofit Analysis of the Portland Quadrangle
Limitations in Applying this Research . . . .
Recommendations for Further Research . . . . . . . . .
240
240
241
242
242
244
247
248
REFERENCES CITED . . . . . . . . . . . , . . . . . . 251
x
Comparison of Optimal Retrofit to Full Retrofit
Building Class Indices in the Optimal Retrofit
Sensitivity Analysis
Development of an Optimal Retrofit Program Based on an Expected
ix
Value Analysis . . . . . . . . . 228
Present Value of Full Retrofit Retrofit Prioritization by
Building Class System Efficiency Parameters of the Optimal
Retrofit Program Building Class Indices in the
Optimal Retrofit Comparison of Expected Value
and Design Basis Retrofits
SUMMARY AND CONCLUSIONS
Earthquake Loss and Retrofit Modeling . . . . . . . .
Regional Retrofit Planning Model
Computer Software Development
Earthquake Loss Estimate of the Portland Quadrangle
Retrofit Analysis of the Portland Quadrangle
Limitations in Applying this Research . . . .
Recommendations for Further Research. . . . . . . . .
240
240
241
242
242
244
247
248
REFERENCES CITED. . . . . . . . . . . . . . . . .. 251
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X
APPENDICES
A POWER CURVE COEFFICIENTS FROM MEAN DAMAGE FACTOR VS. STRUCTURAL SCORE REGRESSIONS 254
B HALF-STEP FRAGILITY CURVES . . . . . . . 261
C BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT FOR A DESIGN BASIS EARTHQUAKE . 274
D BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT BASED OH AH EXPECTED VALUE ANALYSIS . . . . . . . . . . . . . . . . 280
x
APPENDICES
A POWER CURVE COEFFICIENTS FROM MEAN DAMAGE FACTOR VS. STRUCTURAL SCORE REGRESSIONS 254
B HALF-STEP FRAGILITY CURVES. . . . . .. 261
C BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT FOR A DESIGN BASIS EARTHQUAKE . 274
D BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT BASED ON AN EXPECTED VALUE ANALYSIS . . . . . . . . . . . . . . .. 280
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE
I
II
III
IV
v
VI
LIST OF TABLES
Modified Mercalli Intensity Scale
ATC-13 Facility Classes and Numbers
ATC-13 Damage State Descriptions
ATC-13 Injury and Death Rates
ATC-21 Building Types
ATC-21 Performance Modifiers
VII ATC-21 Seismicities and EarthquaKe
VIII
IX
X
Intensities ....... .
ATC-21 Basic Structural Hazard Scores
Ranges of Estimated Number of People
used in ATC-21 and PSHS-93
Post-Benchmark Years used in PSHS-93
XI PGA on Rock and Recurrence Intervals for
PAGE
15
17
21
23
27
28
31
32
39
40
Portland (Geomatrix Consultants 1995) . 43
XII Liquefaction Susceptibility from
GMS-79 Plate 1 . . . . . . .
XIII Lateral Spread Displacement from
GMS-79 Plate 3
XIV Ground Motion Amplification from
GMS-79 Plate 2
48
50
52
TABLE
I
II
III
IV
V
VI
LIST OF TABLES
Modified Mercalli Intensity Scale
ATC-13 Facility Classes and Numbers
ATC-13 Damage State Descriptions
ATC-13 Injury and Death Rates
ATC-21 Building Types
ATC-21 Performance Modifiers
VII ATC-21 Seismicities and EarthquaKe
Intensities. . . . . . . .
VIII
IX
X
ATC-21 Basic Structural Hazard Scores
Ranges of Estimated Number of People
used in ATC-21 and PSHS-93
Post-Benchmark Years used in PSHS-93
XI PGA on Rock and Recurrence Intervals for
PAGE
15
17
21
23
27
28
31
32
39
40
Portland (Geomatrix Consultants 1995). 43
XII
XIII
Liquefaction Susceptibility from
GMS-79 Plate 1 . . . . . . .
Lateral Spread Displacement from
GMS-79 Plate 3 . . . . . . .
XIV Ground Motion Amplification from
GMS-79 Plate 2 • t • , •••• , •••
48
50
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XV
XVI
XVII
XVIII
XIX
XX
XXI
Relative Dynamic Slope Instability from
GMS-79 Plate 3
FEMA-156 Building Group Mean Cost
FEMA-156 Area Adjustment Factors .
FEMA-156 Seismicity/Performance Objective
Adjustment Factor . . . . . . . .
FEMA-156 Location Adjustment Factor
FEMA-156 Time Adjustment Factor
Damage Function for Tilt Up Buildings
from ATC-13 Appendix G ..... .
XXII Weighted Statistics of the Damage Factor
for Tilt-up Buildings from
XXIII
XXIV
ATC-13, Appendix G
Damage Factor Modification Constants for
Non-California Buildings from
ATC-21-1 Table B3
ATC-21 Seismicity by NEHRP Map Area
XXV The Standard Normal Variate vs.
Structural Score
XXVI Structural Score vs. Mean Damage Factor for
xii
53
70
71
71
72
73
86
93
96
97
102
Tilt-up Buildings (PC1) at PGA : 0. 22 . 104
XXVII
XXVIII
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA: .22 with MC: 1. 35 ..... .
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA: .05 with MC: 1. 35 ..... ,
113
113
XV
XVI
XVII
XVIII
XIX
XX
XXI
Relative Dynamic Slope Instability from
GMS-79 Plate 3
FEMA-156 Building Group Mean Cost
FEMA-156 Area Adjustment Factors.
FEMA-156 Seismicity/Performance Objective
Adjustment Factor . . . . . . . .
FEMA-156 Location Adjustment Factor
FEMA-156 Time Adjustment Factor
Damage Function for Tilt Up Buildings
from ATC-13 Appendix G ..... .
XXII Weighted Statistics of the Damage Factor
for Tilt-up Buildings from
XXIII
XXIV
ATC-13, Appendix G
Damage Factor Modification Constants for
Non-California Buildings from
ATC-21-1 Table B3
ATC-21 Seismicity by NEHRP Map Area
XXV The Standard Normal Variate vs.
Structural Score
XXVI Structural Score vs. Mean Damage Factor for
xii
53
70
71
71
72
73
86
93
96
97
102
Tilt-up Buildings (PC1) at PGA : 0.22. 104
XXVII E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA : .22 with MC : 1.35 ..... . 113
XXVIII E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA : .05 with Me = 1.35 ..... . 113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XXIX
XXX
XXXI
XXXII
XXXIII
XXXIV
XXXV
XXXVI
XXXVII
XXXVIII
XXXIX
XL
XLI
XLII
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA = . 10 with MC = 1. 35
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA = .47 with MC = 1. 35
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA = 1.02 with MC = 1. 35
PGA vs. Mean Damage Factor for Tilt-up
at E(PMF) : 0 with MC = 1. 35
ATC-21 Soil Profile and Ground
Amplification .
Amplification Factor (AF)
Liquefaction Factor
Slope Factor
Unit Replacement Values (URV)
Fraction Dead vs. Central Damage Factor
from ATC-13 Table 9. 3
Expected Retrofit Effectiveness (ERE)
for W, S3 Buildings
Expected Retrofit Effectiveness (ERE)
for C1, S1, S2, S4 Buildings
Expected Retrofit Effectiveness (ERE)
for C3/S5, C2, PC2 Buildings
Expected Retrofit Effectiveness (ERE)
for RM Buildings
xiii
114
114
115
115
133
134
136
137
144
145
150
150
151
151
XXIX
XXX
XXXI
XXXII
XXXIII
XXXIV
XXXV
XXXVI
XXXVII
XXXVIII
XXXIX
XL
XLI
XLII
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA = .10 with MC = 1.35
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA = .47 with MC = 1.35
E(PMF) vs. Mean Damage Factor for Tilt-up
at PGA = 1.02 with MC = 1.35
PGA vs. Mean Damage Factor for Tilt-up
at E(PMF) = 0 with MC = 1.35
ATC-21 Soil Profile and Ground
Amplification .
Amplification Factor (AF)
Liquefaction Factor
Slope Factor
Unit Replacement Values (URV)
Fraction Dead vs. Central Damage Factor
from ATC-13 Table 9.3
Expected Retrofit Effectiveness (ERE)
for W. S3 Buildings .
Expected Retrofit Effectiveness (ERE)
for C1. S1. S2. S4 Buildings
Expected Retrofit Effectiveness (ERE)
for C3/S5. C2. PC2 Buildings
Expected Retrofit Effectiveness (ERE)
for RM Buildings . .
xiii
114
114
115
115
133
134
136
137
144
145
150
150
151
151
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XLIII
XLIV
Expected Retrofit Effectiveness (ERE)
for URM, PC1 Buildings
Total Unit Retrofit Costs for
Commercial Buildings, $/SF
XLV Total Unit Retrofit Costs for
Institutional Buildings, $/SF
XLVI Total Unit Retrofit Costs for
XLVII
XLVIII
XLIX
Industrial Buildings, $/SF
Humber of Buildings, Area, and Average
Score by Building Type
Humber of Buildings by Range of
Structural Score (S)
Humber of Buildings by Range of Soil
Modification Factor (SMF)
L Lowpoint and Midpoint of Range of
LI
LII
LIII
LIV
Humber of People and Humber o£
Buildings by Occupancy Type
Summary o£ Losses £or the Portland Quad
at PGA = .08 (100 Year Return)
Summary o£ Losses £or the Portland Quad
at PGA = .20 (500 Year Return)
Summary of Losses £or the Portland Quad
at PGA = .27 (1,000 Year Return)
Summary of Losses £or the Portland Quad
at PGA = . 39 (2, 500 Year Return)
xiv
152
162
162
163
190
191
191
192
194
195
196
197
XLIII
XLIV
Expected Retrofit Effectiveness (ERE)
for URM, PCl Buildings
Total Unit Retrofit Costs for
Commercial Buildings, $/SF
XLV Total Unit Retrofit Costs for
Institutional Buildings, $/SF
XLVI Total Unit Retrofit Costs for
XLVII
XLVIII
XLIX
Industrial Buildings, $/SF
Humber of Buildings, Area, and Average
Score by Building Type
Humber of Buildings by Range of
Structural Score (S)
Humber of Buildings by Range of Soil
Modification Factor (SMF) .
L Lowpoint and Midpoint of Range of
LI
LII
LIII
LIV
Humber of People and Humber of
Buildings by Occupancy Type
Summary of Losses for the Portland Quad
at PGA : .08 (100 Year Return)
Summary of Losses for the Portland Quad
at PGA : .20 (500 Year Return)
Summary of Losses for the Portland Quad
at PGA : .27 (1,000 Year Return)
Summary of Losses for the Portland Quad
at PGA : .39 (2,500 Year Return)
xiv
152
162
162
163
190
191
191
192
194
195
196
197
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LV Loss of Life by Range of
Structural Score
LVI Loss of Life by Range of Soil
Modification Factor (SMF)
LVII
LVIII
Loss of Life by Lowpoint and Midpoint
of Humber of People Range,
by Occupancy Type .
Humber of Buildings in 2 Damage States
by Building Type
LIX Damage State Comparison to Kobe City
1995 Earthquake .
LX Building Damage and Loss of Life
XV
200
201
202
203
204
Attributable to Performance Modifiers 205
LXI Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
LXII
LXIII
LXIV
at PGA = .08 (100 Year Return)
Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
at PGA = . 20 (500 Year Return)
Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
at PGA = . 27 (1,000 Year Return)
Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
at PGA = . 39 (2, 500 Year Return)
208
209
210
211
LV Loss of Life by Range of
Structural Score
LVI Loss of Life by Range of Soil
Modification Factor (SMF)
LVII
LVIII
Loss of Life by Lowpoint and Midpoint
of Humber of People Range,
by Occupancy Type .
Humber of Buildings in 2 Damage States
by Building Type . .
LIX Damage State Comparison to Kobe City
1995 Earthquake. .
LX Building Damage and Loss of Life
xv
200
201
202
203
204
Attributable to Performance Modifiers 205
LXI Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
LXII
LXIII
LXIV
at PGA : .08 (100 Year Return)
Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
at PGA : . 20 (500 Year Return)
Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
at PGA : .27 (1,000 Year Return)
Summary of Losses for the Portland Quad
if All Bldgs are Retrofit
at PGA : . 39 (2, 500 Year Return)
208
209
210
211
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LXV
LXVI
LXVII
LXVIII
LXIX
LXX
LXXI
LXXII
LXXIII
LXXIV
LXXV
Effect of Retrofitting All Buildings .
System Parameters of Optimal Retrofit
Investment for a Scenario Earthquake
xvi
213
Event of PGA = . 20 . . . . 217
Comparison Between Complete Retrofit
and Optimal Retrofit ..... .
Summary of Building Types and Occupancy
Groups Included in the Optimal
Retrofit Program
Percentage of Building Types Included
in Optimal Retrofit Program ...
Percentage of Occupancy Type Groups
Included in Optimal Retrofit Program
Sensitivity Analysis Results .....
Annual Expected Value of Full Retrofit
System Parameters of Optimal Retrofit
Investment Based on an
Expected Value Analysis
Summary of Building Types and Occupancy
Groups Included in the Optimal Retrofit
Program Based on an Expected
Value Analysis
Percentage of Building Types Included in
Optimal Retrofit Program Based on
an Expected Value Analysis
220
221
222
223
225
229
234
234
235
LXV
LXVI
LXVII
LXVIII
LXIX
LXX
LXXI
LXXII
LXXIII
LXXIV
LXXV
Effect of Retrofitting All Buildings.
System Parameters of Optimal Retrofit
Investment for a Scenario Earthquake
xvi
213
Event of PGA = . 20 . . . . 217
Comparison Between Complete Retrofit
and Optimal Retrofit ..... .
Summary of Building Types and Occupancy
Groups Included in the Optimal
Retrofit Program
Percentage of Building Types Included
in Optimal Retrofit Program. . .
Percentage of Occupancy Type Groups
Included in Optimal Retrofit Program
Sensitivity Analysis Results .....
Annual Expected Value of Full Retrofit
System Parameters of Optimal Retrofit
Investment Based on an
Expected Value Analysis
Summary of Building Types and Occupancy
Groups Included in the Optimal Retrofit
Program Based on an Expected
Value Analysis
Percentage of Building Types Included in
Optimal Retrofit Program Based on
an Expected Value Analysis
220
221
222
223
225
229
234
234
235
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LXXVI
LXXVII
Percentage of Occupancy Type Groups
Included in Optimal Retrofit Program
xvii
Based on an Expected Value Analysis 236
Comparison of System Parameters of
Expected Value vs. Design Basis
Optimal Retrofit Programs 238
LXXVI
LXXVII
Percentage of Occupancy Type Groups
Included in Optimal Retrofit Program
xvii
Based on an Expected Value Analysis 236
Comparison of System Parameters of
Expected Value vs. Design Basis
Optimal Retrofit Programs 238
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
PAGE
FIGURE
1. NEHRP Map Areas of the United States
(ATC-21 1988) . 30
2. A Sample Page from the 1993 Portland
Seismic Hazards Survey Database 38
3. Liquefaction Susceptibility Map
(GMS-79, Plate 1) . . . . . . 45
4. Ground Motion Amplification Map
(GMS-79, Plate 2) . . . . . . 46
5. Lateral Spread Displacement and Dynamic
Slope Instability Map (GMS-79 Plate 3). 47
6.
7.
Earthquake Loss Modeling Flowchart ..
Fragility Curve for TILT UP Buildings
Derived from Table XXI
8. Lognormal Distribution of Estimated DF
at PGA = . 22 for Average California
TILT UP bldgs . . .
9. Normal Distribution of ln(DF) at
PGA = . 22 for Average Claifornia
TILT-UP buildings . . . . . . . .
82
87
94
95
FIGURE
1.
LIST OF FIGURES
NEHRP Map Areas of the United States
(ATC-21 1988) . . ..... .
2. A Sample Page from the 1993 Portland
3.
Seismic Hazards Survey Database
Liquefaction Susceptibility Map
(GMS-79, Plate 1) . . . . . .
4. Ground Motion Amplification Map
(GMS-79, Plate 2) . . . . . .
5. Lateral Spread Displacement and Dynamic
PAGE
30
38
45
46
Slope Instability Map (GMS-79 Plate 3). 47
6.
7.
Earthquake Loss Modeling Flowchart . .
Fragility Curve for TILT UP Buildings
Derived from Table XXI
B. Lognormal Distribution of Estimated DF
at PGA = • 22 for Average California
TILT UP bldgs . . .
9. Normal Distribution of In(DF) at
PGA = .22 for Average Claifornia
TILT-UP buildings . . . . . . . .
82
B7
94
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10. Determining the Mean Damage Factor (HDF)
from the Structural Score (S) and the
Normally Distributed ln(DF)
11. Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
for PGA = .05
12. Mean Damage Factor vs. Structural Score
13.
for TILT UP (PC1) buildings,
for PGA : . 10
Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
for PGA = . 22
14. Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
for PGA = .47
15. Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
xix
103
105
106
107
108
for PGA: 1.02 109
16. Mean Damage Factor vs. Structural Score for
TILT-UP at PGA = . 22, Showing Coordinates
for E(PMF) = 0 and E(PMF) = -1 112
17. PC1 - TILT UP fragility curve for
E ( PMF) : 0 11 7
10. Determining the Mean Damage Factor (HDF)
from the Structural Score (S) and the
Normally Distributed In(DF)
11. Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
for PGA = .05
12. Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
for PGA = . 10
13. Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
for PGA = . 22 . .. .
14. Mean Damage Factor vs. Structural Score
15.
for TILT UP (PC1) buildings,
for PGA = .47
Mean Damage Factor vs. Structural Score
for TILT UP (PC1) buildings,
for PGA = 1.02 ...
16. Mean Damage Factor vs. structural Score for
TILT-UP at PGA = • 22, Showing Coordinates
xix
103
105
106
107
108
109
for E(PMF) = 0 and E(PMF) = -1 . 112
17. PC1 - TILT UP fragility curve for
E (PHF) = 0 ••• 117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18. PC1 - TILT UP fragi 1 i ty curves for
integer E(PMF) steps
19. W - WOOD fragility curves for
integer E(PMF} steps
20. S1 - STEEL MRF fragility curves for
integer E(PMF) steps
21. S2 - BR STL FRAME fragility curves for
integer E(PMF) steps
22. S3 - LIGHT MEOTAL fragility curves for
integer E(PMF) steps
23. S4 - STL FRAME W/ COHC SHEARWALLS fragility
curves for integer E(PMF) steps .
24. C1 - REIHF COHC MRF fragility curves for
integer E(PMF) steps
25. C2 - REIHF COHC SHEARWALL fragility curves
26.
£or integer E(PMF) steps
• C3/S5 - URM IHFILL fragility curves £or
integer E(PMF) steps
27. PC2 - PRECAST COHC FRAME fragility curves
for integer E(PMF) steps
28. RM - REIHF MASOHRY fragility curves £or
integer E(PMF) steps
29. URM - UHREIHF MASOHRY fragility curves for
integer E(PMF) steps
XX
119
120
121
122
123
124
125
126
127
128
129
130
18. PCl - TILT UP fragility curves for
integer E(PMF) steps
19. W - WOOD fragility curves for
integer E(PHF) steps
20. S1 - STEEL MRF fragility curves for
integer E(PMF) steps
21. S2 - BR STL FRAME fragility curves for
integer E(PMF) steps
22. S3 - LIGHT MEOTAL fragility curves for
integer E(PMF) steps
23. S4 - STL FRAME wi CONC SHEARWALLS fragility
curves for integer E(PHF) steps.
24. C1 - REINF CONC HRF fragility curves for
integer E(PMF) steps
25. C2 - REINF CONC SHEARWALL fragility curves
26.
for integer E(PHF) steps
• C3/S5 - URH INFILL fragility curves for
integer E(PHF) steps
27. PC2 - PRECAST CONC FRAME fragility curves
for integer E(PHF) steps
28. RH - REINF MASONRY fragility curves for
integer E(PHF) steps
29. URH - UHREINF MASONRY fragility curves for
integer E(PHF) steps
xx
119
120
121
122
123
124
125
126
127
128
129
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30. Comparison of URM Fragility Curve
31.
for "Average" Buildings with
Goettel & Horner (1995)
Damage Factor Estimates
Comparison of C3/S5 Fragility Curve
for "Average" Buildings with
Goettel & Horner (1995)
Damage Factor Estimates
32. Fraction Dead vs. Damage Factor
Relationship Derived from
ATC-13 (1985) Table 9. 3
33. Retrofit Effectiveness for
W and S3 Buildings
34. Retrofit Effectiveness for
C1, S1, S2, and S4 Buildings
35. Retrofit Effectiveness £or
C3/S5, C2, and PC2 Buildings
36. Retrofit Effectiveness for
RM Buildings
37. Retrofit Effectiveness £or
38.
39.
40.
41.
URM and PC1 Buildings
Example of DAM-FACT Module output file
Example of DAMAGE-$ Module output file
Example of LIFELOSS Module output file
Example of RETRO-DF Module output file
xxi
140
141
146
153
154-
155
156
157
176
177
178
182
30. Comparison of URM Fragility Curve
31.
for "Average" Buildings with
Goettel & Horner (1995)
Damage Factor Estimates
Comparison of C3/S5 Fragility Curve
for "Average" Buildings with
Goettel & Horner (1995)
Damage Factor Estimates
32. Fraction Dead vs. Damage Factor
Relationship Derived from
ATC-13 (1985) Table 9.3
33. Retrofit Effectiveness for
Wand S3 Buildings
34. Retrofit Effectiveness for
C1, S1, S2, and S4 Buildings
35. Retrofit Effectiveness for
C3/S5, C2, and PC2 Buildings
36. Retrofit Effectiveness for
RM Buildings
37. Retrofit Effectiveness for
38.
39.
40.
41.
URM and PC1 Buildings
Example of DAM-FACT Module output file
Example of DAMAGE-$ Module output file .
Example of LIFELOSS Module output file .
Example of RETRO-DF Module output file
xxi
140
141
146
153
154
155
156
157
176
177
178
182
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42.
43.
44.
Example of RETRO-$D Module output file .
Example of RETRO-LL Module output file
System Efficiency Curves for Retrofit for
DBE (PGA = .20) for the Portland Quad
45. Lives Saved vs. Retrofit Investment for
Retrofit for DBE (PGA = .20)
46. Building Damage Avoided vs. Retrofit
Investment for Retrofit for
DBE (PGA = .20) ....
47. System Efficiency Curves for Retrofit for
DBE (PGA = .20) for the Portland Quad,
Using the Lowpoint of HP Range
48. System Efficiency Curves for Retrofit for
DBE (PGA = .20) for the Portland Quad,
Using VOL=$ 500,000
System Efficiency Curves for Retrofit for
Expected Value Analysis for
the Portland Quad . . .
xxii
183
184
216
218
219
226
227
233
42.
43.
44.
Example of RETRO-$D Module output file .
Example of RETRO-LL Module output file
System Efficiency Curves for Retrofit for
DBE (PGA = .20) for the Portland Quad
45. Lives Saved vs. Retrofit Investment for
Retrofit for DBE (PGA = .20)
46. Building Damage Avoided vs. Retrofit
Investment for Retrofit for
DBE (PGA = .20) ....
47. System Efficiency Curves for Retrofit for
DBE (PGA = .20) for the Portland Quad,
Using the Lowpoint of HP Range
48. System Efficiency Curves for Retrofit for
DBE (PGA = .20) for the Portland Quad,
Using VOL = $ 500,000
System Efficiency Curves for Retrofit for
Expected Value Analysis for
the Portland Quad. . .
xxii
183
184
216
218
219
226
227
233
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER I
INTRODUCTION
PROBLEM DEFINITION
Newly discovered geologic evidence shows that the
entire Pacific Northwest, including Portland, is subject
to periodic large earthquakes (Geomatrix Consultants
1995). Unlike the plate tectonic mechanisms in California,
which generate large earthquakes every decade or so, the
geomechanics of the Pacific Northwest produce large events
on the order of hundreds of years apart. Very recent
studies have shown that the average interval between
occurrances of the large events is 450 years, with a 90Y.
confidence interval of 200 years; the last event occurred
about 300 years ago (Geomatrix Consultants 1995).
Since our historical record is less than two hundred
years old, comparison to the active faults in California
had previously led the scientific and engineering
community to assume Portland is in a region of low to
moderate seismicity. Prior to the 1994 edition of the
Uniform Building Code (International Conference of
Building Officials 1994), buildings in Portland were only
required to resist lateral forces equal to two-thirds of
that now required for design. Further, prior to the early
CHAPTER I
INTRODUCTION
PROBLEM DEFINITION
Newly discovered geologic evidence shows that the
entire Pacific Northwest, including Portland, is subject
to periodic large earthquakes (Geomatrix Consultants
1995). Unli~e the plate tectonic mechanisms in California,
which generate large earthquakes every decade or so, the
geomechanics of the Pacific Northwest produce large events
on the order of hundreds of years apart. Very recent
studies have shown that the average interval between
occurrances of the large events is 450 years, with a 901-
confidence interval of 200 years; the last event occurred
about 300 years ago (Geomatrix Consultants 1995).
Since our historical record is less than two hundred
years old, comparison to the active faults in California
had previously led the scientific and engineering
community to assume Portland is in a region of low to
moderate seismicity. Prior to the 1994 edition of the
Uniform Building Code (International Conference of
Building Officials 1994), buildings in Portland were only
required to resist lateral forces equal to two-thirds of
that now required for design. Further, prior to the early
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1970's, much less was Known about seismic design, so that
all construction in the first two-thirds of this century
is non-seismic resistant construction, i.e., without
special attention given to seismic resistance.
2
Portland is home to thousands of existing buildings
which have not been designed to resist the larger
earthquakes which are now expected. Since we have rarely
had earthquakes producing even slight damage, the problem
is made worse in that grossly deficient old buildings have
not been automatically "culled out" (as they are in
California) by moderate quakes, and in addition public
awareness of the risk potential is lacking. Many
structural engineers now feel that we are sitting on a
powder keg; sometime in the future, Portland will
experience a large earthquake, and the damage and loss of
life will be much greater than seen during a similar event
in California, because of our deficient building stock.
A responsible public safety policy, in view of
current risk awareness, should require many buildings in
Portland to be strengthened in an effort to reduce damage
and casualties. However, retroactively requiring
improvements in a community's existing building stock is
among the most conflictual and difficult types of public
policy decisions, requiring careful engineering and
economic analysis and consideration of societal
1970's, much less was known about seismic design, so that
all construction in the first two-thirds of this century
is non-seismic resistant construction, i.e., without
special attention given to seismic resistance.
2
Portland is home to thousands of existing buildings
which have not been designed to resist the larger
earthquakes which are now expected. Since we have rarely
had earthquakes producing even slight damage, the problem
is made worse in that grossly deficient old buildings have
not been automatically "culled out" (as they are in
California) by moderate quakes, and in addition public
awareness of the risk potential is lacking. Many
structural engineers now feel that we are sitting on a
powder keg; sometime in the future, Portland will
experience a large earthquake, and the damage and loss of
life will be much greater than seen during a similar event
in California, because of our deficient building stock.
A responsible public safety policy, in view of
current risk awareness, should require many buildings in
Portland to be strengthened in an effort to reduce damage
and casualties. However, retroactively requiring
improvements in a community's existing building stock is
among the most conflictual and difficult types of public
policy decisions, requiring careful engineering and
economic analysis and consideration of societal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
priorities. In most cases, the value of life is the
principal motivation for implementing seismic retrofit
programs, while in some instances property protection or
continued function may be the driving force.
The basic problem, from a city manager or
decision-maker viewpoint, is to decide which buildings
"should" be retrofit, which should not be retrofit,
how much a retrofit program will cost, and how should it
be financed.
"Should be retrofit" refers to sufficiently high
benefits from retrofit --- savings in life and property
--- to warrant the expense. Some buildings should not be
retrofit, either because the ris~ is low, or because
3
the ris~ is so high that retrofit will not be effective in
making them safe, or because the cost is too high to
justify the expected benefits. Clearly, some form of
bene£it-cost analysis is needed to make rational choices.
Sufficiently accurate modeling of earthquake losses,
avoided losses (benefits) resulting £rom retrofitting, and
the cost of retrofit must precede and provide the input to
the benefit-cost analysis. The required modeling is
probabilistic, complex, and multi-disciplinary.
It will be desirable to optimize the sequence of
buildings to be strengthened in a retrofit program. The
first dollars spent should go to retrofit those buildings
priorities. In most cases, the value of life is the
principal motivation for implementing seismic retrofit
programs, while in some instances property protection or
continued function may be the driving force.
The basic problem, from a city manager or
decision-maker viewpoint, is to decide which buildings
"should" be retrofit, which should not be retrofit,
how much a retrofit program will cost, and how should it
be financed.
"Should be retrofit" refers to sufficiently high
benefits from retrofit --- savings in life and property
--- to warrant the expense. Some buildings should not be
retrofit, either because the risK is low, or because
3
the risK is so high that retrofit will not be effective in
maRing them safe, or because the cost is too high to
justify the expected benefits. Clearly, some form of
benefit-cost analysis is needed to make rational choices.
Sufficiently accurate modeling of earthquake losses,
avoided losses (benefits) resulting from retrofitting, and
the cost of retrofit must precede and provide the input to
the benefit-cost analysis. The required modeling is
probabilistic, complex, and multi-disciplinary.
It will be desirable to optimize the sequence of
buildings to be strengthened in a retrofit program. The
first dollars spent should go to retrofit those buildings
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
where savings of life will be greatest. An efficient
prioritization of buildings earmarked for retrofit, based
on benefit-cost analysis, should be established. If
constraints in available resources --- time and money
dictate a staged retrofit program over time, the
prioritization becomes very meaningful indeed.
GENERAL NATURE OF THE PROBLEM
This research focuses on assessing and avoiding
potential seismic damage and casualties (losses) to
non-residential buildings in a region or metropolitan
area. Methodologies for creating a detailed inventory of
data on the buildings, accurately modeling losses,
modeling reduced losses anticpated from retrofitting,
establishing costs, prioritizing buildings for retrofit,
and planning resource allocation are multi-disciplinary,
complex, and uncertain in nature. High-quality decision
making will require a decision maker to "see through" the
complexities of the problem. These characteristics of the
nature of the problem point toward utilizing a systems
problem solving approach.
Systems Approach
A "system" is defined as a time-varying
configuration of people, hardware, and procedures
organized for the purpose of accomplishing certain
where savings of life will be greatest. An efficient
prioritization of buildings earmarked for retrofit, based
on benefit-cost analysis, should be established. If
constraints in available resources --- time and money
dictate a staged retrofit program over time, the
prioritization becomes very meaningful indeed.
GENERAL NATURE OF THE PROBLEM
This research focuses on assessing and avoiding
potential seismic damage and casualties (losses) to
non-residential buildings in a region or metropolitan
area. Methodologies for creating a detailed inventory of
data on the buildings, accurately modeling losses,
modeling reduced losses anticpated from retrofitting,
establishing costs, prioritizing buildings for retrofit,
and planning resource allocation are multi-disciplinary,
complex, and uncertain in nature. High-quality decision
making will require a decision maker to "see through" the
complexities of the problem. These characteristics of the
nature of the problem point toward utilizing a systems
problem solving approach.
Systems Approach
A "system" is defined as a time-varying
configuration of people, hardware, and procedures
organized for the purpose of accomplishing certain
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
functions (Robertshaw, Mecca and Rerick 1978).
Inherent in this functional definition of system is
the idea that there are many alternative system
descriptions of the same "reality," and that the system
description selected is a representation of reality, not
reality itself. The choice of system description is made
by the analyst, based on the problem to be solved.
In assessing regional earthquake damage, it has
become customary to classify losses as belonging to one of
three general categories (National Institute of Building
Sciences 1995), as follows:
Buildings: residential, non-residential.
Transportation Systems: highways, railways, light rail, bus, ports and harbors, ferry, and airports.
Lifeline Utility Systems: potable water supply, waste water, oil, natural gas, electric power, communications.
Decisions regarding loss mitigation are likely to be
made by individual government agencies responsible for
each category: city/county building departments, state
transportation departments, and individual utilities.
Clearly, the above categories are system descriptions
designed to facilitate earthquake hazard mitigation
decision-making. The present research follows this
current practice by considering one category of loss:
non-residential buildings in a region.
5
functions (Robertshaw, Mecca and Rerick 1978).
Inherent in this functional definition of system is
the idea that there are many alternative system
descriptions of the same "reality, II and that the system
description selected is a representation of reality, not
reality itself. The choice of system description is made
by the analyst, based on the problem to be solved.
In assessing regional earthquake damage, it has
become customary to classify losses as belonging to one of
three general categories (National Institute of Building
SCiences 1995), as follows:
Buildings: residential, non-residential.
Transportation Systems: highways, railways, light rail, bus, ports and harbors, ferry, and airports.
Lifeline Utility Systems: potable water supply, waste water, 011, natural gas, electriC power, communications.
Decisions regardlng loss mitigation are likely to be
made by individual government agencies responsible for
each category: city/county building departments, state
transportation departments, and individual utilities.
Clearly, the above categories are system descriptions
designed to facilitate earthquake hazard mitigation
decision-making. The present research follows this
current practice by considering one category of loss:
non-residential buildings in a region.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
Inter-dependence between the NIBS loss categories is
a potential topic for further research, to explore
"cross-category" loss synergisms and consequent potential
changes in retrofit priorities.
System Defined
The system description selected for use in the
present research is defined as follows:
SYSTEM: Non-residential buildings of a region.
SUB-SYSTEMS: Building owners. Building occupants.
ENVIRONMENT: Earthquake magnitudes and probabilities. Soil conditions at building sites.
DECISION-MAKER: Municipal Managers; State Legislature.
Multi-Disiplinary Aspects
Although earthquake rehabilitation seems at first
glance to be essentially a structural engineering problem,
it is actually a tasK that crosses a number of traditional
technical discipline boundaries, including the following:
Geology Geotechnical Engineering Structural Engineering Architecture Construction Management Economics Urban planning Systems Science.
Geologists are depended upon to provide estimates of
magnitude and expected frequency of occurrance for seismic
6
Inter-dependence between the NIBS loss categories is
a potential topic for further research, to explore
"cross-category" loss synergisms and consequent potential
changes in retrofit priorities.
System Defined
The system description selected for use in the
present research is defined as follows:
SYSTEM: Non-residential buildings of a region.
SUB-SYSTEMS: Building owners. Building occupants.
ENVIRONMENT: Earthquake magnitudes and probabilities. Soil conditions at building sites.
DECISION-MAKER: Municipal Managers; State Legislature.
Multi-Disiplinary Aspects
Although earthquake rehabilitation seems at first
glance to be essentially a structural engineering problem,
it is actually a tasK that crosses a number of traditional
technical discipline boundaries, including the following:
Geology Geotechnical Engineering Structural Engineering Architecture Construction Management Economics Urban planning Systems Science.
Geologists are depended upon to provide estimates of
magnitude and expected frequency of occurrance for seismic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
events; indeed, as previously discussed, new geologic
information about the Pacific Northwest is the genesis of
the increased local interest in this subject. Geotechnical
engineering addresses the response of subsurface soils,
which often significantly increases the damaging effect of
the ground shaKing. Structural engineering is of course
the "center-stage" discipline: structural engineers are
the group providing estimates of damage and techniques for
rehabilitation of existing buildings. Architects address
issues of building occupancies and building finishes which
must be removed and/or restored as part of the retrofit
process. Construction managers provide input on the cost
required for seismic rehabilitation, and carry out the
work of retrofitting the buildings. Economic theories of
benefit-cost analysis are used to determine where
retrofits are and are not justified, and provide tools to
analyze the efficiency of a retrofit program. Urban
planners and municipal authorities are the professionals
who must maKe recommendations and implement a retrofit
program.
The multi-disciplinary nature of a seismic
rehabilitation program may impose difficulties in current
practice. The professionals in the various technical
disciplines described are problem solving specialists,
adept in their own disipline but not necessarily adept at
7
eventsj indeed, as previously discussed, new geologic
information about the Pacific Northwest is the genesis of
the increased local interest in this subject. Geotechnical
engineering addresses the response of subsurface soils,
which often significantly increases the damaging effect of
the ground shaKing. Structural engineering is of course
the "center-stage" discipline: structural engineers are
the group providing estimates of damage and techniques for
rehabilitation of existing buildings. Architects address
issues of building occupancies and building finishes which
must be removed and/or restored as part of the retrofit
process. Construction managers provide input on the cost
required for seismic rehabilitation, and carry out the
work of retrofitting the buildings. Economic theories of
benefit-cost analysis are used to determine where
retrofits are and are not justified, and provide tools to
analyze the efficiency of a retrofit program. Urban
planners and municipal authorities are the professionals
who must maKe recommendations and implement a retrofit
program.
The multi-disciplinary nature of a seismic
rehabilitation program may impose difficulties in current
practice. The professionals in the various technical
disciplines described are problem solving speCialists,
adept in their own disipline but not necessarily adept at
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
cross-disciplinary problem solving. A general problem
solving approach is needed to put the whole package
together. Systems science is that general problem solving
approach, and can form the backbone to synthesize the
various disciplines. A systems approach would provide
tools to safeguard against leaving out important elements
of the problem, plus tools to bring the diverse elements
to a solution.
Complexity
Complexity can be defined as a state that involves
uncertainty, indescribability, intangibles, and many
interactions (Robertshaw, Mecca and Rerick, 1978). The
losses sustained by a building in an earthquake depend on
the combined effect of the interactions between many
parameters of the SYSTEM and of the EHVIROHMEHT.
The SYSTEM parameters (attributes belonging to the
non-residential buildings of a region) used in the present
research are:
12 Building Structure Types 10 Characteristics of Design & Construction
8 Occupancy Types 7 Occupancy Levels 4 Building Size Groups
The number of possible subsets of building
performance categories created by combinations of these
attributes are:
12 x 210 x 8 x 7 x 4 = 2,752,512 combinations.
8
cross-disciplinary problem solving. A general problem
solving approach is needed to put the whole package
together. Systems science is that general problem solving
approach, and can form the backbone to synthesize the
various disciplines. A systems approach would provide
tools to safeguard against leaving out important elements
of the problem, plus tools to bring the diverse elements
to a solution.
Complexity
Complexity can be defined as a state that involves
uncertainty, indescribability, intangibles, and many
interactions (Robertshaw, Mecca and Rerick, 1978). The
losses sustained by a building in an earthquake depend on
the combined effect of the interactions between many
parameters of the SYSTEM and of the EHVI ROHHEHT.
The SYSTEM parameters (attributes belonging to the
non-residential buildings of a region) used in the present
research are:
12 Building Structure Types 10 Characteristics of Design & Construction
8 Occupancy Types 7 Occupancy Levels 4 Building Size Groups
The number of possible subsets of building
performance categories created by combinations of these
attributes are:
12 x 210 x 8 x 7 x 4 : 2,752,512 combinations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
The proposed ENVIRONMENTAL parameters are:
3 Soil Profile Characteristics
6 Ground Motion/Probability values.
The environmental parameters generates the following
number of scenarios for consideration.
6 X 23 : 48
Finally, the total number of possible combinations
of system and environment parameters is a staggering:
2,752,512 X 48: 132,120,576.
The number of combinations of attributes creates
difficulties at two levels in optimizing retrofit
priorities. First, it makes the analysis of l9sses and
retrofit effectiveness more difficult. Second, it makes a
practical scheme to classify the results without losing
the detail of the analysis more difficult as well.
A major intangible in retrofit benefit-cost analysis
is the value of loss of life. It becomes necessary to
compare the value of human life (intangible) to the value
of building damages and retrofit costs (tangibles).
Additionally, cost data on the buildings and the
retrofits, and a method for rating retrofit effectiveness,
are required. These items contribute greatly to the
complexity of establishing a retrofit program.
Uncertainty
As an engineering community, in the USA we are
9
The proposed ENVIRONMENTAL parameters are:
3 Soil Profile Characteristics
6 Ground Motion/Probability values.
The environmental parameters generates the following
number of scenarios for consideration.
6 x 23 = 48
Finally, the total number of possible combinations
of system and environment parameters is a staggering:
2,752,512 x 48 = 132,120,576.
The number of combinations of attributes creates
difficulties at two levels in optimizing retrofit
priorities. First, it makes the analysis of 19sses and
retrofit effectiveness more difficult. Second, it makes a
practical scheme to classify the results without losing
the detail of the analysis more difficult as well.
A major intangible in retrofit benefit-cost analysis
is the value of loss of life. It becomes necessary to
compare the value of human life (intangible) to the value
of building damages and retrofit costs (tangibles).
Additionally, cost data on the buildings and the
retrofits, and a method for rating retrofit effectiveness,
are required. These items contribute greatly to the
complexity of establishing a retrofit program.
Uncertainty
As an engineering community, in the USA we are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
successful at designing new structures to withstand
earthquakes. For example, consider two earthquakes of
similar magnitude that tooK place in 1988 and 89: in the
Lorna Prieta (San Francisco) earthquake, approximately 60
people were Killed (BenusKa 1990), while in the Armenian
earthquake the dead numbered approximately 25,000 (Wyllie
and Filson 1989).
However, modeling earthquake loss estimates for a
large number of existing buildings is not the same thing
as designing new buildings to withstand earthquakes. The
loss modeling discipline is in its infancy. As
demonstrated above, loss modeling is complex, and
complexity leads to uncertainty.
At the present time, existing methodologies
generally address the probability of average damage for a
given group of similar structures, and do not apply to
individual, specific buildings. This failure to address
the complexity of the problem increases the uncertainty.
Additionally, the magnitude and return period of
earthquakes is a highly probabilistic science, with large
uncertainties. For example, the earthquakes now being
predicted for Portland and the Pacific Northwest have not
occurred in historic times, but are postulated based on
geologic evidence and probabilistic modeling.
The uncertainties are clearly compounded. For
10
successful at designing new structures to withstand
earthquakes. For example, consider two earthquakes of
similar magnitude that tooK place in 1988 and 89: in the
Loma Prieta (San Francisco) earthquake, approximately 60
people were Killed (BenusKa 1990), while in the Armenian
earthquake the dead numbered approximately 25,000 (Wyllie
and Filson 1989).
However, modeling earthquake loss estimates for a
large number of existing buildings is not the same thing
as designing new buildings to withstand earthquakes. The
loss modeling discipline is in its infancy. As
demonstrated above, loss modeling is complex, and
complexity leads to uncertainty.
At the present time, existing methodologies
generally address the probability of average damage for a
given group of similar structures, and do not apply to
individual, specific buildings. This failure to address
the complexity of the problem increases the uncertainty.
Additionally, the magnitude and return period of
earthquakes is a highly probabilistic SCience, with large
uncertainties. For example, the earthquakes now being
predicted for Portland and the Pacific Northwest have not
occurred in historiC times, but are postulated based on
geologic evidence and probabilistic modeling.
The uncertainties are clearly compounded. For
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
specified earthquake characteristics (magnitude, type, and
location), the response of a given building is uncertain.
However, the magnitude of the earthquake cannot be
specified without its own contribution of uncertainty.
Furthermore, there is the uncertainty of how closely the
entire regional system of real buildings fits the model of
damage for the "average" building.
Decision Theoretic Considerations
Establishing a retrofit program will involve
decision making under conditions of uncertainty. There
are two primary categories of "decision rule" which are
typically considered in making decisions under uncertainty
(Robertshaw, Mecca and Rerick 1978):
1. Maximum Expected Value, and
2. Maximin.
These two decision rules may produce opposite results,
which leads to more uncertainty about the decision.
Maximum Expected Value weights the expected utility
of outcomes according to the probability of occurance of a
particular state of nature. Maximin ignores
probabilities, and pessimistically chooses the alternative
which minimizes the maximum possible loss.
As a simplified example, suppose a hypothetical
building stands to cause its owner a loss of $10. M if a
"design basis" earthquake occurs (a design basis
11
specified earthquake characteristics (magnitude, type, and
location), the response of a given building is uncertain.
However, the magnitude of the earthquake cannot be
specified without its own contribution of uncertainty.
Furthermore, there is the uncertainty of how closely the
entire regional system of real buildings fits the model of
damage for the "average" building.
Decision Theoretic Considerations
Establishing a retrofit program will involve
decision making under conditions of uncertainty. There
are two primary categories of "decision rule" which are
typically considered in making deCisions under uncertainty
(Robertshaw, Mecca and Rerick 1978):
1. Maximum Expected Value, and
2. Maximin.
These two decision rules may produce opposite results,
which leads to more uncertainty about the decision.
Maximum Expected Value weights the expected utility
of outcomes according to the probability of occurance of a
particular state of nature. Maximin ignores
probabilities, and pessimistically chooses the alternative
which minimizes the maximum possible loss.
As a simplified example, suppose a hypothetical
building stands to cause its owner a loss of $10. M if a
"design basis" earthquake occurs (a design basis
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
earthquake has a 10-percent probability of occurrance in
the 50 year "life" of the building). Suppose also that if
the building is retrofit, at a cost of $1. M, the
earthquake losses are reduced from $10. M to $4. M.
If the owner decides to retrofit, there are two
possibilities for his losses, which are:
a) If the earthquake occurs (p = . 1):
reduced loss + retrofit cost = $ 4. M + 1. M
= $ 5.0 M
b) If the earthquake does not occur (p = .9):
retrofit cost only=$ 1.0 M
On the other hand, if he does not retrofit, his two
possible losses are:
c) If the earthquake occurs (p = . 1):
total earthquake loss = $ 10. M
d) If the earthquake does not occur (p = .9):
loss = 0.
If the maximum expected value decision rule is used,
the expected loss if the owner chooses to retrofit is
found by multiplying the losses from a) and b) by their
corresponding probabilities:
( . 1 IE $ 5. 0 M ) + ( • 9 IE $ 1. 0 M ) : $ 1. 4 M
The expected loss of choosing to not retrofit is found
similarly, from c) and d):
( . 1 IE $ 10 M ) + ( • 9 IE $ 0. ) : $ 1. 0 M
12
earthquake has a 10-percent probability of occurrance in
the 50 year "life" of the building). Suppose also that if
the building is retrofit, at a cost of $1. H, the
earthquake losses are reduced from $10. H to $4. H.
If the owner decides to retrofit, there are two
possibilities for his losses, which are:
a) If the earthquake occurs (p = .1):
reduced loss + retrofit cost = $ 4. H + 1. H
= $ 5.0 H
b) If the earthquake does not occur (p = .9):
retrofit cost only = $ 1.0 H
On the other hand, if he does not retrofit, his two
possible losses are:
c) If the earthquake occurs (p = .1):
total earthquaKe loss = $ 10. H
d) If the earthquake does not occur (p = .9):
loss = O.
If the maximum expected value decision rule is used,
the expected loss if the owner chooses to retrofit is
found by multiplying the losses from a) and b) by their
corresponding probabilities:
( . 1 * $ 5. 0 H ) + ( • 9 * $ 1. 0 H ) = $ 1. 4 H
The expected loss of choosing to not retrofit is found
similarly, from c) and d):
( . 1 * $ 10 H ) + ( .9 * $ O. ) = $ 1. 0 H
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Since the expected loss of retrofitting is 40-percent
greater than the expected loss of not retrofitting, the
maximum expected value decision rule indicates that the
building owner should not retrofit.
However, the maximin decision rule seeks to insure
avoidance of the worst outcome, pessimistically assuming
the worst state of nature (the earthquake occurs).
means choosing between a) and c) above; clearly, the
decision is to retrofit, to avoid the $ 10. M loss.
This
If maximum expected value and maximin produce
paradoxical results, as in this example, selecting the
correct decision rule will be a major point of debate in
retrofit decision making.
13
Since the expected loss of retrofitting is 40-percent
greater than the expected loss of not retrofitting, the
maximum expected value decision rule indicates that the
building owner should not retrofit.
However, the maximin decision rule seeks to insure
avoidance of the worst outcome, pessimistically assuming
the worst state of nature (the earthquake occurs).
means choosing between a) and c) above; clearly, the
decision is to retrofit, to avoid the $ 10. M loss.
This
If maximum expected value and maximin produce
paradoxical results, as in this example, selecting the
correct decision rule will be a major point of debate in
retrofit deCision making.
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER II
LITERATURE REVIEW
ATC-13: EARTHQUAKE DAMAGE EVALUATION DATA FOR CALIFORNIA
ATC-13 (1985) was developed by the Applied
Technology Council (ATC), funded by the Federal Emergency
Management Agency (FEHA), to estimate the economic impact
of a major California earthquaKe. ATC-13 presents
expert-opinion earthquaKe damage and loss estimates for
existing industrial, commercial, residential, utility and
transportation facilities in California.
ATC-13 is a crucial landmark study, establishing the
basic methodology of earthquaKe loss estimation, and
setting the stage for many research projects to follow. It
is still the most complete source of earthquaKe damage and
loss estimating information available for a wide variety of
structure types. Three tasks in the study of particular
importance to the present research are:
1. Identification of the earthquaKe shaKing
characterization most appropriate for damage and
loss estimation.
2. Development of facility classification schemes.
3. Development of earthquaKe damage and loss
CHAPTER II
LITERATURE REVIEW
ATC-13: EARTHQUAKE DAMAGE EVALUATION DATA FOR CALIFORNIA
ATC-13 (1985) was developed by the Applied
Technology Council (ATC) , funded by the Federal Emergency
Management Agency (FEHA) , to estimate the economic impact
of a major California earthquake. ATC-13 presents
expert-opinion earthquake damage and loss estimates for
existing industrial, commerCial, reSidential, utility and
transportation facilities in California.
ATC-13 is a crucial landmark study, establishing the
basic methodology of earthquake loss estimation, and
setting the stage for many research projects to follow. It
is still the most complete source of earthquake damage and
loss estimating information available for a wide variety of
structure types. Three tasKs in the study of particular
importance to the present research are:
1. Identification of the earthquake shaking
characterization most appropriate for damage and
loss estimation.
2. Development of facility classification schemes.
3. Development of earthquake damage and loss
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
estimates in terms of the earthquake shaking
characterization selected and facility classes
identified.
15
Earthquake Shaking Characterization
The Modified Mercalli Intensity (MMI) Scale (Wood
and Newmann 1931) was selected as the most appropriate
earthquake shaKing characterization, because most expert
knowledge and existing motion-damage data for earthquakes
in the USA exists in that form. The upper portion of the
MMI scale is summarized in Table I, following:
TABLE I
MODIFIED MERCALLI INTENSITY SCALE
MMI DESCRIPTION
VI. Felt by all, many frightened and run outdoors.
Some heavy furniture moved; a few instances of
fallen plaster or damaged chimneys; damage
slight.
VII. Everybody runs outdoors. Damage slight to
moderate in well-built ordinary structures; some
chimneys broken. Noticed by persons driving
cars.
VIII. Alarm approaches panic. Damage considerable in
ordinary substantial buildings with partial
estimates in terms of the earthquake shaking
characterization selected and facility classes
identified.
15
Earthquake Shaking Characterization
The Modified Mercalli Intensity (MMI) Scale (Wood
and Newmann 1931) was selected as the most appropriate
earthquake shaking characterization, because most expert
knowledge and existing motion-damage data for earthquakes
in the USA exists in that form. The upper portion of the
MMI scale is summarized in Table I, following:
TABLE I
MODIFIED MERCALLI INTENSITY SCALE
MMI DESCRIPTION
VI. Felt by all, many frightened and run outdoors.
Some heavy furniture movedj a few instances of
fallen plaster or damaged chimneysj damage
Slight.
VII. Everybody runs outdoors. Damage slight to
moderate in well-built ordinary structuresj some
chimneys broken. Noticed by persons driving
cars.
VIII. Alarm approaches panic. Damage considerable in
ordinary sUbstantial buildings with partial
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IX.
X.
XL
16
collapse. Fall of chimneys, factory stacKs,
columns, monuments, walls. Heavy furniture
overturned; disturbing to persons driving cars.
Panic general. Damage great in substantial
buildings, with partial collapse. Well design-
ed frame structures thrown out of plumb.
Buildings shifted off foundations. Ground
cracKed conspicuously.
Some well-built wooden structures destroyed;
most masonry and frame structures destroyed.
Ground badly cracKed. Landslides considerable
from river banKs and steep slopes.
Few if any masonry structures remain standing.
Bridges destroyed. Broad fissures in ground.
Earth slumps and land slips in soft ground.
XII. Damage total. Waves seen on ground surface.
Lines of sight and level distorted. Objects
thrown upward into the air.
Facility Classifications
The step of classifying facilities was fundamental
to the ATC-13 study, because earthquaKe-induced physical
damage is dependent upon structural properties. The
classification developed contains 78 classes of
structures, ~0 of which are buildings and 38 of which are
IX.
X.
16
collapse. Fall of chimneys, factory stacKs,
columns, monuments, walls. Heavy furniture
overturned; disturbing to persons driving cars.
Panic general. Damage great in substantial
buildings, with partial collapse. Well design-
ed frame structures thrown out of plumb.
Buildings shifted off foundations. Ground
cracKed conspicuously.
Some well-built wooden structures destroyed;
most masonry and frame structures destroyed.
Ground badly cracKed. Landslides considerable
from river banKs and steep slopes.
XI. Few if any masonry structures remain standing.
Bridges destroyed. Broad fissures in ground.
Earth slumps and land slips in soft ground.
XII. Damage total. Waves seen on ground surface.
Lines of sight and level distorted. Objects
thrown upward into the air.
Facility Classifications
The step of classifying facilities was fundamental
to the ATC-13 study, because earthquake-induced physical
damage is dependent upon structural properties. The
classification developed contains 78 classes of
structures, ~o of which are buildings and 38 of which are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
other structure types, such as bridges, pipelines, dams,
tunnels, etc. The classes were selected on the basis of
expected uniqueness in seismic performance. The facility
classes were assigned Facility Numbers to which the data
developed is referred. The facility numbers are not
consecutive, because facility classifications were added
after the expert questionnaire process was started, and
numbers assigned initially were maintained to preclude
error. The ATC-13 Facility Numbers for buildings only are
presented in Table II, following.
TABLE II
ATC-13 FACILITY CLASSES AND NUMBERS
FACILITY CLASS FACILITY HUMBER
Wood Frame (Low Rise)
Light Metal (Low Rise)
Unreinforced Masonry (Bearing Wall)
a) Low Rise (1-3 Stories)
b) Medium Rise (4-7 Stories)
Unreinforced Masonry (with Load Bearing Frame)
a) Low Rise
b) Medium Rise
c) High Rise (8+ Stories)
1
2
75
76
78
79
80
17
other structure types, such as bridges, pipelines, dams,
tunnels, etc. The classes were selected on the basis of
expected uniqueness in seismic performance. The facility
classes were assigned Facility Numbers to which the data
developed is referred. The facility numbers are not
consecutive, because facility classifications were added
after the expert questionnaire process was started, and
numbers assigned initially were maintained to preclude
error. The ATC-13 Facility Numbers for buildings only are
presented in Table II, following.
TABLE II
ATC-13 FACILITY CLASSES AND NUMBERS
FACILITY CLASS FACILITY NUMBER
Wood Frame (Low Rise)
Light Metal (Low Rise)
Unreinforced Masonry (Bearing Wall)
a) Low Rise (1-3 Stories)
b) Medium Rise (4-7 Stories)
Unreinforced Masonry (with Load Bearing Frame)
a) Low Rise
b) Medium Rise
c) High Rise (8+ Stories)
1
2
75
76
78
79
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reinforced Concrete Shear Wall
(with Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Reinforced Concrete Shear Wall
(without Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Reinforced Masonry Shear Wall
(without Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Reinforced Masonry Shear Wall
(with Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Braced Steel Frame
a) Low Rise
b) Medium Rise
c) High Rise
3
4
5
6
7
8
9
10
i1
84
85
86
12
13
14
18
Reinforced Concrete Shear Wall
(with Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Reinforced Concrete Shear Wall
(without Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Reinforced Masonry Shear Wall
(without Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Reinforced Masonry Shear Wall
(With Moment-Resisting Frame)
a) Low Rise
b) Medium Rise
c) High Rise
Braced Steel Frame
a) Low Rise
b) Medium Rise
c) High Rise
3
4
5
6
7
8
9
10
i1
84
85
86
12
13
14
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
Moment-Resisting Steel Frame (Perimeter Frame)
a) Low Rise 15
b) Medium Rise 16
c) High Rise 17
Moment-Resisting Steel Frame (Distributed Frame)
a) Low Rise 72
b) Medium Rise 73
c) High Rise 74
Moment-Resisting Ductile Concrete Frame
(Distributed Frame)
a) Low Rise 18
b) Medium Rise 19
C) High Rise 20
Moment-Resisting Non-Ductile Concrete Frame
(Distributed Frame}
a) Low Rise 87
b) Medium Rise 88 ..
c) High Rise 89
Precast Concrete (other than Tilt-up)
a) Low Rise 81
b) Medium Rise 82
c) High Rise 83
Long Span (Low Rise) 91
Tilt-up (Low Rise) 21
Mobile Homes 23
19
Moment-Resisting Steel Frame (Perimeter Frame)
a) Low Rise 15
b) Medium Rise 16
c) High Rise 17
Moment-Resisting Steel Frame (Distributed Frame)
a) Low Rise 72
b) Medium Rise 73
c) High Rise 74
Moment-Resisting Ductile Concrete Frame
(Distributed Frame)
a) Low Rise 18
b) Medium Rise 19
c) High Rise 20
Moment-Resisting Non-Ductile Concrete Frame
(Distributed Frame)
a) Low Rise 87
b) Medium Rise 88 ..
c) High Rise 89
Precast Concrete (other than Tilt-up)
a) Low Rise 81
b) Medium Rise 82
c) High Rise 83
Long Span (Low Rise) 91
Tilt-up (Low Rise) 21
Mobile Homes 23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
Physical Damage Caused by Ground Shaking
To express earthquake damage, ATC-13 introduced an
expression for Damage Factor (DF), the percentage of dollar
loss to buildings:
Damage Factor (DF) = Dollar Loss ( 2. 1) Replacement Value
Expected damage ($Damage) for a given facility is then
calculated as follows:
$Damage = DF x (Replacement Value)
Damage Factor is useful and flexible; other types of
earthquake loss can be rationally deduced from it. Ranges
of DF were correlated to Damage States; Central Damage
Factor (CDF) is defined as the midpoint of the DF range,
as shown in Table III, following.
20
Physical Damage Caused by Ground Shaking
To express earthquake damage, ATC-13 introduced an
expression for Damage Factor (DF) , the percentage of dollar
loss to buildings:
Damage Factor (DF) = Dollar Loss (2. 1) Replacement Value
Expected damage ($Damage) for a given facility is then
calculated as follows:
$Damage = DF x (Replacement Value)
Damage Factor is useful and flexiblej other types of
earthquake loss can be rationally deduced from it. Ranges
of DF were correlated to Damage Statesj Central Damage
Factor (CDF) is defined as the midpoint of the DF range,
as shown in Table III, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
TABLE III
ATC-13 DAMAGE STATE DESCRIPTIONS
DAMAGE STATE DF RANGE
1 - None 0
2 - Slight 0-1
3 - Light 1-10
4 - Moderate 10-30
5 - Heavy 30-60
6 - Major 60-100
7 - Destroyed 100
CDF
0
0. 5
5
20
45
80
100
DAMAGE DESCRIPTION
No damage.
Limited Localized minor damage not requiring repair.
Significant localized damage of some components generally not requiring repair.
Significant localized damage of many components warranting repair.
Extensive damage re-quiring major repairs.
Major widespread damage that may result in the facility being razed, demolished or repaired.
Total destruction of the majority of the facility.
Estimates of DF versus MMI were developed for all 78
facility classes, through a multiple questionnaire process
involving a Project Engineering Panel of 13 senior-level
specialists in earthquake engineering, plus 58 other
selected earthquake engineering experts. The questionnaire
process was modeled after the Delphi method for expert
21
TABLE III
ATC-13 DAMAGE STATE DESCRIPTIONS
DAMAGE STATE DF RANGE CDF DAMAGE DESCRIPTION
1 - None 0 0 No damage.
2 - Slight 0-1 o. 5 Limited Localized minor damage not requiring repair.
3 - Light 1-10 5 Significant localized damage of some components generally not requiring repair.
4 - Moderate 10-30 20 Significant localized damage of many components warranting repair.
5 - Heavy 30-60 45 Extensive damage re-quiring major repairs.
6 - Major 60-100 80 Major widespread damage that may result in the facility being razed, demolished or repaired.
7 - Destroyed 100 100 Total destruction of the majority of the facility.
Estimates of DF versus HMI were developed for all 78
facility classes, through a multiple questionnaire process
involving a Project Engineering Panel of 13 senior-level
specialists in earthquake engineering, plus 58 other
selected earthquake engineering experts. The questionnaire
process was modeled after the Delphi method for expert
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
opinion solicitation, developed for the Air Force by the
Rand Corporation in the 1950's. The procedure as used in
ATC-13 consisted of formulating and obtaining individual
answers to questionnaires from experts, iterating the
questionnaires two or three times while carefully
controlling the information feedback between rounds, and
finally aggregating the responses by statistical
operations.
One of the most important aspects of the Delphi
method is the selection of the experts. The experts were
carefully chosen by the Project Engineering Panel. As a
further safeguard each expert was asked to respond only on
the facility classes with which they had extensive
familiarity. This resulted in, out of a total of 71
selected experts, at most only eight experts giving
answers for any given facility class.
For each facility class, the experts were asked to
provide a low, best, and high estimate of DF at seven
shaking intensities, from MMI VI through XII. The low and
high estimates were defined to be the 90Y. probability
bounds of the DF distribution, while the best estimate was
defined for the experts as the DF most likely to be
observed for a given MMI and facility class. Each expert
was also asked to evaluate his level of experience in the
facility class being evaluated and to provide a
22
opinion solicitation, developed for the Air Force by the
Rand Corporation in the 1950's. The procedure as used in
ATC-13 consisted of formulating and obtaining individual
answers to questionnaires from experts, iterating the
questionnaires two or three times while carefully
controlling the information feedback between rounds, and
finally aggregating the responses by statistical
operations.
One of the most important aspects of the Delphi
method is the selection of the experts. The experts were
carefully chosen by the Project Engineering Panel. As a
further safeguard each expert was asked to respond only on
the facility classes with which they had extensive
familiarity. This resulted in, out of a total of 71
selected experts, at most only eight experts giving
answers for any given facility class.
For each facility class, the experts were asked to
provide a low, best, and high estimate of DF at seven
shaking intenSities, from MHI VI through XII. The low and
high estimates were defined to be the 90Y. probability
bounds of the DF distribution, while the best estimate was
defined for the experts as the DF most likely to be
observed for a given HHI and facility class. Each expert
was also asked to evaluate his level of experience in the
facility class being evaluated and to provide a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
self-evaluated degree of certainty in the low, best, and
high estimates, which were used as weighting factors in
the statistical compilation of the results.
Appendix G of ATC-13, "Weighted Statistics of Damage
Factor," is a complete listing of the expert opinion data,
including the weighted means of the experts' low, best, &
high DF estimates for the 78 facility classes, subjected
to each of the 7 levels of shaking intensity.
Death and Inlury Estimates
ATC-13 reviewed the existing literature to determine
the rates of deaths and injuries as a function of damage
to various facilities. On the basis of this information,
death and injury rates were developed as a function of
Central Damage Factor (CDF), shown in Table IV, following.
DAMAGE STATE
1 2 3 4 5 6 7
TABLE IV
ATC-13 INJURY AND DEATH RATES
CDF (Y.)
0 0. 5
5 20 45 80
100
SERIOUS INJURIES
0 1/250,000 1/25,000 1/2,500 1/250 1/25 2/5
FRACTION DEAD
0 1/1,000,000 1/100,000 1/10,000 1/1,000 1/100 1/5
23
self-evaluated degree of certainty in the low, best, and
high estimates, which were used as weighting factors in
the statistical compilation of the results.
Appendix G of ATC-13, "Weighted Statistics of Damage
Factor," is a complete listing of the expert opinion data,
including the weighted means of the experts' low, best, &
high DF estimates for the 78 facility classes, subjected
to each of the 7 levels of shaking intensity.
Death and Injury Estimates
ATC-13 reviewed the existing literature to determine
the rates of deaths and injuries as a function of damage
to various facilities. On the basis of this information,
death and injury rates were developed as a function of
Central Damage Factor (CDF) , shown in Table IV, following.
DAMAGE STATE
1 2 3 4 5 6 7
TABLE IV
ATC-13 INJURY AND DEATH RATES
CDF (X)
0 O. 5
5 20 45 80
100
SERIOUS INJURIES
o 1/250,000 1/25,000 1/2,500 1/250 1/25 2/5
FRACTION DEAD
o 1/1,000,000 1/100,000 1/10,000 1/1,000 1/100 1/5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
These estimates are for all types of construction
except light steel construction and wood-frame
construction. For light steel construction and wood-frame
construction, multiply all numerators by 0. 1.
Significance and Limitations
Earthquake loss estimation is more an art than a
science, because there is not sufficient observational
data on earthquake losses to characterize it as a hard
science. The ATC-13 loss estimates are based on judgment,
subject to the following limitations:
1. Damage estimates represent "average" conditions.
2. Damage estimates provided are for California.
3. Great amounts of experimental data from actual
earthquakes would be needed to verify or improve
these estimates.
Because the MMI scale was selected to characterize
the earthquake intensities used in the ATC-13 expert
opinions, the shaking intensity reference is not specific
to California. However, the building response is specific
to California, because building codes in California have
for many years required a higher level of seismic
resistant design than in most regions. This specificity of
the design basis of California buildings must be adjusted
to apply the results of ATC-13 to other regions.
These estimates are for all types of construction
except light steel construction and wood-frame
construction. For light steel construction and wood-frame
construction, multiply all numerators by 0.1.
Significance and Limitations
Earthquake loss estimation is more an art than a
SCience, because there is not sufficient observational
data on earthquake losses to characterize it as a hard
science. The ATC-13 loss estimates are based on judgment,
subject to the following limitations:
1. Damage estimates represent "average" conditions.
2. Damage estimates provided are for California.
3. Great amounts of experimental data from actual
earthquakes would be needed to verify or improve
these estimates.
Because the HHI scale was selected to characterize
the earthquake intensities used in the ATC-13 expert
opinions, the shaking intensity reference is not specific
to California. However, the building response is specific
to California, because building codes in California have
for many years required a higher level of seismic
resistant design than in most regions. This specificity of
the design basis of California buildings must be adjusted
to apply the results of ATC-13 to other regions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
Although the expert loss estimates were weighted and
statistically analyzed in a scientific manner, the loss
estimates themselves are not experimentally verifiable.
Despite its limitations, the ATC-13 loss estimates
reflect the best judgement of a group of highly prominent
earthquaKe engineers and represent the only available
information for a wide variety of structure types. The
seventy plus participants in the project represent more
than a thousand man-years of professional experience in
earthquaKe engineering; thus the judgement evaluations
made are of significance.
25
Although the expert loss estimates were weighted and
statistically analyzed in a scientific manner, the loss
estimates themselves are not experimentally verifiable.
Despite its limitations, the ATC-13 loss estimates
reflect the best judgement of a group of highly prominent
earthquaKe engineers and represent the only available
information for a wide variety of structure types. The
seventy plus participants in the project represent more
than a thousand man-years of professional experience in
earthquaKe engineeringi thus the judgement evaluations
made are of significance.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ATC-21: RAPID VISUAL SCREENING OF BUILDINGS FOR POTENTIAL SEISMIC HAZARDS: A HANDBOOK
ATC-21 (1988) was developed to provide a standard
rapid visual screening procedure (RSP) to identify those
buildings that might pose a potentially serious risK of
loss of life and injury, or of severe curtailment of
community services, in case of a damaging earthquaKe.
26
ATC-21 extends the loss estimating results of ATC-13, and
presents:
1. A Rapid Screening Procedure (RSP), based on
visual observation, to classify the building as
to structural type and identify significant
seismic-related defects in each building.
2. A Scoring System, based on the field survey data,
which relates to the probability of each building
sustaining major life-threatening structural
damage during a major earthquaKe.
Rapid Screening Procedure
ATC-21 utilizes a methodology based on a "sidewalK
survey" of a building using a Data Collection Form, which
an inspector completes based on visual observation of the
building. Three versions of Data Collection Form are
employed, for Low, Moderate and High Seismicity areas.
The methodology begins by collapsing the 40 building
classes of ATC-13 into 12 Building Types, based on the
ATC-21: RAPID VISUAL SCREENING OF BUILDINGS FOR POTENTIAL SEISMIC HAZARDS: A HANDBOOK
ATC-21 (1988) was developed to provide a standard
rapid visual screening procedure (RSP) to identify those
buildings that might pose a potentially serious risK of
loss of life and injury, or of severe curtailment of
community services, in case of a damaging earthquake.
26
ATC-21 extends the loss estimating results of ATC-13, and
presents:
1. A Rapid Screening Procedure (RSP) , based on
visual observation, to classify the building as
to structural type and identify significant
seismic-related defects in each building.
2. A Scoring System, based on the field survey data,
which relates to the probability of each building
sustaining major life-threatening structural
damage during a major earthquake.
Rapid Screening Procedure
ATC-21 utilizes a methodology based on a "sidewalK
survey" of a building using a Data Collection Form, which
an inspector completes based on visual observation of the
building. Three versions of Data Collection Form are
employed, for Low, Moderate and High Seismicity areas.
The methodology begins by collapsing the 40 building
classes of ATC-13 into 12 Building Types, based on the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
characteristics of the primary structural lateral load
resisting system. The 12 Building Types used, along with
their identifiers, are presented in Table V, following.
BUILDING IDENTIFIER
w S1
S2
S3
S4
C1
C2
C3/S5
PC1
PC2
RM
~M
TABLE V
ATC-21 BUILDING TYPES
DESCIPTIOH
Wood buildings of all types.
Steel moment resisting frames.
Braced steel frames.
Light metal buildings.
Steel frame with concrete shear walls.
Concrete moment resisting frames.
Concrete shear wall buildings.
Concrete or steel frame W/~M infill walls.
Tilt-up concrete buildings.
Precast concrete frame buildings.
Reinforced masonry.
Unreinforced masonry.
Eleven Performance Modifiers are identified as being
present or absent in each building. This is a significant
advance, making ATC-21 the first of the earthquake loss
estimation procedures to include the characteristics o£
27
characteristics of the primary structural lateral load
resisting system. The 12 Building Types used, along with
their identifiers, are presented in Table V, following.
BUILDING IDENTIFIER
W
S1
S2
S3
S4
C1
C2
C3/S5
PCI
PC2
RM
URM
TABLE V
ATC-2l BUILDING TYPES
DESCIPTIOH
Wood buildings of all types.
Steel moment resisting frames.
Braced steel frames.
Light metal buildings.
Steel frame with concrete shear walls.
Concrete moment resisting frames.
Concrete shear wall buildings.
Concrete or steel frame w/URM infill walls.
Tilt-up concrete buildings.
Precast concrete frame buildings.
Reinforced masonry.
Unreinforced masonry.
Eleven Performance Modifiers are identified as being
present or absent in each building. This is a significant
advance, making ATC-21 the first of the earthquake loss
estimation procedures to include the characteristics of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
design and construction of individual buildings into the
estimates. The Performance Modifiers are presented in
Table VI, following.
TABLE VI
ATC-21 PERFORMANCE MODIFIERS
HR High rise
PC Poor condition
VI Vertical irregularity
SS Soft story
T Torsion
PI Plan irregularity
P Pounding
LH Large heavy cladding
SC Short columns
PB Post benchmark year
SL Soil profile
28
Foundation type (deep vs. shallow foundations) has
not been included as a performance modifier because it is
impractical to include it as part of a "sidewalk" survey.
Also included on the Data Collection Form are number
of stories, square footage, year built, occupancy, number
of occupants, and non-structural falling hazards.
28
design and construction of individual buildings into the
estimates. The Performance Modifiers are presented in
Table VI, following.
TABLE VI
ATC-21 PERFORMANCE MODIFIERS
HR High rise
PC Poor condition
VI Vertical irregularity
SS Soft story
T Torsion
PI Plan irregularity
P Pounding
LH Large heavy cladding
SC Short columns
PB Post benchmark year
SL Soil profile
Foundation type (deep vs. shallow foundations) has
not been included as a performance modifier because it is
impractical to include it as part of a "sidewalk" survey.
Also included on the Data Collection Form are number
of stories, square footage, year built, occupancy, number
of occupants, and non-structural falling hazards.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
Seismicity
It has been necessary to correlate MMI intensity to
horizontal acceleration, which has become the commonly
accepted scientific measure of ground motion. Peak Ground
Acceleration (PGA) is defined as the maximum acceleration
recorded from an accelerogram on rock during an earthquake.
Effective Peak Acceleration (EPA) can be thought of as the
maximum acceleration after high frequency accelerations
(that do not affect sizeable structures) have been
discounted. ATC-21-1 uses the following correlations
between MMI, PGA, and EPA:
PGA (cm;sec2) = 1o(MMI-1)/3
EPA: 0.75 X PGA
The United States has been divided into 7 "HEHRP Map
Areas," corresponding to estimated maximum EPA levels
likely to occur during the life of a building. ATC-21
uses the HEHRP Map Areas to define it's Low, Moderate, and
High Seismicities, as shown on the map in Figure 1. Table
VII, following, shows the HEHRP Map Areas, the defining
EPA, and corresponding back-calculated PGA and MMI values.
29
Seismicity
It has been necessary to correlate MMI intensity to
horizontal acceleration, which has become the commonly
accepted scientific measure of ground motion. Peak Ground
Acceleration (PGA) is defined as the maximum acceleration
recorded from an accelerogram on rock during an earthquake.
Effective Peak Acceleration (EPA) can be thought of as the
maximum acceleration after high frequency accelerations
(that do not affect sizeable structures) have been
discounted. ATC-21-1 uses the following correlations
between MMI, PGA, and EPA:
PGA (cm/sec2 ) : 10(MMI-l)/3
EPA: 0.75 x PGA
The United States has been divided into 7 "HEHRP Map
Areas," corresponding to estimated maximum EPA levels
liKely to occur during the life of a building. ATC-21
uses the HEHRP Map Areas to define it's Low, Moderate, and
High Seismicities, as shown on the map in Figure 1. Table
VII, following, shows the HEHRP Map Areas, the defining
EPA, and corresponding back-calculated PGA and MMI values.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fif!ure 1. (ATC-21,
'I "' ~ ~
•:i c "' ffi ~
i ~
lo § 2 •i ~
I I' "' ::E ...J . I z
.---~:)',
\ ! :,_, ;j,'
NEHRP Map Areas of the United States 1988).
30
Figure 1. (ATC-21,
NEHRP Map Areas 1988) .
of
30
'I '" ~ ~ ~ I:i C ~ , z ~ .
I W . I, Cl <
~ ~ 2 'i ~
~ I I' .z ::IE ...J I Z
.... ~:)',
"-! ~ ij,-
--'..!.~-~
the United States
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
TABLE VII
ATC-21 SEISHICITIES AND EARTHQUAKE INTENSITIES
ATC-21 HEHRP SEISMICITY HAP AREA EPA PGA HHI
Low 1 . 05 . 07 6.4
Low 2 . 05 . 07 6.4
Moderate 3 . 10 . 13 7.4
Moderate 4 . 15 . 20 7.9
High 5 . 20 . 27 8. 3
High 6 . 30 . 40 8.78
High 7 . 40 . 53 9. 2
Scoring System
After identifying the NEHRP Hap area and Seismicity,
a Basic Structural Hazard (BSH) score, ranging from 1 to
8. 5, is assigned to each building, depending on the
Building Type and the NEHRP Hap Area (high BSH values
reflect good seismic performance, and low BSH values
reflect a seismically hazardous building). The Basic
Structural Hazard Scores are presented in Table VIII,
following.
31
TABLE VII
ATC-21 SEISMICITIES AND EARTHQUAKE INTENSITIES
ATC-21 NEHRP SEISMICITY HAP AREA EPA PGA HHI
Low 1 .05 .07 6.4
Low 2 .05 .07 6.4
Moderate 3 .10 . 13 7.4
Moderate 4 . 15 .20 7.9
High 5 .20 .27 8. 3
High 6 .30 .40 8.78
High 7 .40 .53 9. 2
Scoring System
After identifying the NEHRP Map area and Seismicity,
a Basic Structural Hazard (BSH) score, ranging from 1 to
8.5, is assigned to each building, depending on the
Building Type and the NEHRP Map Area (high BSH values
reflect good seismic performance, and low BSH values
reflect a seismically hazardous building). The Basic
Structural Hazard Scores are presented in Table VIII,
following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
32
TABLE VIII
ATC-21 BASIC STRUCTURAL HAZARD SCORES
NEHRP HAP AREAS
Low Moderate High BUILDING IDENTIFIER ( 1, 2) (3,4) (5,6,7)
w WOOD FRAME 8. 5 6.0 4. 5
S1 STEEL MRF 3. 5 4.0 4. 5
S2 BRACED STEEL FRAME 2. 5 3.0 3.0
S3 LIGHT METAL 6. 5 6.0 5. 5
S4 STEEL FRAME W/CONC sw 4. 5 4.0 3. 5
C1 RC MRF 4.0 3. 0 2.0
C2 RCSW NO MRF 4.0 3. 5 3.0
C3/S5 URM INFILL 3.0 2.0 1.5
PC1 TILT UP 3. 5 3. 5 2.0
PC2 PC FRAME 2. 5 2.0 1.5
RM REINFORCED MASONRY 4.0 3. 5 3. 0
URM UNREINFORCED MASONRY 2. 5 2.0 1.0
Each of the Performance Modifiers present in a
building is assigned a Performance Modification Factor
(PMF), from -2.5 to +2.0, depending on the specific
Performance Modifier, Building Type, and NEHRP Map Area
(most PMF's are "detractors," indicating a reduction in
the seismic performance of a building, and are therefore
32
TABLE VIII
ATC-21 BASIC STRUCTURAL HAZARD SCORES
NEHRP HAP AREAS
Low Moderate High BUILDING IDENTIFIER ( 1, 2) (3,4) (5,6,7)
w WOOD FRAME 8. 5 6.0 4. 5
Sl STEEL MRF 3. 5 4.0 4.5
S2 BRACED STEEL FRAME 2. 5 3.0 3.0
S3 LIGHT METAL 6. 5 6.0 5. 5
S4 STEEL FRAME W/CONC SW 4. 5 4.0 3. 5
Cl RC MRF 4.0 3. 0 2.0
C2 RCSW NO MRF 4.0 3. 5 3.0
C3/S5 URM INFILL 3.0 2.0 1.5
PCl TILT UP 3. 5 3. 5 2.0
PC2 PC FRAME 2. 5 2.0 1.5
RM REINFORCED MASONRY 4.0 3. 5 3.0
URM UNREINFORCED MASONRY 2. 5 2.0 1.0
Each of the Performance Modifiers present in a
building is assigned a Performance Modification Factor
(PMF), from -2.5 to +2.0, depending on the specific
Performance Modifier, Building Type, and NEHRP Map Area
(most PMF's are "detractors," indicating a reduction in
the seismic performance of a building, and are therefore
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33
negative values).
Finally, each building is assigned a Structural
Score (S), equal to the BSH score plus the sum of all the
PMF values for the building:
S : BSH + E(PMF)
The Structural Score s is a "performance score"
(higher numbers mean better seismic resistance), measuring
the liKlihood of building resistance to major damage
during a 1 arge earthquaKe.
ATC-21 recommends a Structural Score of 2 as a
"cut-off": that is, buildings with an S of 2 or less
should be flagged for detailed investigation by a
professional engineer experienced in seismic design. The
cut-off score is a cost-benefit division between the costs
of detailed engineering investigation versus the benefits
of increased seismic safety. The cut-off of S = 2
recommended by ATC-21 is a preliminary value, which could
be adjusted by a particular community based on its own
cost-benefit analysis of seismic safety.
Basis of the Scoring System
ATC-21's Structural ScoreS was developed from the
Damage Factor DF established in ATC-13. The Structural
Score s is defined as the negative of the logarithm (base
10) of the probability of damage (DF) exceeding 60-percent
of the building value for a specified HEHRP Effective PeaK
33
negative values).
Finally, each building is assigned a Structural
Score (S), equal to the BSH score plus the sum of all the
PMF values for the building:
S : BSH + E(PMF)
The Structural Score S is a "performance score"
(higher numbers mean better seismic resistance), measuring
the liKlihood of building resistance to major damage
during a I arge earthquaKe.
ATC-21 recommends a Structural Score of 2 as a
"cut-off": that is, buildings with an S of 2 or less
should be flagged for detailed investigation by a
professional engineer experienced in seismic design. The
cut-off score is a cost-benefit division between the costs
of detailed engineering investigation versus the benefits
of increased seismic safety. The cut-off of S : 2
recommended by ATC-21 is a preliminary value, which could
be adjusted by a particular community based on its own
cost-benefit analysis of seismic safety.
Basis of the Scoring System
ATC-21's Structural Score S was developed from the
Damage Factor DF established in ATC-13. The Structural
Score S is defined as the negative of the logarithm (base
10) of the probability of damage (DF) exceeding 50-percent
of the building value for a specified HEHRP Effective PeaK
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Acceleration (EPA) loading, as:
Structural ScoreS = log [Pr (DF >= 60Y.))
Sixty percent damage was selected as the generally
accepted threshold of major damage, roughly the point at
which many structures are a "total loss," and the
approximate lower bound at which there begins to be a
significant potential for building collapse and
significant life-safety threat.
34
ATC-21's development of S from DF was accomplished
by treating ATC-13's DF as a random variable, and modeling
it with a lognormal probability distribution. From
ATC-13's Appendix G (The Weighted Statistics of the Damage
Factor), the mean low and mean high estimates of the DF
were taKen as the 90Y. probability bounds on the damage
factor distribution, while the mean best estimate was
interpreted as the median DF. The value of the mean best
estimate of DF was multiplied by a Modification Constant
to adjust for differences in building practices for
non-California buildings. The probability of the DF being
greater than 60 percent was then calculated from a
polynomial approximation. MMI intensities were transformed
to EPA values, so the resulting structural scores could be
applied to NEHRP Map Areas.
Where several ATC-13 building types correspond to
one in ATC-21, the results were averaged. Inconsistencies
Acceleration (EPA) loading, as:
Structural Score S = log [Pr (DF >= 60Y.))
Sixty percent damage was selected as the generally
accepted threshold of major damage, roughly the point at
which many structures are a "total loss," and the
approximate lower bound at which there begins to be a
significant potential for building collapse and
significant life-safety threat.
34
ATC-21's development of S from DF was accomplished
by treating ATC-13's DF as a random variable, and modeling
it with a lognormal probability distribution. From
ATC-13's Appendix G (The Weighted StatistiCS of the Damage
Factor), the mean low and mean high estimates of the DF
were taKen as the gOY. probability bounds on the damage
factor distribution, while the mean best estimate was
interpreted as the median DF. The value of the mean best
estimate of DF was multiplied by a Modification Constant
to adjust for differences in building practices for
non-California buildings. The probability of the DF being
greater than 60 percent was then calculated from a
polynomial approximation. HMI intensities were transformed
to EPA values, so the resulting structural scores could be
applied to NEHRP Hap Areas.
Where several ATC-13 building types correspond to
one in ATC-21, the results were averaged. Inconsistencies
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
were smoothed and adjusted on the basis of judgment, and
the final BSH Score S rounded to the nearest 0. 5.
Significance and Limitations
35
The landmar~ achievement of ATC-21 is the development
of the first methodology that rates the performance of
individual buildings, accounting for structural type,
expected seismicity, and combinations of individual
variations in building design and construction
characteristics.
However, ATC-21/RSP is by definition an approximate
procedure. The goal is to broadly identify most of the
potentially seismically hazardous buildings, at a
relatively modest expenditure of time and effort, and to
eliminate most of the relatively adequate buildings from
further review. It is intended for rapidly evaluating the
hundreds or thousands of buildings in a community, and is
definitely not intended for the full determination of the
seismic safety of individual buildings.
ATC-21's Structural Score s was designed to be a
cutoff score for buildings with a potential life-safety
ris~ during major earthquaKes. Despite utilizing the
Damage Factor of ATC-13 in developing S, ATC-21 does not
model damage. A continuous function does not exist,
allowing DF to be calculated from S, structure type, and
earthquaKe intensity.
were smoothed and adjusted on the basis of judgment, and
the final BSH Score S rounded to the nearest O. 5.
Significance and Limitations
35
The landmar~ achievement of ATC-21 is the development
of the first methodology that rates the performance of
individual buildings, accounting for structural type,
expected seismicity, and combinations of individual
variations in building design and construction
characteristics.
However, ATC-21/RSP is by definition an approximate
procedure. The goal is to broadly identify most of the
potentially seismically hazardous buildings, at a
relatively modest expenditure of time and effort, and to
eliminate most of the relatively adequate buildings from
further review. It is intended for rapidly evaluating the
hundreds or thousands of buildings in a community, and is
definitely not intended for the full determination of the
seismiC safety of individual buildings.
ATC-21's Structural Score 8 was designed to be a
cutoff score for buildings with a potential life-safety
ris~ during major earthquakes. Despite utilizing the
Damage Factor of ATC-13 in developing 8, ATC-21 does not
model damage. A continuous function does not eXist,
allowing DF to be calculated from S, structure type, and
earthquake intensity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
PORTLAND SEISMIC HAZARDS SURVEY (PSHS)
In 1993, Portland State University's Department o£
Civil Engineering, funded by METRO, conducted a survey of
the seismic hazards o£ 4,500 non-residential buildings in
the USGS Portland Quadrangle (Rad and McCormack 1994).
Eight sections in the SE part of the quadrangle
sections 1N1E33, 34, 35,36 and 1S1E01,02,03,04 ---were
inventoried by the Bureau of Buildings, City of Portland,
bringing the total number of buildings surveyed in the
Portland Quadrangle to about 10,000. In 1995, an
additional 12,000 buildings were surveyed by the PSU/CE
team in the Mt. Tabor, Gladstone, Lake Oswego, and Linnton
Quadrangles. This may be the largest survey of its kind
made anywhere to date.
About a dozen civil engineering graduate and
upper-division students were trained and supervised by PSU
faculty conducting the survey, utilizing ATC-21's Rapid
Screening Procedure (RSP). In the 1995 phase of the work,
about twice that many students were employed.
Building Inventory Database
The PSU team created an inventory of the survey data
on the database ACCESS, wherein one data-line is used for
each building. Each data-line contains the map section
number, survey sequence number, tax lot parcel (R) number,
36
PORTLAND SEISMIC HAZARDS SURVEY (PSHS)
In 1993, Portland State University's Department of
Civil Engineering, funded by METRO, conducted a survey of
the seismic hazards of 4,500 non-residential buildings in
the USGS Portland Quadrangle (Rad and McCormack 1994).
Eight sections in the SE part of the quadrangle
sections 1N1E33, 34, 35, 36 and 1S1E01,02,03,04 --- were
inventoried by the Bureau of Buildings, City of Portland,
bringing the total number of buildings surveyed in the
Portland Quadrangle to about 10,000. In 1995, an
additional 12,000 buildings were surveyed by the PSU/CE
team in the Mt. Tabor, Gladstone, Lake Oswego, and Linnton
Quadrangles. This may be the largest survey of its kind
made anywhere to date.
About a dozen civil engineering graduate and
upper-division students were trained and supervised by PSU
faculty conducting the survey, utilizing ATC-21's Rapid
Screening Procedure (RSP). In the 1995 phase of the worK,
about twice that many students were employed.
Building Inventory Database
The PSU team created an inventory of the survey data
on the database ACCESS, wherein one data-line is used for
each building. Each data-line contains the map section
number, survey sequence number, tax lot parcel (R) number,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37
address, year built, area, city use code, number of
stories, building name, use, occupancy, estimated range of
number of people, state occupancies (special, major,
essential, and hazardous), non-structural falling hazards,
RSP building type, and RSP performance modifiers. A sample
database spreadsheet page is shown in Figure 2 (with no
R-number, address nor building name, but with the three
soil modifers added).
In the database, the Building Type (using the
symbols indicated previously) was recorded, and the
Performance Modifiers were identified as either present
(Y) or absent (blank). The survey did not record the
ATC-21 Soil Profile (SL) modifer for soil effects. Three
performance modifiers specific to soil profiles in
Portland, based on the Oregon DOGAMI Earthquake Hazard
Maps (GMS-79 1993), were developed as part of the present
study. These were matched to the specific building sites
and added to the database.
Range of Estimated Humber of People
The ATC-21 estimated ranges of Humber of Persons was
expanded for use in the Portland Seismic Hazards Survey,
as shown in Table IX, following.
37
address, year built, area, city use code, number of
stories, building name, use, occupancy, estimated range of
number of people, state occupancies (special, major,
essential, and hazardous), non-structural falling hazards,
RSP building type, and RSP performance modifiers. A sample
database spreadsheet page is shown in Figure 2 (with no
R-number, address nor building name, but with the three
soil modifers added).
In the database, the Building Type (using the
symbols indicated previously) was recorded, and the
Performance Modifiers were identified as either present
(Y) or absent (blank). The survey did not record the
ATC-21 Soil Profile (SL) modi fer for soil effects. Three
performance modifiers specific to soil profiles in
Portland, based on the Oregon DOGAMI Earthquake Hazard
Maps (GMS-79 1993), were developed as part of the present
study. These were matched to the specific building Sites
and added to the database.
Range of Estimated Humber of People
The ATC-21 estimated ranges of Humber of Persons was
expanded for use in the Portland Seismic Hazards Survey,
as shown in Table IX, following.
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited without perm
ission.
Cll~ tD ..... ..... ~~ ..... tD 0
N ::x:· Ill N ~> p.r.o
:~ ~:; <: ID'tl "<Ill au P,ID Ill r+l-h
~6 Ill 8 to tDr+
::t tD
.... \0 \0 (JJ
'tl 0 '":S r+ ..... ~ p.
Section 1N1EOB 1N1EOB 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1EOB
-
Seq. Year No.
1 1895 2 1925 3 1925 4 1957 5 27 6 1895 7 1922 8 1925 9 1970 10 1973 11 1950 12 1973 13 1949 14 47 15 1955 16 1901 17 57 18 1965 19 1924 20 1970 21 1904 22 1947 23 1929 24 1896 25 1908 26 1908 27 1925 28 1924 29 1925 30 1980 31 1906
Area No. Use Occupancy Sl
2970 2 House Residential 2860 1 Vacant Commercial 630 1 Tavern Commercial 1986 1 Restaurant Commercial 2436 2 Firestation Emer. Serv. 6668 2 Community Office 2116 1 Community Residential 2500 1 Shop Commercial 6000 1 Office Office 4000 1 Mental He Office 3440 1 Shop Commercial 1288 1 Restaurant Commercial 1650 1 Tavern Commercial 1432 1 Insurance/ Commercial 6230 1 Grocery St Commercial 9356 3 Church Pub.Assem. 3060 1 Church Pub. Assem. 2100 1 Shop Commercial 1540 1 AutoRepal Commercial 3772 1 Church Pub.Assem. 6596 2 Car Dealer Commercial 5792 1 Warehous Industrial 936 1 Realtor Commercial 2516 2 Office Office 2756 1 Realtors Commercial 10932 2 Shop Commercial 6037 1 Shop Commercial 1970 1 Shop Commercial
12180 3 Shop Commercial 3600 1 Restaurant Commercial 7291 2 Shop Commercial
Number Sl NSFH Str. Mixed People Occ. Type
0-10 y w 0-10 C1 C2 11·30 y w 11-30 y w 0-10 Esse y w 11·30 y w 11-30 w 11-30 y URM C2 11-30 RM
30-100 y RM 0-10 y w 11-30 RM 11·30 y RM 11-30 y URM w 11-30 C2
30-100 y w 30-100 y w 0-10 RM 0-10 w
30-100 y w 11·30 y w 11-30 y C2 0-10 y w 0-10 y w 0-10 y w
30-100 w 30-100 y C2 11-30 y w
30-100 y C2 11·30 RM 11·30 y w
PERFORMANCE MODIFIERS HR PC VI SS T PI P LH sc PB
y y y y y
yy
y y y
y y
y y y
y y y y
y y y
y y y
y yy
y y yy
y y y y
y y
SMF LF AF 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.4 0 0.2 0 0.2 0 0.2 0 0.2 0 0.2 0 0.4 0 0.4 0 0.2 0 0.2 0 0.2 0 0.4 0 0.4 0 0.4
SF 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
(JJ ())
Section Seq. Year Area No. Use Occupancy Number St NSFH Str. Mixed PERFORMANCE MODIFIERS SMF 1N1EOB No. St People Occ. Type HR PC VI S5 T PI P lH SC PB IF AF SF 1N1E08 1 1895 2970 2 House Residential 0-10 Y W Y Y 0 0.4 0 1N1E08 2 1925 2860 1 Vacant Commercial 0-10 C1 C2 Y Y 0 0.4 0 1N1E08 3 1925 630 1 Tavem Commercial 11-30 Y W Y 0 0.4 0 1N1E08 4 1957 1986 1 Reslaurant Commercial 11-30 Y W YY 0 0.4 0 1N1E08 5 27 2436 2 Firestation Emer. Servo 0-10 Esse Y W 0 0.4 0 1N1E08 6 1895 6668 2 Community Office 11-30 Y W 0 0.4 0 1N1E08 7 1922 2116 1 Community Residential 11-30 W Y 0 0.4 0 1N1E08 8 1925 2500 1 Shop Commercial 11-30 Y URM C2 Y Y 0 0.4 0 1N1E08 9 1970 6000 1 Office Office 11-30 RM Y 0 0.4 0 1N1E08 10 1973 4000 1 Mental He Office 30-100 Y RM Y 0 0.4 0 1N1E08 11 1950 3440 1 Shop Commercial 0-10 V W V V 0 0.4 0 1N1E08 12 1973 1288 1 Restaurant Commercial 11-30 RM Y 0 0.4 0 1N1E08 13 1949 1650 1 Tavem Commercial 11-30 V RM V Y 0 0.4 0 1N1E08 14 47 1432 1 Insurance/ Commercial 11-30 V URM W V Y 0 0.4 0 1N1E08 15 1955 6230 1 Grocery 51 Commercial 11-30 C2 V 0 0.4 0 1N1E08 16 1901 9356 3 Church Pub. Assem. 30-100 V W Y Y 0 0.4 0 1N1E08 17 57 3060 1 Church Pub. Assem. 30-100 V W Y 0 0.4 0 1N1E08 18 1965 2100 1 Shop Commercial 0-10 RM V 0 0.4 0 1N1E08 19 1924 1540 1 Auto Repal Commercial 0-10 W Y 0 0.2 0 1N1E08 20 1970 3772 1 Church Pub. Assem. 30-100 V W 0 0.2 0 1N1E08 21 1904 6596 2 Car Dealer Commercial 11-30 V W V 0 0.2 0 1N1E08 22 1947 5792 1 Warehous Industrial 11-30 V C2 YV 0 0.2 0 1N1E08 23 1929 936 1 Realtor Commercial 0-10 V W 0 0.2 0 1N1E08 24 1896 2516 2 Office Office 0-10 V W 0 0.4 0 1N1E08 25 1908 2756 1 Reallors Commercial 0-10 V W V 0 0.4 0 lN1E08 26 1908 10932 2 Shop Commercial 30-100 W Y 0 0.2 0 lN1E08 27 1925 6037 1 Shop Commercial 30-100 V C2 VY 0 0.2 0 lN1E08 28 1924 1970 1 Shop Commercial 11-30 V W V 0 0.2 0 lN1E08 29 1925 12180 3 Shop Commercial 30-100 V C2 Y 0 0.4 0 lN1EOB 30 1980 3600 1 Restauranl Commercial 11-30 RM Y V 0 0.4 0 1N1E08 31 1906 7291 2 Shop Commercial 11-30 V W Y Y 0 0.4 0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE IX
RANGES OF ESTIMATED HUMBER OF PEOPLE USED IN ATC-21 AND PSHS-93
ATC-21
0-10 11-100 100+
PSHS-93
0-10 11-30 31-100 100-500 500-2000 2000-5000 5000+
39
The range o£ number o£ people was based on the field
surveyors estimate o£ the building occupant load while the
building is in "full use." This occupant load would, £or
most o£ the buildings surveyed, occur during regular
business hours, 9:00AM - 5:00PM Monday through Friday.
Post-Benchmark Year
The Post Benchmark Year £or each building type used
in the survey are presented in Table X, following.
The post benchmark years were determined in consultation
with the City o£ Portland Bureau o£ Buildings.
TABLE IX
RANGES OF ESTIMATED NUMBER OF PEOPLE USED IN ATC-21 AND PSHS-93
ATC-21
0-10 11-100 100+
PSHS-93
0-10 11-30 31-100 100-500 500-2000 2000-5000 5000+
39
The range of number of people was based on the field
surveyors estimate of the building occupant load while the
building is in "full use." This occupant load would, for
most of the buildings surveyed, occur during regular
business hours, 9:00 AM - 5:00 PM Monday through Friday.
Post-Benchmark Year
The Post Benchmark Year for each building type used
in the survey are presented in Table X, following.
The post benchmark years were determined in consultation
with the City of Portland Bureau of Buildings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE X
POST-BENCHMARK YEARS USED IN PSHS-93
BUILDING TYPE
w Si S2 S3 S4 C1 C2 C3/S5 PC1 PC2 RM URM
Significance and Limitations
POST BENCHMARK YEAR
1965 1978 1990 N/A 1978 1978 1978 N/A 1975 N/A 1978 N/A
This is probably the largest ATC-21/RSP survey
completed to date. The earthquake loss estimates and
retrofit programming in the present research are,
therefore, based on the most complete and detailed
building inventory data available.
This worK refers to a scoring system that has a
40
probabilistic basis. Therefore, it should not be used to
tag specific buildings as hazardous or safe. It is
intended for use in estimating losses for categories or
classes of buildings.
TABLE X
POST-BENCHMARK YEARS USED IN PSHS-93
BUILDING TYPE
W S1 S2 S3 S4 C1 C2 C3/S5 PCl PC2 RM URN
Significance and Limitations
POST BENCHMARK YEAR
1965 1978 1990 N/A 1978 1978 1978 N./A 1975 N/A 1978 N/A
This is probably the largest ATC-21/RSP survey
completed to date. The earthquake loss estimates and
retrofit programming in the present research are,
therefore, based on the most complete and detailed
building inventory data available.
This worK refers to a scoring system that has a
40
probabilistic basis. Therefore, it should not be used to
tag specific buildings as hazardous or safe. It is
intended for use in estimating losses for categories or
classes of buildings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
STATE OF OREGON SEISMIC DESIGN MAPPING
Geomatrix Consultants (1995) of San Francisco,
California, presented to the Oregon Department of
Transportation a probabilistic assessment of seismic
hazard in the state of Oregon. The probabilistic approach
to mapping earthquake hazards is a standard approach and
is used by the U.S. Geological Survey to prepare national
seismic hazard maps that form the basis for setting
seismic design levels in most building codes. The
Geomatrix report is the best currently available summary
of the seismicity of Oregon, applicable to any type of
facility, and includes:
1. A detailed desciption of the characterization of
earthquaKe sources throughout Oregon.
2. A set of seismic hazard maps showing contours of
peaK ground acceleration (PGA) with 500, 1,000,
and 2, 500 year returns.
3. Detailed seismic hazard analyses for fifteen
locations in Oregon (including Portland).
Earthquake Sources
The principal tectonic feature of the Pacific
Northwest is the active subduction zone formed by the Juan
de Fuca Plate subducting beneath the North American Plate
along the Cascadia margin, the western coast of North
41
STATE OF OREGON SEISMIC DESIGN MAPPING
Geomatrix Consultants (1995) of San Francisco,
California, presented to the Oregon Department of
Transportation a probabilistic assessment of seismic
hazard in the state of Oregon. The probabilistic approach
to mapping earthquake hazards is a standard approach and
is used by the U.S. Geological Survey to prepare national
seismic hazard maps that form the basis for setting
seismic design levels in most building codes. The
Geomatrix report is the best currently available summary
of the seismicity of Oregon, applicable to any type of
facility, and includes:
1. A detailed desciption of the characterization of
earthquake sources throughout Oregon.
2. A set of seismic hazard maps showing contours of
peak ground acceleration (PGA) with 500, 1,000,
and 2, 500 year returns.
3. Detailed seismic hazard analyses for fifteen
locations in Oregon (including Portland).
Earthquake Sources
The principal tectonic feature of the Pacific
Northwest is the active subduction zone formed by the Juan
de Fuca Plate subducting beneath the North American Plate
along the Cascadia margin, the western coast of North
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
America stretching from northern California to central
Vancouver Island. In 1987, evidence was discovered for
what appear to have been repeated large magnitude
earthquakes along the coast of Oregon, Washington, and
northern California in the recent geological past (Atwater
1987). These large earthquakes are now attributed to the
subduction zone.
Three possible earthquake source zones that could
affect the Portland are outlined in the Geomatrix
report ( 1995):
1. Cascadia subduction zone earthquakes which can
occur on the interface between the Juan de Fuca
Plate and the North American Plate, called
"interplate" sources.
2. Deep earthquakes which can occur within the
subducting plate, called "intraplate" sources.
3. Earthquakes that occur on shallow crustal faults
within the North American Plate, called "crustal"
sources.
Seismic Hazard Analysis
The seismic hazard maps and detailed hazard analyses
determine annual exceedance probabilities based on
combining the probabilities of occurrance from the three
earthquake sources. The PGA on rocK for 100, 500, 1,000,
and 2, 500 year return periods for Portland are presented
42
America stretching from northern California to central
Vancouver Island. In 1987, evidence was discovered for
what appear to have been repeated large magnitude
earthquakes along the coast of Oregon, Washington, and
northern California in the recent geological past (Atwater
1987). These large earthquakes are now attributed to the
subduction zone.
Three possible earthquake source zones that could
affect the Portland are outlined in the Geomatrix
report (1995):
1. Cascadia subduction zone earthquakes which can
occur on the interface between the Juan de Fuca
Plate and the North American Plate, called
"interplate" sources.
2. Deep earthquakes which can occur within the
subducting plate, called "intraplate" sources.
3. Earthquakes that occur on shallow crustal faults
within the North American Plate, called "crustal"
sources.
Seismic Hazard Analysis
The seismic hazard maps and detailed hazard analyses
determine annual exceedance probabilities based on
combining the probabilities of occurrance from the three
earthquake sources. The PGA on rocK for 100, 500, 1,000,
and 2, 500 year return periods for Portland are presented
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in Table XI, following.
TABLE XI
PGA ON ROCK AND RECURRENCE INTERVALS FOR PORTLAND (GEOMATRIX CONSULTANTS 1995)
RECURRENCE INTERVAL (years)
100
500
1,000
2,500
PEAK GROUND ACCELERATION ON ROCK
(PGA as :1. of g)
. 08
. 20
. 27
. 39
Limitations and Significance
Characterization of the magnitude of ground motion
and corresponding recurrence intervals is a fundamental
first step in any earthquake loss estimation methodology.
The Geomatrix report (1995) is the best currently
available summary of the probabilistic seismology of
Oregon.
in Table XI, following.
TABLE XI
PGA ON ROCK AND RECURRENCE INTERVALS FOR PORTLAND (GEOMATRIX CONSULTANTS 1995)
RECURRENCE INTERVAL (years)
100
500
1,000
2,500
PEAK GROUND ACCELERATION ON ROCK
(PGA as Yo of g)
.08
.20
.27
.39
Limitations and Significance
Characterization of the magnitude of ground motion
and corresponding recurrence intervals is a fundamental
first step in any earthquake loss estimation methodology.
The Geomatrix report (1995) is the best currently
available summary of the probabilistic seismology of
Oregon.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
EARTHQUAKE HAZARD MAPS OF THE PORTLAND QUADRANGLE
The Oregon Department of Geology and Mineral
Industries (1993) published an earthquake hazard map
series, designated GMS-79, consisting of 4 maps of the
USGS Portland Quadrangle, as follows:
1. Liquefaction Susceptibility Map (Plate 1)
2. Ground Motion Amplification Map (Plate 2)
3. Lateral Spread Displacement and Dynamic Slope
Instability Map (Plate 3)
4. Relative Earthquake Hazard Map
These maps delineate hazards due to site-specific
soil conditions derived from a three-dimensional digital
geologic model coupled with borehole records, geotechnical
properties, and shear-wave velocity data collected from
purpose-drilled boreholes at 20 locations. Each map is
color-coded for gradations in the severity of hazards
depicted, as shown in Figures 3, 4, and 5, following.
Liquefaction Susceptibility Map
Liquefaction-induced ground failure is a major cause
of earthquake damage. Liquefaction is a phenomenon during
which ground shaking causes a complete loss of shear
strength in loose, water-saturated granular soils,
typically fine sands and silty sands near river channels.
Foundations supported on liquefied soils may sink or be
EARTHQUAKE HAZARD MAPS OF THE PORTLAND QUADRANGLE
The Oregon Department of Geology and Mineral
Industries (1993) published an earthquake hazard map
series, designated GMS-79, consisting of 4 maps of the
USGS Portland Quadrangle, as follows:
1. Liquefaction Susceptibility Map (Plate 1)
2. Ground Motion Amplification Map (Plate 2)
3. Lateral Spread Displacement and Dynamic Slope
Instability Map (Plate 3)
4. Relative Earthquake Hazard Map
These maps delineate hazards due to site-specific
soil conditions derived from a three-dimensional digital
geologic model coupled with borehole records, geotechnical
properties, and shear-wave velocity data collected from
purpose-drilled boreholes at 20 locations. Each map is
color-coded for gradations in the severity of hazards
depicted, as shown in Figures 3, 4, and 5, following.
Liquefaction Susceptibility Map
Liquefaction-induced ground failure is a major cause
of earthquake damage. Liquefaction is a phenomenon during
which ground shaking causes a complete loss of shear
strength in loose, water-saturated granular SOils,
typically fine sands and silty sands near river channels.
Foundations supported on liquefied soils may sink or be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 3. (GMS-79,
~ ••. '..... .... . . . . • •• . ..... I·- •.
' ' ...... ~ -·· .. ;. -~ ...... ----~ . -
........... r-···· ! .. : .. ---.. ···: .. . . . -..... ·~ .
.•.• , ..... lit·~~ ~r ~ ~~ = f l
':~···1f··· ... ' l'f.r,~~---··iH,_[~~:T~::~:T ..... T ..
·--··:r:·:~-.-·::,:.·:.--.--·~~;::: ... : · .. : :
Liquefaction Susceptibility Map Plate 1).
45
Figure 3. (GMS-79.
' .. -... ; -........ ~ ....... __ .... -
.... --..... ~ -" .: .. ,. -. -........ ~ .
• ~,!-"" . ~.- • .. ,
l.~:T~::~:T'··· .. , .... _ .. _:!.:.J. ___ ",:"_ .. _ .... _1'.'
" . . " ' i i ,," i"··y' .. ··, .. · ..
Liquefaction Susceptibility Map Plate 1).
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4. Ground Motion Amplification Map (GMS-79, Plate 2).
46
Figure 4. Ground Motion Amplification Map (GHS-79, Plate 2).
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
---.........
--- ·-·-.
-.'
. i
... : .... ·:·. ~. '• . '• ..... ,
··.:
Figure 5. Lateral Spread Displacement and Dynamic Slope Instability Map (GMS-79 Plate 3). Figure 5. Lateral Spread Displacement and Dynamic Slope Instability Hap (GHS-79 Plate 3).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
pulled apart laterally. The amount of permanent ground
displacement resulting from liquefaction is a function of
the thickness of the liquefiable layer. The thickest and
most liquefiable deposits in the Portland Quadrangle lie
beneath the extended flood plains of the Columbia and
Willamette Rivers and generally conform to the lowlands
along those rivers.
The purpose of the Liquefaction Susceptibility Map
(Plate 1) is to show the areas likely to be underlain by
liquefiable deposits. Plate 1 indicates lique-faction
susceptibility by showing color codes indicating
combinations of thickness of liquefiable sediment and
depth to water table, as shown in Table XII, following.
TABLE XII
LIQUEFACTION SUSCEPTIBILITY FROM GMS-79 PLATE 1
MAP THICKNESS OF DEPTH TO COLOR LIQUEFIABLE SEDIMENT WATER TABLE
Red • More than 30-ft Less than 15-ft
Pink D More than 30-ft 15-ft to 30-ft
Orange D 10-ft to 20-£t Less than 15-ft
Yellow D 10-£t to 20-£t 15-£t to 30-ft
Dark Green • Less than 10-ft Less than 15-ft
Light Green D Less than 10-£t 15-ft to 30-ft
48
pulled apart laterally. The amount of permanent ground
displacement resulting from liquefaction is a function of
the thickness of the liquefiable layer. The thickest and
most liquefiable deposits in the Portland Quadrangle lie
beneath the extended flood plains of the Columbia and
Willamette Rivers and generally conform to the lowlands
along those rivers.
The purpose of the Liquefaction Susceptibility Map
(Plate 1) is to show the areas likely to be underlain by
liquefiable deposits. PI ate 1 indicates liquefaction
susceptibility by showing color codes indicating
combinations of thickness of liquefiable sediment and
depth to water table, as shown in Table XII, following.
TABLE XII
LIQUEFACTION SUSCEPTIBILITY FROM GMS-79 PLATE 1
HAP THICKNESS OF DEPTH TO COLOR LIQUEFIABLE SEDIMENT WATER TABLE
Red • More than 30-ft Less than 15-ft
Pink D More than 30-ft 15-ft to 30-ft
Orange D 10-ft to 20-ft Less than 15-ft
Yellow D 10-ft to 20-ft 15-ft to 30-ft
Dark Green Less than 10-ft Less than 15-ft
Light Green Less than 10-ft 15-ft to 30-ft
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
The order shown in Table XII is roughly from worst
to decreasing liquefaction hazard. Liquefaction seldom
occurs in areas where the depth to the water table exceeds
30-feet.
Lateral Spread Displacement Map
Lateral spread is one of the most widespread and
devastating forms of ground failure triggered by
liquefaction. These failures are characterized by surface
soils underlain by liquefied soils that slide laterallY
down mild slopes or toward a free face, such as a river
channel bank. Differential movements within the spreading
mass usually create such features as fissures, scarps, and
grabens. Lateral displacements may range from a few
centimeters to several meters, and differential vertical
movements (settlements) may be up to about half of the
horizontal movement. These displacements commonly inflict
severe damage to structures built on the unstable terrain.
Lateral spread displacements are driven by a
combination of static gravitational and dynamic seismic
forces. Localities most vulnerable to lateral spread lie
near river channels or on gently sloping ground underlain
by loose, water-saturated granular sediments.
The purpose of the Lateral Spread Displacement Map
(Plate 3) is to quantitatively estimate maximum lateral
ground displacements, by combining ground slope and/or
49
The order shown in Table XII is roughly from worst
to decreasing liquefaction hazard. Liquefaction seldom
occurs in areas where the depth to the water table exceeds
30-feet.
Lateral Spread Displacement Map
Lateral spread is one of the most widespread and
devastating forms of ground failure triggered by
liquefaction. These failures are characterized by surface
soils underlain by liquefied soils that slide laterally
down mild slopes or toward a free face. such as a river
channel bank. Differential movements within the spreading
mass usually create such features as fissures. scarps. and
grabens. Lateral displacements may range from a few
centimeters to several meters. and differential vertical
movements (settlements) may be up to about half of the
horizontal movement. These displacements commonly inflict
severe damage to structures built on the unstable terrain.
Lateral spread displacements are driven by a
combination of static gravitational and dynamic seismic
forces. Localities most vulnerable to lateral spread lie
near river channels or on gently sloping ground underlain
by loose, water-saturated granular sediments.
The purpose of the Lateral Spread Displacement Map
(Plate 3) is to quantitatively estimate maximum lateral
ground displacements. by combining ground slope and/or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
proximity to a free face with the liquefaction hazard
data, utilizing a technique developed by Bartlett and Youd
(1992). Bartlett and Youd formed empirical regression
models relating ground displacement to seismic,
topographic, and geotechnical factors with data compiled
from more than 450 sites where lateral spread has
occurred. Plate 3 color-codes estimated maximum lateral
displacement, as shown in Table XIII, following.
TABLE XIII
LATERAL SPREAD DISPLACEMENT FROM GMS-79 PLATE 3
MAP COLOR GROUND DISPLACEMENT
White 0
Light Green D 1 to 2 feet
Yellow D 2 to 3 feet
Orange D 3 to 4 feet
Red • more than 4 feet
Ground Motion Amplification Map
Recent earthquaKes have shown the effect of local
ground shaKing amplification on the distribution of major
damage. Unconsolidated geologic materials tend to amplify
the ground shaKing from earthquaKes, as much as 2 to 3
times the accelerations of underlying bedroc~ (~i 1988).
50
proximity to a free face with the liquefaction hazard
data, utilizing a technique developed by Bartlett and Youd
(1992). Bartlett and Youd formed empirical regression
models relating ground displacement to seismic, .-
topographic, and geotechnical factors with data compiled
from more than 450 sites where lateral spread has
occurred. Plate 3 color-codes estimated maximum lateral
displacement, as shown in Table XIII, following.
TABLE XIII
LATERAL SPREAD DISPLACEMENT FROM GMS-79 PLATE 3
MAP COLOR GROUND DISPLACEMENT
White 0
Light Green 0 1 to 2 feet
Yellow 0 2 to 3 feet
Orange D 3 to 4 feet
Red • more than 4 feet
Ground Motion Amplification Map
Recent earthquaKes have shown the effect of local
ground shaKing amplification on the distribution of major
damage. Unconsolidated geologic materials tend to amplify
the ground shaKing from earthquaKes, as much as 2 to 3
times the accelerations of underlying bedrocK (AKi 1988).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The amount of amplification depends primarily on the
thickness and shear-wave velocity of the unconsolidated
materials overlying bedrock.
The Ground Motion Amplification Map (Plate 2) was
developed by combining information on the thickness and
distribution of geologic units with measurements of the
physical properties of the units. The shear modulus is
one of the most critical properties and was determined
directly by down-hole measurements of the shear-wave
velocity in twenty boreholes throughout the quadrangle.
51
The degree of ground shaking amplification that
might occur in future earthquakes was estimated using the
computer program SHAKE (Schnable and others 1972). Version
SHAXE88 was selected because it is widely used and
accepted for this type of modeling. In order to account
for both the variation of shaking from one earthquake to
another and to account for the three distinct types of
seismic sources in the region, Plate 2 was created by
taking the maximum amplification from the SHAKE analysis
of five time histories of bedrock acceleration.
Plate 2 uses a color-coded map to depict the ground
motion amplification of the Portland Quadrangle. The
Amplification is defined as the ratio of the peak ground
surface acceleration to the peak bedrock acceleration, as
shown in Table XIV, following.
The amount of amplification depends primarily on the
thickness and shear-wave velocity of the unconsolidated
materials overlying bedrock.
The Ground Motion Amplification Map (Plate 2) was
developed by combining information on the thickness and
distribution of geologic units with measurements of the
physical properties of the units. The shear modulus is
one of the most critical properties and was determined
directly by down-hole measurements of the shear-wave
velocity in twenty boreholes throughout the quadrangle.
51
The degree of ground shaking amplification that
might occur in future earthquakes was estimated using the
computer program SHAKE (Schnable and others 1972). Version
SHAXE88 was selected because it is widely used and
accepted for this type of modeling. In order to account
for both the variation of shaking from one earthquake to
another and to account for the three distinct types of
seismic sources in the region, Plate 2 was created by
taking the maximum amplification from the SHAKE analysis
of five time histories of bedrock acceleration.
Plate 2 uses a color-coded map to depict the ground
motion amplification of the Portland Quadrangle. The
Amplification is defined as the ratio of the peak ground
surface acceleration to the peak bedrock acceleration, as
shown in Table XIV, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
TABLE XIV
GROUND MOTION AMPLIFICATION FROM GMS-79 PLATE 2
MAP COLOR AMPLIFICATION
DarK Blue 1 or less
DarK Green 1 to 1.4
Light Green D 1.4 to 1.8
YellOW D 1.8 to 2. 2
Orange D 2. 2 to 2. 5
Red [] . . 2. 5 or more
Dynamic Slope Instability Map
The dynamic loading imposed by earthquaKes tends to
destabilize steep slopes that are stable under static
conditions. The objective o£ the Dynamic Slope
Instability Map (also on Plate 3) is to identify areas
liKely to experience dynamic instability. The only
significant area o£ steep slopes is in the West Hills,
which are covered by a varying thicKness o£ the loessial
Portland Hills Silt. The map was developed by analyzing
slope steepness, thicKness o£ unconsolidated material
overlying bedrocK, and potential seismic loading.
The steepness o£ slopes was derived £rom USGS
elevation data. The thicKness o£ unconsolidated material,
which is all loess, was interpreted £rom boreholes in the
52
TABLE XIV
GROUND MOTION AMPLIFICATION FROM GMS-79 PLATE 2
MAP COLOR AMPLIFICATION
DarK Blue 1 or less
DarK Green 1 to 1.4-
Light Green D 1.4- to 1.8
Yellow 0 1.8 to 2. 2
Orange 0 2. 2 to 2. 5
Red [] ~ ... / ... 2. 5 or more
Dynamic Slope Instability Map
The dynamiC loading imposed by earthquaKes tends to
destabilize steep slopes that are stable under static
conditions. The objective of the Dynamic Slope
Instability Map (also on Plate 3) is to identify areas
liKely to experience dynamic instability. The only
significant area of steep slopes is in the West Hills,
which are covered by a varying thicKness of the loessial
Portland Hills Silt. The map was developed by analyzing
slope steepness, thicKness of unconsolidated material
overlying bedrocK, and potential seismic loading.
The steepness of slopes was derived from USGS
elevation data. The thicKness of unconsolidated material,
which is all loess, was interpreted from boreholes in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
53
area and seismic refraction profiling done for this
purpose. The strength properties of the loess were based
on extensive laboratory testing done for the west-side
light-rail project. A soil friction angle of 33-degrees
and a moist unit weight of 115 pcf were used in the
stability analysis, based on Tri-Met (1993).
The dynamic slope stability was evaluated with a
pseudo-static infinite slope model, comparing the results
obtained from four seismic coefficients: 0.05, 0. 15, and
o. 25. Factors of safety were assigned to map areas derived
from the seismic coefficient of 0. 15, which was judged to
most realistically represent the shaking likely to be
experienced. Four hazard zones are identified by color-code
on the map, as shown in Table XV, following.
TABLE XV
RELATIVE DYNAMIC SLOPE INSTABILITY FROM GMS-79 PLATE 3
MAP COLOR RELATIVE DYNAMIC SLOPE INSTABILITY
White Hone
Dark Green • Slope : 15X or greater
Blue Ill ' Factor of Safety : 1. 25 to 2.0
Dark Purple • Factor of Safety less than 1. 25
Light Purple • Existing Landslide
53
area and seismic refraction profiling done for this
purpose. The strength properties of the loess were based
on extensive laboratory testing done for the west-side
light-rail project. A soil friction angle of 33-degrees
and a moist unit weight of 115 pcf were used in the
stability analysis, based on Tri-Met (1993).
The dynamic slope stability was evaluated with a
pseudo-static infinite slope model, comparing the results
obtained from four seismic coefficients: 0.05, 0.15, and
0.25. Factors of safety were assigned to map areas derived
from the seismic coefficient of O. 15, which was judged to
most realistically represent the shaking likely to be
experienced. Four hazard zones are identified by color-code
on the map, as shown in Table XV, following.
TABLE XV
RELATIVE DYNAMIC SLOPE INSTABILITY FROM GHS-79 PLATE 3
HAP COLOR RELATIVE DYNAMIC SLOPE INSTABILITY
White None
Dark Green • Slope : 15X or greater
Blue III i~ Factor of Safety : 1. 25 to 2.0
Dark Purple • Factor of Safety less than 1. 25
Light Purple • Existing Landslide
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
The darK green merely represents areas with steep
slopes, which could experience some instability just
because of their steepness. Blue represents areas that
should be suspect during earthquaKes. DarK purple
represents areas with the greatest hazard based on the
stability modeling, with factors of safety less than 1. 25;
these areas are liKely to experience instability during
earthquaKes with a rocK PGA 0.20g, or less. The light
purple represents existing landslides, which are liKely to
be reactivated in an earthquaKe.
Plate 3 combines two seismic hazards on one map:
lateral spread displacement and dynamic slope instability.
These two seismic hazards involve ground slopes that are
different by an order of magnitude, and always occur in
mutually exclusive map areas. Therefore, the two hazards
have been logically combined on one map, to reduce the
total number of maps required.
Relative EarthquaKe Hazard Map
This is a composite map developed to show which
areas have the greatest tendency to experience damage due
to any one, or any combination of three different
earthquaKe-related hazards. The map was developed by
combining the single hazard maps for ground motion
amplification, liquefaction, and slope instability. The
objective is to generalize the hazards in a way useful to
54
The darK green merely represents areas with steep
slopes, which could experience some instability just
because of their steepness. Blue represents areas that
should be suspect during earthquakes. DarK purple
represents areas with the greatest hazard based on the
stability modeling, with factors of safety less than 1.25;
these areas are liKely to experience instability during
earthquakes with a rocK PGA O.20g, or less. The light
purple represents existing landslides, which are liKely to
be reactivated in an earthquake.
Plate 3 combines two seismic hazards on one map:
lateral spread displacement and dynamic slope instability.
These two seismic hazards involve ground slopes that are
different by an order of magnitude, and always occur in
mutually exclusive map areas. Therefore, the two hazards
have been logically combined on one map, to reduce the
total number of maps required.
Relative Earthquake Hazard Map
This is a composite map developed to show which
areas have the greatest tendency to experience damage due
to anyone, or any combination of three different
earthquake-related hazards. The map was developed by
combining the single hazard maps for ground motion
amplification, liquefaction, and slope instability. The
objective is to generalize the hazards in a way useful to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
a non-technical audience such as planners, lenders, and
emergency responders, and for hazard mitigation response
planning.
The three individual hazards were categorized as
zones 0, 1,2, or 3, with 3 being the greatest hazard, as
follows:
For ground motion amplification:
1) Amplification Factor less than 1. 25
2) Amplification factor 1. 25 to 1. 50
3) Amplification factor greater than 1. 50
For liquefaction susceptibility:
1) Materials liquefiable when they are
intermittently saturated.
55
2) Liquefiable material less than 20-ft thic~,
with a water table 15 to 30-ft deep.
3} Liquefiable material more than 30-ft thic~,
with a water table 15 to 30-ft deep, or when
the water table is less than 15-feet deep.
For dynamic slope instability:
1} Slopes steeper than 15-percent.
2) Dynamic safety factor from 2.0 to 1. 25
3) Dynamic safety factor less than 1. 25, or the
vicinity of an existing landslide.
The relative hazard map methodology taKes a "vector
sum, " or square root of the sum of the squares, of the
a non-technical audience such as planners, lenders, and
emergency responders, and for hazard mitigation response
planning.
The three individual hazards were categorized as
zones 0,1,2, or 3, with 3 being the greatest hazard, as
follows:
For ground motion amplification:
i) Amplification Factor less than i. 25
2) Amplification factor i. 25 to i. 50
3) Amplification factor greater than i. 50
For liquefaction susceptibility:
1) Materials liquefiable when they are
intermittently saturated.
55
2) Liquefiable material less than 20-ft thick,
with a water table i5 to 30-ft deep.
3) Liquefiable material more than 30-ft thick,
with a water table 15 to 30-ft deep, or when
the water table is less than i5-feet deep.
For dynamic slope instability:
i) Slopes steeper than is-percent.
2) Dynamic safety factor from 2.0 to 1.25
3) Dynamic safety factor less than 1.25, or the
vicinity of an existing landslide.
The relative hazard map methodology takes a "vector
sum," or square root of the sum of the squares, of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
three individual hazard categories, rounding the result to
the nearest whole number. The result is assigned to
hazard categories, as follows:
VECTOR SUM RELATIVE HAZARD MAP ZONE
4 Zone A - Greatest Hazard
3 Zone B
2 Zone C
1 Zone D - Least Hazard
The result of this system is that areas with a high
hazard from a single local effect are assigned the rating
of Zone B, as well as areas with a combination of lesser
single ratings.
of high ratings.
The rating of A represents a combination
The hazard Zone B should not be
underrated, since it can result from the great severity of
a single hazard.
Significance and Limitations
The quantitative data provided by these maps is the
best currently available information on probabilistic
seismic soil hazards in Portland.
The most severe damage done by an earthquaKe is
sometimes concentrated in areas which respond adversely
due to the subsurface soil conditions. Clearly,
improvements in regionally defining and quantifying the
soil strata response to earthquake shaKing is important in
modeling the total damage to structures.
56
three individual hazard categories, rounding the result to
the nearest whole number. The result is assigned to
hazard categories, as follows:
VECTOR SUM RELATIVE HAZARD MAP ZONE
4 Zone A - Greatest Hazard
3 Zone B
2 Zone C
1 Zone D - Least Hazard
The result of this system is that areas with a high
hazard from a single local effect are assigned the rating
of Zone B, as well as areas with a combination of lesser
single ratings.
of high ratings.
The rating of A represents a combination
The hazard Zone B should not be
underrated, since it can result from the great severity of
a single hazard.
Significance and Limitations
The quantitative data provided by these maps is the
best currently available information on probabilistic
seismic soil hazards in Portland.
The most severe damage done by an earthquake is
sometimes concentrated in areas which respond adversely
due to the subsurface soil conditions. Clearly,
improvements in regionally defining and quantifying the
soil strata response to earthquake shaking is important in
modeling the total damage to structures.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The earthquaKe hazard maps are intended to provide
an estimate of regional variations in the geologic
earthquaKe hazards, but were not intended for use as the
basis for earthquaKe resistant design of individual
structures. Additional site-specific investigations and
analyses are generally required for such purposes.
57
The earthquaKe hazard maps are intended to provide
an estimate of regional variations in the geologic
earthquaKe hazards, but were not intended for use as the
basis for earthquaKe resistant design of individual
structures. Additional site-specific investigations and
analyses are generally required for such purposes.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FEMA-227: A BENEFIT-COST MODEL FOR SEISMIC REHABILITATION OF BUILDINGS
FEMA-227 (1992) presents the results of a two year
experimental research project undertaken to establish a
benefit-cost model for seismic rehabilitation of
buildings. The objective of the project was to encourage
local decision makers, the design professions, and other
58
interested groups to undertake a program of mitigating the
risks posed by future earthquakes.
The intended application of the model is to provide
quick and preliminary estimates of whether a prospective
seismic retrofit program is economically justifiable. The
benefit-cost model presented produces three indices for
judging whether the rehabilitation of a building or group
of buildings is economically justified:
1. The expected net present value of the retrofit
without the value of life.
2. The expected net present value of the retrofit
with the value of life.
3. The benefit/cost ratio of the retrofit.
When expected benefits exceed costs, the net present
value is positive and the rehabilitation investment is
economically justified. When expected benefits are less
than costs, the net present value is negative and the
rehabilitation investment is not economically justified.
FEMA-227: A BENEFIT-COST MODEL FOR SEISMIC REHABILITATION OF BUILDINGS
FEHA-227 (1992) presents the results of a two year
experimental research project undertaKen to establish a
benefit-cost model for seismic rehabilitation of
buildings. The objective of the project was to encourage
local decision maKers, the design professions, and other
58
interested groups to undertaKe a program of mitigating the
risks posed by future earthquaKes.
The intended application of the model is to provide
quick and preliminary estimates of whether a prospective
seismic retrofit program is economically justifiable. The
benefit-cost model presented produces three indices for
judging whether the rehabilitation of a building or group
of buildings is economically justified:
1. The expected net present value of the retrofit
without the value of life.
2. The expected net present value of the retrofit
with the value of life.
3. The benefit/cost ratio of the retrofit.
When expected benefits exceed costs, the net present
value is positive and the rehabilitation investment is
economically justified. When expected benefits are less
than costs, the net present value is negative and the
rehabilitation investment is not economically justified.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
59
Expected Net Present Value Model without the Value of Life
The expected net present value (HPV) of a seismic
rehabilitation investment is the sum of the present value
of benefits expected to accrue each year
(probabilistically) over the planning period, plus the
present value of the salvage value of the rehabilitation
investment at the end of the planning period, minus the
initial cost of the rehabilitation. FEMA-227 presents the
following equation for NPV:
NPV = Bt [
where:
1-(i+i)-t i ] + Vt
(""i+T)t - INV
Bt = the expected annual benefit attributed to the retrofit in year t;
Vt = any change that the retrofit will have on the salvage
value of the building in the terminal year t;
t = the length of the planning horizon which should
reflect the effective life of the retrofit of the
buildings;
i = the discount rate; and
IHV = the cost (engineering/construction) of the retrofit.
In the NPV equation above, the term in brackets is
the Uniform Series -Present Worth Factor (PWF), which
discounts each years benefit over the planning horizon to
its present value. The choice of an appropriate discount
rate, i, is one of the most difficult aspects of
59
Expected Net Present Value Model without the Value of Life
The expected net present value (HPV) of a seismic
rehabilitation investment is the sum of the present value
of benefits expected to accrue each year
(probabilistically) over the planning period, plus the
present value of the salvage value of the rehabilitation
investment at the end of the planning period, minus the
initial cost of the rehabilitation. FEMA-227 presents the
following equation for HPV:
HPV : Bt [
where:
1-(i+i)-t i
Vt --rr+TIt - INV
Bt : the expected annual benefit attributed to the retrofit in year t;
Vt : any change that the retrofit will have on the salvage
value of the building in the terminal year t;
t : the length of the planning horizon which should
reflect the effective life of the retrofit of the
buildings;
i = the discount rate; and
IHV = the cost (engineering/construction) of the retrofit.
In the NPV equation above, the term in brackets is
the Uniform Series - Present Worth Factor (PWF) , which
discounts each years benefit over the planning horizon to
its present value. The choice of an appropriate discount
rate, i, is one of the most difficult aspects of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
benefit-cost analysis. Increasing the discount rate
lowers the present value of future benefits and lowers
Benefit/Cost (B/C) ratios. FEMA-227 recommends using a
discount rate between 3/. and 6/.i for public sector
considerations, the lower end of the range is appropriate,
whereas for private sector, the higher end is more
appropriate. The planning horizon also significantly
impacts benefit-cost results. Longer planning horizons
capture more future benefits, resulting in higher present
value of benefits. Typical planning horizons suggested by
FEMA-227 are 20 or 30 years.
Cost records of renovated buildings rarely indicate
what would have been the cost of seismic retrofit alone,
separate from other Kinds of renovation that generally
occurs concurrently. In estimating the cost of seismic
retrofit, a decision must be made whether to use the
estimated cost of seismic retrofit as if it were the only
building improvement, or a share of costs that occur in
conjunction with other improvements. The retrofit cost
question is covered in more detail under the discussion of
FEMA-156, to follow.
Typically, the net present value of the salvage
value of the retrofit investment is only a few percent of
retrofit costs and, thus, has only a minor impact on B/C
ratios.
60
benefit-cost analysis. Increasing the discount rate
lowers the present value of future benefits and lowers
Benefit/Cost (B/C) ratios. FEHA-227 recommends using a
discount rate between 31. and 61.j for public sector
considerations, the lower end of the range is appropriate,
whereas for private sector, the higher end is more
appropriate. The planning horizon also significantly
impacts benefit-cost results. Longer planning horizons
capture more future benefits, resulting in higher present
value of benefits. Typical planning horizons suggested by
FEMA-227 are 20 or 30 years.
Cost records of renovated buildings rarely indicate
what would have been the cost of seismic retrofit alone,
separate from other Kinds of renovation that generally
occurs concurrently. In estimating the cost of seismic
retrofit, a decision must be made whether to use the
estimated cost of seismic retrofit as if it were the only
building improvement, or a share of costs that occur in
conjunction with other improvements. The retrofit cost
question is covered in more detail under the discussion of
FEMA-156, to follow.
Typically, the net present value of the salvage
value of the retrofit investment is only a few percent of
retrofit costs and, thus, has only a minor impact on B/C
ratios.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
The expected annual benefit which accrues from the
retrofit (Bt) is the sum of expected avoided losses, both
direct and indirect, accounting for the expected annual
probability of damaging earthquakes, as follows:
Bt : E [ (EAEm) (BJ)IIl + ID) ]
where:
EAEm =
BJ)IIl =
ID =
the expected number of earthquakes annually by
MMI, ranging from VI to XII;
building damages avoided by facility class
MMI; and
indirect losses avoided, such as rental,
relocation, income, business inventory, and
personal property.
and
In this equation, the summation indicates that
expected annual benefits are summed over the expected
earthquakes with MMis ranging from VI to XII. Avoided
damages and losses means the reduction in expected damages
and losses in retrofit buildings versus those expected in
unrehabilitated buildings of the same facility class.
Building damages avoided (BJ)IIl) are the product
of the replacement value of the building times the
expected mean damage factor for the facility class and
MMI, times the expected rehabilitation effectiveness in
reducing building damage, as follows:
61
The expected annual benefit which accrues from the
retrofit (Bt) is the sum of expected avoided losses, both
direct and indirect, accounting for the expected annual
probability of damaging earthquaKes, as follows:
Bt = E [ (EAEm) (BJ)IIl + ID) ]
where:
EAEm =
BJ)IIl =
ID =
the expected number of earthquaKes annually by
MMI, ranging from VI to XIIj
building damages avoided by facility class
MMlj and
indirect losses avoided, such as rental,
relocation, income, business inventory, and
personal property.
and
In this equation, the summation indicates that
expected annual benefits are summed over the expected
earthquaKes with MMls ranging from VI to XII. Avoided
damages and losses means the reduction in expected damages
and losses in retrofit buildings versus those expected in
unrehabilitated buildings of the same facility class.
Building damages avoided (BJ)IIl) are the product
of the replacement value of the building times the
expected mean damage factor for the facility class and
MMI, times the expected rehabilitation effectiveness in
reducing building damage, as follows:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
where:
BVAL = the replacement value of the building;
MDFm = the mean damage factor by facility class
and MMI, per ATC-13; and
EREm = expected retrofit effectiveness by
facility class & MMI.
62
To account roughly for the greater damage expected
at poor soil sites, the model adjusts the mean damage
factor upward by one MMI level for sites on "soft" soils.
The retrofit effectiveness, EREm, is defined as
the percent reduction in expected damages in the
strengthened facility compared to the expected damages in
the unstrengthened facility. This factor was determined
as part of FEMA-227. These estimates were based on
engineering experience and judgement, assuming life-safety
as the objective of the retrofit.
"Life-Safety" retrofits provide the minimum
strengthening to prevent building collapse. Other
objectives are "Damage Control," for additional reduction
in damage, and "Immediate Occupancy," appropriate for
emergency service facilities. FEMA-227 invites users to
adjust the retrofit effectiveness estimates for alternate
performance objectives.
where:
BVAL = the replacement value of the buildingj
HDFm = the mean damage factor by facility class
and MHI, per ATC-13j and
EREm = expected retrofit effectiveness by
facility class & HMI.
62
To account roughly for the greater damage expected
at poor soil sites, the model adjusts the mean damage
factor upward by one HMI level for sites on "soft" soils.
The retrofit effectiveness, EREm, is defined as
the percent reduction in expected damages in the
strengthened facility compared to the expected damages in
the unstrengthened facility. This factor was determined
as part of FEHA-227. These estimates were based on
engineering experience and judgement, assuming life-safety
as the objective of the retrofit.
"Life-Safety" retrofits provide the minimum
strengthening to prevent building collapse. Other
objectives are "Damage Control," for additional reduction
in damage, and "Immediate Occupancy," appropriate for
emergency service facilities. FEHA-227 invites users to
adjust the retrofit effectiveness estimates for alternate
performance objectives.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
Expected Net Present Value Model with the Value of Life
The model discussed above does not include the value
of life. When the value of life is included, the value of
avoided deaths is frequently one of the principal factors
in producing high B/C ratios for prospective strengthening
programs, especially for high occupancy facilities.
The expected net present value including the value
of life (NPvvol) is the expected net present value
without the value of life, plus the present value of
expected deaths avoided by seismic retrofit, as follows:
NPvvol : NPV + [ (VDAt) (PWF) J where:
NPV = the expected net present value without the value
of life;
VDAt = the annual value of expected deaths avoided by
retrofitting buildings to life-safety
standards.
PWF = the Uniform Series - Present Worth Factor, as
previously defined.
The annual value of avoided earthquake death loss is
assumed to be the product of the building occupancy times
the difference in expected death rates between unretrofit
and retrofit buildings, times the dollar value of one
human life, as follows:
63
Expected Net Present Value Model with the Value of Life
The model discussed above does not include the value
of life. When the value of life is included, the value of
avoided deaths is frequently one of the principal factors
in producing high B/C ratios for prospective strengthening
programs, especially for high occupancy facilities.
The expected net present value including the value
of life (NPVvol) is the expected net present value
without the value of life, plus the present value of
expected deaths avoided by seismic retrofit, as follows:
NPVvol = NPV + [ (VDAt) (PWF) ]
where:
NPV = the expected net present value without the value
of lifej
VDAt = the annual value of expected deaths avoided by
retrofitting buildings to life-safety
standards.
PWF = the Uniform Series - Present Worth Factor, as
previously defined.
The annual value of avoided earthquake death loss is
assumed to be the product of the building occupancy times
the difference in expected death rates between unretrofit
and retrofit buildings, times the dollar value of one
human life, as follows:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VDAt : E [ (EAEID) (OCC) (DRm - DRRffi) ] (VOL)
where:
OCC = the average occupancy of the building, number of
people;
DRm = the expected death rate by central damage
factor, according to ATC-13 Table 3-9
DRRm = is the expected death rate for rehabilitated
buildings by central damage factor; and
VOL = the dollar value of one human life.
The summation symbol indicates that the expected
deaths avoided must be summed over the range of MMI
earthquakes for each facility. The occupancy OCC is
determined from local occupancy data, or by applying
ATC-13 Table 4. 12 to the building area.
64
The expected death rate for the rehabilitated
buildings, DRRm, are estimated based on engineering
experience and judgement, assuming that the death rates
are lowered to those that would be expected if the
building damage states were three states lower. For most
damage states, this assumption is equivalent to reducing
deaths by a factor of 1,000. This benefit-cost model did
not include as a benefit the economic value of injuries
avoided by retrofitting, concluding that this benefit was
insignificant compared with the dollar value of life.
VDAt = E [ (EAEm) (CCC) (DRm - DRRm) ] (VOL)
where:
OCC = the average occupancy of the building, number of
people;
DRm = the expected death rate by central damage
factor, according to ATC-13 Table 3-9
DRRm = is the expected death rate for rehabilitated
buildings by central damage factor; and
VOL = the dollar value of one human life.
The summation symbol indicates that the expected
deaths avoided must be summed over the range of HHI
earthquakes for each facility. The occupancy OCC is
determined from local occupancy data, or by applying
ATC-13 Table 4. 12 to the building area.
64
The expected death rate for the rehabilitated
buildings, DRRm, are estimated based on engineering
experience and judgement, assuming that the death rates
are lowered to those that would be expected if the
building damage states were three states lower. For most
damage states, this assumption is equivalent to reducing
deaths by a factor of 1,000. This benefit-cost model did
not include as a benefit the economic value of injuries
avoided by retrofitting, concluding that this benefit was
insignificant compared with the dollar value of life.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The value of one human life, VOL, has been the
subject of much study, with values ranging from
65
$ 1,100,000 to$ 8,000,000 per life. FEMA-227 recommends a
consensus value obtained from a review of 25 updated
studies for the FAA, of$ 1,740,000.
Benefit/Cost Ratios
Benefit/cost (B/C) ratios are an alternative way of
viewing net present value results, which may make it
easier to compare and prioritize prospective
rehabilitation projects. B/C ratios are the expected
present value of future benefits divided by retrofit cost.
B/C ratios greater than one correspond to positive
expected net present values, indicating that a prospective
retrofit is economically justified, while ratios less than
one correspond to negative expected net present values and
a prospective retrofit that is not justified on the basis
of the assumptions in the model. The extent to which B/C
ratios are greater or less than one provides important
guidance as to the relative economic justification of
prospective retrofit projects.
Significance and Limitations
Benefit-cost analysis is a widely-used economic tool
for helping to maKe economic decisions. The central
economic question about retrofitting earthquake-hazardous
The value of one human life, VOL, has been the
subject of much study, with values ranging from
65
$ 1,100,000 to $ 8,000,000 per life. FEMA-227 recommends a
consensus value obtained from a review of 25 updated
studies for the FAA, of $ 1,740,000.
Benefit/Cost Ratios
Benefit/cost (B/C) ratios are an alternative way of
viewing net present value results, which may make it
easier to compare and prioritize prospective
rehabilitation projects. B/C ratios are the expected
present value of future benefits divided by retrofit cost.
B/C ratios greater than one correspond to positive
expected net present values, indicating that a prospective
retrofit is economically justified, while ratios less than
one correspond to negative expected net present values and
a prospective retrofit that is not justified on the basis
of the assumptions in the model. The extent to which B/C
ratios are greater or less than one provides important
guidance as to the relative economic justification of
prospective retrofit projects.
Significance and Limitations
Benefit-cost analysis is a widely-used economic tool
for helping to make economic decisions. The central
economic question about retrofitting earthquake-hazardous
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
buildings is whether the benefits which accrue from
retrofit are sufficiently valuable to warrant the expense.
FEMA-227 was an important step in establishing a model to
answer this question.
Since the benefit-cost analysis is based on the
ATC-13 loss model, the limitations of ATC-13 are carried
into this model, namely, that loss estimates are based on
''average" buildings of each class, with the
characteristics of individual buildings not accounted for;
additionally, important soil profile effects, often an
important component of earthquake losses, are only roughly
accounted for.
Two other areas add uncertainty to the model.
The first is uncertainty in the evaluation of retrofit
effectiveness. Estimating the effectiveness of retrofit
would be difficult even if clear and well-defined seismic
strengthening standards were available. Currently, some
of the facility classes considered do not have well
defined standards, and very few buildings that have
received seismic improvements have ever been tested by
earthquakes strong enough to demonstrate their
effectiveness.
A second fundamental limitation with the model is
imposed by the expected value methodology used. For
Portland, the return period for damaging earthquakes (the
66
buildings is whether the benefits which accrue from
retrofit are sufficiently valuable to warrant the expense.
FEMA-227 was an important step in establishing a model to
answer this question.
Since the benefit-cost analysis is based on the
ATC-13 loss model, the limitations of ATC-13 are carried
into this model, namely, that loss estimates are based on
"average" buildings of each class, with the
characteristics of individual buildings not accounted forj
additionally, important soil profile effects, often an
important component of earthquake losses, are only roughly
accounted for.
Two other areas add uncertainty to the model.
The first is uncertainty in the evaluation of retrofit
effectiveness. Estimating the effectiveness of retrofit
would be difficult even if clear and well-defined seismic
strengthening standards were available. Currently, some
of the facility classes considered do not have well
defined standards, and very few buildings that have
received seismic improvements have ever been tested by
earthquakes strong enough to demonstrate their
effectiveness.
A second fundamental limitation with the model is
imposed by the expected value methodology used. For
Portland, the return period for damaging earthquakes (the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
inverse of the annual exceedance probability) is far
greater than any reasonable planning horizon. This
results in total losses, and hence total benefits from
retrofit, being significantly less than the losses and
retrofit benefits from a moderate earthquaKe. At constant
retrofit cost, the resulting net present value and B/C
ratios could produce misleading conclusions regarding the
cost effectiveness of retrofit.
67
inverse of the annual exceedance probability) is far
greater than any reasonable planning horizon. This
results in total losses, and hence total benefits from
retrofit, being significantly less than the losses and
retrofit benefits from a moderate earthquaKe. At constant
retrofit cost, the resulting net present value and B/C
ratios could produce misleading conclusions regarding the
cost effectiveness of retrofit.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FEMA-156: TYPICAL COSTS FOR SEISMIC REHABILITATION OF EXISTING BUILDINGS, SECOND EDITION
FEMA-156 (1994) presents a methodology to estimate
typical costs for seismic retrofit of buildings. The
methodology was derived from a computerized database of
2,088 retrofit costs.
Each retrofit cost data point is from either an
68
actual retrofit project or the estimated cost of retrofit
of a building subjected to a detailed analysis by an
experienced design professional. Cost estimates based on
studies were excluded, as were cost estimates that lacked
certain critical information. Consistency was achieved by
weighting the quality of the data and the source.
Therefore, the resulting database is extensive, objective,
reliable, and consistent.
FEMA-156 provides three methods of retrofit cost
estimation. Option 1 requires a minimum of input
information, and is intended for rough preliminary
estimates. Option 3 uses raw cost data from the
computerized database, and is the most accurate for
costing individual buildings, but contains gaps that make
it less suitable for use in regional studies. Option 2 is
the most suitable for the present research, because it
provides values in tables that are derived using a
smoothing of the cost data to provide values for all
FEMA-156: TYPICAL COSTS FOR SEISMIC REHABILITATION OF EXISTING BUILDINGS, SECOND EDITION
FEHA-156 (1994) presents a methodology to estimate
typical costs for seismic retrofit of buildings. The
methodology was derived from a computerized database of
2,088 retrofit costs.
Each retrofit cost data point is from either an
68
actual retrofit project or the estimated cost of retrofit
of a building subjected to a detailed analysis by an
experienced design professional. Cost estimates based on
studies were excluded, as were cost estimates that lacked
certain critical information. Consistency was achieved by
weighting the quality of the data and the source.
Therefore, the resulting database is extensive, objective,
reliable, and consistent.
FEMA-156 provides three methods of retrofit cost
estimation. Option 1 requires a minimum of input
information, and is intended for rough preliminary
estimates. Option 3 uses raw cost data from the
computerized database, and is the most accurate for
costing individual buildings, but contains gaps that make
it less suitable for use in regional studies. Option 2 is
the most suitable for the present research, because it
provides values in tables that are derived using a
smoothing of the cost data to provide values for all
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
variable combinations, and to provide logical
relationships between changes in variables and changes in
costs. Option 2 requires the following input:
Building Group
Building Area (size)
State
Year o£ Retrofit Construction
HEHRP Seismic Map Area
Performance Objective
The typical structural cost per square foot of
retrofit (C), using option 2, is calculated according to
the following formula:
C = (C1) (C2) (C3) (CL) (CT)
where:
C1 = Building Group Mean Cost
C2 = Area Adjustment Factor
C3 = Seismicity/Performance Objective Adjustment
Factor
CL = Location Adjustment Factor
CT = Time Adjustment Factor
The methodology starts with the identification o£
the building type, which is assigned to one of eight
Building Groups, according to similarities in retrofit
mean costs, C1, as shown in Table XVI, following.
69
variable combinations, and to provide logical
relationships between changes in variables and changes in
costs. Option 2 requires the following input:
Building Group
Building Area (size)
State
Year of Retrofit Construction
HEHRP Seismic Map Area
Performance Objective
The typical structural cost per square foot of
retrofit (C), using option 2, is calculated according to
the following formula:
C = (Cl) (C2) (C3) (CL) (CT)
where:
Cl = Building Group Mean Cost
C2 = Area Adjustment Factor
C3 = Seismicity/Performance Objective Adjustment
Factor
CL = Location Adjustment Factor
CT = Time Adjustment Factor
The methodology starts with the identification of
the building type, which is assigned to one of eight
Building Groups, according to similarities in retrofit
mean costs, Ci' as shown in Table XVI, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
TABLE XVI
FEMA-156 BUILDING GROUP MEAN COST
BUILDING ATC-21 GROUP MEAN COST GROUP BUILDING TYPE ($ I sq. ft. )
1 URM 15.29
2 W1, W2 12.29
3 PC1, RM1 14.02
4 C1,C3 20.02
5 S1 18.86
6 S2,S3 7.23
7 S5 24.01
8 C2,PC2,RM2,S4 17. 31
The size (area) of a building affects its typical
cost per square foot. The buildings are assigned to an
area group, as follows:
Small Less than 10,000 sq. ft.
Medium 10,000 to 49,999 sq. ft.
Large 50,000 to 99,999 sq. ft.
Very Large 100,000 sq. ft. or more
The Area Adjustment Factor, C2, is a function of
the area group and the building group, as shown in Table
XVII, following.
70
TABLE XVI
FEMA-156 BUILDING GROUP HEAN COST
BUILDING ATC-21 GROUP MEAN COST GROUP BUILDING TYPE ($ / sq. ft. )
1 URM 15.29
2 Wi, W2 12.29
3 PC1,RM1 14.02
4 Ci, C3 20.02
5 S1 18.86
6 S2,S3 7.23
7 S5 24.01
8 C2,PC2,RM2,S4 17. 31
The size (area) of a building affects its typical
cost per square foot. The buildings are assigned to an
area group, as follows:
Small Less than 10,000 sq. ft.
Medium 10,000 to 49.999 sq. ft.
Large 50,000 to 99,999 sq. ft.
Very Large 100,000 sq. ft. or more
The Area Adjustment Factor, C2' is a function of
the area group and the building group, as shown in Table
XVII, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XVII
FEMA-156 AREA ADJUSTMENT FACTORS
BUILDING GROUP
1 2 3 5 6 7 8
Small 1. 01 0.97 1. 13 1. 09 1. 16 1. 18 1. 04 1.11
Medium 1. 00 1. 02 1. 07 1. 06 1. 14 1. 12 1. 03 1. 08
Large 0.95 1. 28 0.92 1. 01 1. 09 o. 90 o. 99 1. 02
v. Lrg 0.80 1. 64 o. 57 0. 84 0.83 o. 51 0.87 0.83
The Seismicity/Performance Objective Adjustment
Factor, C3, incorporates the the seismicity and the
desired performance objective, as shown in Table XVIII.
TABLE XVIII
FEMA-156 SEISMICITY/PERFORMANCE OBJECTIVE ADJUSTMENT FACTOR
71
SEISMICITY LIFE DAMAGE IMMEDIATE SAFETY CONTROL OCCUPANCY
Low 0.61 0.71 1. 21
Moderate 0.70 0.85 1. 40
High 0.89 1. 09 1. 69
Very High 1. 18 1. 43 2.08
TABLE XVII
FEHA-156 AREA ADJUSTMENT FACTORS
BUILDING GROUP
1 2 3 5 6 7 8
Small 1. 01 0.97 1. 13 1. 09 1. 16 1. 18 1. 04 L 11
Medium L 00 1. 02 1. 07 1. 06 1. 14 1. 12 1. 03 1. 08
Large O. 95 1. 28 0.92 1. 01 1. 09 O. 90 O. 99 1. 02
V. Lrg 0.80 1. 64 O. 57 O. 84 O. 83 0.51 0.87 O. 83
The Seismicity/Performance Objective Adjustment
Factor, C3' incorporates the the seismicity and the
desired performance objective, as shown in Table XVIII.
TABLE XVIII
FEMA-156 SEISMICITY/PERFORMANCE OBJECTIVE ADJUSTMENT FACTOR
71
SEISMICITY LIFE DAMAGE IMMEDIATE SAFETY CONTROL OCCUPANCY
Low 0.61 0.71 1. 21
Moderate 0.70 0.85 1. 40
High 0.89 L 09 1. 69
Very High 1. 18 1. 43 2.08
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
The seismicity is identified by the NEHRP Seismicity
Map Areas: map areas 1 & 2 are Low, 3 & 4 are Moderate,
5 & 6 are High, and 7 is Very High. Life Safety allows for
unrepairable damage as long as life is not jeopardized and
egress routes are not blocKed. Damage Control protects
some feature or function of the building beyond
life-safety, such as protecting building contents or
preventing the release of toxic material. Immediate
Occupancy allows only minimal post-earthquaKe damage and
disruption, with some nonstructural repairs and cleanup
done while the building remains occupied and safe.
The location adjustment factor, CL, compares the
purchasing power in each state as compared to Missouri;
Missouri was chosen as the base because of its central
geographic location. A factor for all 50 states is given;
of interest in the proposed research is Oregon, and
perhaps several western states, as shown in Table XIX.
TABLE XIX
FEMA-156 LOCATION ADJUSTMENT FACTOR
Missouri Oregon California Washington Idaho Utah Montana Nevada
1. 00 o. 99 1. 12 1. 02 o. 91 o. 89 o. 90 1. 03
72
The seismicity is identified by the NEHRP Seismicity
Map Areas: map areas 1 & 2 are Low, 3 & 4 are Moderate,
5 & 6 are High, and 7 is Very High. Life Safety allows for
unrepairable damage as long as life is not jeopardized and
egress routes are not blocKed. Damage Control protects
some feature or function of the building beyond
life-safety, such as protecting building contents or
preventing the release of toxic material. Immediate
Occupancy allows only minimal post-earthquake damage and
disruption, with some nonstructural repairs and cleanup
done while the building remains occupied and safe.
The location adjustment factor, CL' compares the
purchasing power in each state as compared to Missouri;
Missouri was chosen as the base because of its central
geographic location. A factor for all 50 states is given;
of interest in the proposed research is Oregon, and
perhaps several western states, as shown in Table XIX.
TABLE XIX
FEMA-156 LOCATION ADJUSTMENT FACTOR
Missouri Oregon California Washington Idaho Utah Montana Nevada
1. 00 O. 99 1. 12 1. 02 O. 91 O. 89 O. 90 1. 03
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Time Adjustment Factor, CT, projects costs
beyond the 1993 cost database assuming various rates of
inflation. A selection of years vs. inflation rates is
shown in Table XX.
YEAR
1996
2000
2004
TABLE XX
FEMA-156 TIME ADJUSTMENT FACTOR
01.
1. 00
1. 00
1. 00
2/.
1. 06
1. 15
1. 24
4/.
1. 12
1. 32
1. 54
61.
1. 19
1. 50
1. 90
8/.
1. 26
1. 71
2.33
73
Volume 2 of FEMA-156 presents unit costs related to
seismic retrofit, including restoration of architectural
finishes, non-structural retrofit, design fees, permits,
management, insurance, and occupant relocation.
Significance and Limitations
The FEMA-156 methodology for estimating typical
seismic retrofit costs is the most authoritaive study
currently available, based on ATC-21 building types. The
scientifically and carefully compiled cost database upon
which it is based is the bacKbone of the study. This is
the best source of retrofit cost information for use in
the benefit-cost analysis portion of the present research.
The Time Adjustment Factor, CT' projects costs
beyond the 1993 cost database assuming various rates of
inflation. A selection of years vs. inflation rates is
shown in Table XX.
YEAR
1996
2000
2004
TABLE XX
FEMA-156 TIME ADJUSTMENT FACTOR
01.
1. 00
1. 00
1. 00
21.
1. 06
1. 15
1. 24
1. 12
1. 32
1. 54
61.
1. 19
1. 50
1. 90
81.
1. 26
1. 71
2.33
73
Volume 2 of FEMA-156 presents unit costs related to
seismic retrofit, including restoration of architectural
finishes, non-structural retrofit, design fees, permits,
management, insurance, and occupant relocation.
Significance and Limitations
The FEMA-156 methodology for estimating typical
seismic retrofit costs is the most authoritaive study
currently available, based on ATC-21 building types. The
scientifically and carefully compiled cost database upon
which it is based is the backbone of the study. This is
the best source of retrofit cost information for use in
the benefit-cost analysis portion of the present research.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER III
FOCUS OF RESEARCH
LIMITATIONS OF CURRENT PRACTICE
current methods of earthquaKe loss analysis (Goettel
and Horner 1995) are based only on "average" buildings of
a given structural type, ignoring individual variations in
design and construction, while incompletely modeling the
effect on damage of site-specific soil characteristics.
Decision analysis for retrofit is currently based on a
probabilistic expected value approach (Goettel and Horner
1995; FEMA-227 1992), which computes losses far less than
the losses that would be seen in an actual deterministic
scenario earthquake event, leading to uncertainty about
whether the most appropriate decision rule is being used.
Current cost-benefit methods for retrofit decision
analysis looK at buildings one at a time (FEMA-227 1992),
and do not prioritize an entire system of buildings, to
insure that the greatest savings in life and property are
obtained per dollar spent. Furthermore, the combinations
of attributes contributing to seismic performance are so
numerous that a worKable system to classify the buildings
for retrofit has yet to emerge.
CHAPTER III
FOCUS OF RESEARCH
LIMITATIONS OF CURRENT PRACTICE
Current methods of earthquaKe loss analysis (Goettel
and Horner 1995) are based only on "average" buildings of
a given structural type, ignoring individual variations in
design and construction, while incompletely modeling the
effect on damage of site-specific soil characteristics.
Decision analysis for retrofit is currently based on a
probabilistic expected value approach (Goettel and Horner
1995; FEMA-227 1992), which computes losses far less than
the losses that would be seen in an actual deterministic
scenario earthquaKe event, leading to uncertainty about
whether the most appropriate decision rule is being used.
Current cost-benefit methods for retrofit decision
analysis looK at buildings one at a time (FEHA-227 1992),
and do not prioritize an entire system of buildings, to
insure that the greatest savings in life and property are
obtained per dollar spent. Furthermore, the combinations
of attributes contributing to seismic performance are so
numerous that a worKable system to classify the buildings
for retrofit has yet to emerge.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
There does not currently exist a comprehensive
systems approach to finding the optimal way to proceed with
seismic rehabilitation of the buildings in a large city.
The preceding review of the current methodologies point to
shortcomings in two fundamental areas: 1) earthquake loss
modeling, and 2) retrofit decision making.
Earthquake Loss Modeling
Damage Estimation. Current methodologies of
earthquake loss modeling do not utilize data on the actual
buildings being analyzed. FEMA-227, for example, follows
the lead of ATC-13, basing its loss estimates on damage
factors for "average" buildings of a given structural type.
The only modification made to "average" building losses is
an approximate adjustment in shaking intensity to account
for amplified ground motion due to "poor soils".
Individual buildings are often different from
"average", and commonly exhibit characteristics of design
and construction that may modify their earthquake losses to
a level well above or below the average for its class.
Examples of such characteristics are building height,
maintenance condition, irregularities in configuration and
stiffness, proximity to adjacent structures, heavy
cladding, column stiffness variations, and age. Because of
the large number of possible combinations of
characteristics, a model based only on typical response for
75
There does not currently exist a comprehensive
systems approach to finding the optimal way to proceed with
seismic rehabilitation of the buildings in a large city.
The preceding review of the current methodologies point to
shortcomings in two fundamental areas: 1) earthquake loss
modeling, and 2) retrofit deciSion making.
Earthquake Loss Modeling
Damage Estimation. Current methodologies of
earthquake loss modeling do not utilize data on the actual
buildings being analyzed. FEMA-227, for example, follows
the lead of ATC-13, basing its loss estimates on damage
factors for "average" buildings of a given structural type.
The only modification made to "average" building losses is
an approximate adjustment in shaking intensity to account
for amplified ground motion due to "poor soils".
Individual buildings are often different from
"average", and commonly exhibit characteristics of design
and construction that may modify their earthquake losses to
a level well above or below the average for its class.
Examples of such characteristics are building height,
maintenance condition, irregularities in configuration and
stiffness, proximity to adjacent structures, heavy
cladding, column stiffness variations, and age. Because of
the large number of possible combinations of
characteristics, a model based only on typical response for
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
a given structural type may he very incorrect. For
example, in work on the Portland Seismic Hazards Survey
(PSHS), which catalogues these characteristics, the author
has seen many examples of structural performance scores
reflecting damage of only half of what is considered
"average" for the structural type; he has also seen even
more instances of scores indicating damage that is double,
triple, or more than what is "average" for the structure
type.
Clearly, it would he a major improvement if, in
addition to the effect of the structure type, an earthquake
loss model accounted for and combined the effect of
variations in design, construction, and soil condition
hazards.
Soil Profile Effects. It is well known that the
subsurface soil profile at a building site may have an
important effect on building response and consequent
losses. However, the existing models of this aspect of
seismic response relative to building losses are
incomplete. As noted, FEMA-227 includes a very approximate
model for ground motion amplification due to "poor soils. "
Hone of the current loss estimation models include
the most catastrophic of all seismic soil hazards:
liquefaction potential and dynamic slope instability.
Additionally, these hazards may have a combined effect with
76
a given structural type may he very incorrect. For
example, in work on the Portland Seismic Hazards Survey
(PSHS), which catalogues these characteristics, the author
has seen many examples of structural performance scores
reflecting damage of only half of what is considered
"average" for the structural type; he has also seen even
more instances of scores indicating damage that is douhle,
triple, or more than what is "average" for the structure
type.
Clearly, it would he a major improvement if, in
addition to the effect of the structure type, an earthquake
loss model accounted for and combined the effect of
variations in deSign, construction, and soil condition
hazards.
Soil Profile Effects. It is well known that the
subsurface soil profile at a huilding site may have an
important effect on huilding response and consequent
losses. However, the existing models of this aspect of
seismic response relative to huilding losses are
incomplete. As noted, FEHA-227 includes a very approximate
model for ground motion amplification due to "poor soils."
Hone of the current loss estimation models include
the most catastrophic of all seismic soil hazards:
liquefaction potential and dynamic slope instability.
Additionally, these hazards may have a comhined effect with
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
ground motion amplification. A better model would develop
a method to include the effect of all three of these soil
profile effects.
Retrofit Decision Analysis
Expected Value vs. Scenario Event. FEMA-227 uses an
"expected value" approach to compute net benefits and B/C
ratios to determine whether retrofit is economically
justified. In this method, avoided losses from retrofit
for a range of possible earthquake magnitudes are
multiplied by the annual probability of each magnitude. The
resulting annual benefit is summed over a planning horizon
by applying a Present Worth Factor to determine the present
value of the benefit.
Because there are orders of magnitude differences
between the earthquake return periods and any reasonable
planning horizon used for buildings, the value of
beneficial avoided losses using the expected value method
are far less than the value of avoided loss that would be
computed in a deterministic analysis for a single "scenario
earthquake event, " such as a Design Basis Earthquake
(usually considered as an event with a 10-percent chance
of exceedance in 50 years).
Since the retrofit cost is a constant, whether the
expected value or the scenario event approach is used makes
a big difference in calculation of net benefit and B/C
77
ground motion amplification. A better model would develop
a method to include the effect of all three of these soil
profile effects.
Retrofit Decision Analysis
Expected Value vs. Scenario Event. FEMA-227 uses an
"expected value" approach to compute net benefits and B/C
ratios to determine whether retrofit is economically
justified. In this method, avoided losses from retrofit
for a range of possible earthquake magnitudes are
multiplied by the annual probability of each magnitude. The
resulting annual benefit is summed over a planning horizon
by applying a Present Worth Factor to determine the present
value of the benefit.
Because there are orders of magnitude differences
between the earthquake return periods and any reasonable
planning horizon used for buildings, the value of
beneficial avoided losses using the expected value method
are far less than the value of avoided loss that would be
computed in a deterministic analysis for a single "scenario
earthquake event," such as a Design Basis Earthquake
(usually considered as an event with a i0-percent chance
of exceedance in 50 years).
Since the retrofit cost is a constant, whether the
expected value or the scenario event approach is used makes
a big difference in calculation of net benefit and B/C
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ratio. The retrofit plans resulting from using one
decision rule versus the other can lead to paradoxical
conclusions, as discussed in Chapter I.
This anomalous situation --- which in decision
theoretic terms is the difference between a "maximum
expected value" and a "maximin" decision rule --- is a
major point of debate in current retrofit decision
analysis.
78
The Heed for a Systems Approach. FEMA-227, the most
current benefit-cost methodology for retrofit decision
analysis, computes a retrofit B/C ratio one building at a
time. If the B/C ratio for a particular building exceeds
one, retrofit is deemed economically justified.
A systems approach would looK at establishing
priorities of retrofit for a group of buildings, in order
to maximize benefits and optimize investment from a
regional viewpoint.
Effective decision analysis must be defined with
respect to a decision maKer. The decision maKer in
regional seismic retrofit is the city, county or state
government legislating the retrofits. Currently
(1995-96), the City of Portland and State of Oregon are
searching for criteria by which to establish legislation
to encourage or require retrofit. The methods that
presently exist would be enhanced by integrating the
ratio. The retrofit plans resulting from using one
decision rule versus the other can lead to paradoxical
conclusions, as discussed in Chapter I.
This anomalous situation --- which in decision
theoretic terms is the difference between a "maximum
expected value" and a "maximin" decision rule --- is a
major point of debate in current retrofit decision
anal ysis.
78
The Heed for a Systems Approach. FEMA-227, the most
current benefit-cost methodology for retrofit decision
analysis, computes a retrofit B/C ratio one building at a
time. If the B/C ratio for a particular building exceeds
one, retrofit is deemed economically justified.
A systems approach would looK at establishing
priorities of retrofit for a group of buildings, in order
to maximize benefits and optimize investment from a
regional viewpoint.
Effective decision analysis must be defined with
respect to a decision maker. The decision maker in
regional seismic retrofit is the City, county or state
government legislating the retrofits. Currently
(1995-96), the City of Portland and State of Oregon are
searching for criteria by which to establish legislation
to encourage or require retrofit. The methods that
presently exist would be enhanced by integrating the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
systems approach.
Building Classification. Currently, buildings are
classified according to their structural types, i.e., wood
frame, steel frame, etc. However, structural type is one
of a number of attributes that contribute to earthquaKe
loss. As previously discussed, other attributes are
various characteristics of design and construction, the
site-specific soil profile, occupancy type, and occupant
load. currently, no classification system exists which
captures the combinations of these attributes.
OUTLINE OF THE RESEARCH
The present research formulates a seismic loss
estimation and retrofit planning methodology for buildings,
to estimate regional earthquaKe losses and determine
retrofit priorities and budgets.
The methodology utilizes seismologic ground motion
data, soil hazard maps, and ATC-21/RSP inventory data as
input to a soil effects model, an earthquake loss model,
and a retrofit direct benefit model. A building
classification system was formulated to aggregate buildings
which share similar loss and retrofit direct benefit
magnitudes.
A benefit-cost model is used as the basis for a
retrofit prioritization model and efficiency analysis.
79
systems approach.
Building Classification. Currently, buildings are
classified according to their structural types, i.e., wood
frame, steel frame, etc. However, structural type is one
of a number of attributes that contribute to earthquaKe
loss. As previously discussed, other attributes are
various characteristics of design and construction, the
site-specific soil profile, occupancy type, and occupant
load. Currently, no classification system exists which
captures the combinations of these attributes.
OUTLINE OF THE RESEARCH
The present research formulates a seismic loss
estimation and retrofit planning methodology for buildings,
to estimate regional earthquaKe losses and determine
retrofit priorities and budgets.
The methodology utilizes seismologic ground motion
data, soil hazard maps, and ATC-21/RSP inventory data as
input to a soil effects model, an earthquaKe loss model,
and a retrofit direct benefit model. A building
classification system was formulated to aggregate buildings
which share similar loss and retrofit direct benefit
magnitudes.
A benefit-cost model is used as the basis for a
retrofit prioritization model and efficiency analysis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
80
Results are compared from an Expected Value and an
EarthquaKe Scenario Event retrofit analysis. The goal is
to obtain the maximum system (regional) benefit (reduced
loss) for each retrofit dollar spent. Retrofit
prioritization also enables the scope of a regional
retrofit program to be established: at what point does the
decision-maker opt to "cut-off" the retrofit plan the
point of optimum benefit-cost efficiency --- and what is
the estimated budgets for the optimal retrofit program.
The research consists of the following sequence:
1. Development of an improved EarthquaKe Loss Model:
a) Seismic vulnerability model for buildings.
b) Soil effects model for buildings.
c) Building damage and loss of life models.
2. Development of a retrofit direct benefit model.
a) Retrofit effectiveness model.
b) Retrofit direct benefit and cost models.
3. Development of a Retrofit Planning methodology:
a) Building Classification System.
b) Retrofit cost-benefit analysis models.
c) Retrofit prioritization model.
d) Retrofit system optimization model.
3. Utilizing the methodologies developed, computer
software was developed to estimate earthquake
losses and plan a retrofit program for a region.
80
Results are compared from an Expected Value and an
EarthquaKe Scenario Event retrofit analysis. The goal is
to obtain the maximum system (regional) benefit (reduced
loss) for each retrofit dollar spent. Retrofit
prioritization also enables the scope of a regional
retrofit program to be established: at what point does the
decision-maKer opt to "cut-off" the retrofit plan the
point of optimum benefit-cost efficiency --- and what is
the estimated budgets for the optimal retrofit program.
The research consists of the following sequence:
1. Development of an improved Earthquake Loss Hodel:
a) Seismic vulnerability model for buildings.
b) Soil effects model for buildings.
c) Building damage and loss of life models.
2. Development of a retrofit direct benefit model.
a) Retrofit effectiveness model.
b) Retrofit direct benefit and cost models.
3. Development of a Retrofit Planning methodology:
a) Building Classification System.
b) Retrofit cost-benefit analysis models.
c) Retrofit prioritization model.
d) Retrofit system optimization model.
3. Utilizing the methodologies developed, computer
software was developed to estimate earthquake
losses and plan a retrofit program for a region.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
4. Losses were estimated for 7, 500 non-residential
buildings in the Portland, Oregon Quadrangle, for
four earthquake scenario events, with emphasis on
the Design Basis Earthquake event (a magnitude
with a 10Y. chance of exceedance in 50 years).
5. A retrofit analysis was made of 7, 500 non-
residential buildings in the Portland Quadrangle.
Results from a Design Basis Earthquake Scenario
are compared to an Expected Value Analysis.
A flowchart depicting the components and general
process in the earthquake loss model is shown in the first
page of Figure 6, following.
A flowchart depicting the components and general
process in the retrofit planning methodology is shown in
the second page of Figure 6, following.
81
4. Losses were estimated for 7.500 non-residential
buildings in the Portland. Oregon Quadrangle. for
four earthquake scenario events. with emphasis on
the Design Basis Earthquake event (a magnitude
with a 10Y. chance of exceedance in 50 years).
5. A retrofit analysis was made of 7.500 non-
residential buildings in the Portland Quadrangle.
Results from a Design Basis Earthquake Scenario
are compared to an Expected Value Analysis.
A flowchart depicting the components and general
process in the earthquake loss model is shown in the first
page of Figure 6. following.
A flowchart depicting the components and general
process in the retrofit planning methodology is shown in
the second page of Figure 6. following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
~-- -- -PSHS-93 INVENTORY DATA-- -- -- I I I HUMBER OF PEOPLE I I I
OCCUPANCY TYPE
BUILDING SIZE
.----------. STRUCTURE TYPE
PERFORMANCE MODIFIERS
BUILDING LOCATION
L---- ------------- __ _l
,----l SOIL 1 SOIL HAZARD
MAPS I EFFECTS --;----1 MODEL ~----------~
I I I FRAGILITY H-~:: ::~_j
EQ DAMAGE EXPERT OPINION
~-~ EQ INTENSITY DATA
~--------D~G-Eil
I ESTIMATED FACTOR I VALUE OF BUILDING I I DAMAGE I
II ESTIMATED I !
~~F~F~ -------- J
CONSTRUCTION COST DATA
EQ CASUALTY DATA
Figure 6. Earthquake loss modeling flowchart.
82
1---- -PSHS-93
I I HUMBER OF PEOPLE I
INVENTORY DATA -- -- -- I STRUCTURE TYPE
I I
OCCUPANCY TYPE
BUILDING SIZE
PERFORMANCE MODIFIERS
BUILDING LOCATION
L ___________________ -1
,----l SOIL I SOIL HAZARD
MAPS I EFFECTS ~I MODEL ~----------~
I I I FRAGILITY H-~::~:_J
EQ DAMAGE EXPERT OPINION
~_~ EQ INTENSITY DATA
I--------D~G-E l I ESTIMATED FACTOR I
VALUE OF BUILDING t------~---..:I:..---__I CONSTRUCTION I DAMAGE I COST DATA
II ESTIMATED l ! ~~F~F~ ________ J EQ CASUALTY
DATA
Figure 6. Earthquake loss modeling flowchart.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I PSHS-93 INVENTORY DATA I EARTHQUAKE LOSS MODEL
• RETROFIT RETROFIT BENEFIT MODEL EFFECTIVENESS
EXPERT OPINION
-----------~ I ESTIMATED ESTIMATED VALUE OF LIVES SAVED BUILDING DAMAGE I
AVOIDED
~ I L --- ---·i----:.J
RETROFIT RETROFIT COST COST-BENEFIT
MODEL VALUE OF LIFE
BUILDING CLASSIFICATION
SYSTEM
RETROFIT PRIORITIZATION
MODEL
RETROFIT DISCOUNT RATE EFFICIENCY
DATA
DATA
ANALYSIS PLANNING HORIZON
OPTIMAL RETROFIT PROGRAM
Figure 6, continued.
83
I PSHS-93 INVENTORY DATA
EARTHQUAKE LOSS MODEL
• RETROFIT RETROFIT BENEFIT MODEL EFFECTIVENESS
EXPERT OPINION
-----------~ I ESTIMATED ESTIMATED VALUE OF LIVES SAVED BUILDING DAMAGE I
AVOIDED
~ I L --- ___ .~ ____ :.J
RETROFIT RETROFIT COST DATA COST-BENEFIT
MODEL VALUE OF LIFE DATA
BUILDING CLASSIFICATION
SYSTEM
RETROFIT PRIORITIZATION
MODEL
RETROFIT DISCOUNT RATE EFFICIENCY
ANALYSIS PLANNING HORIZON
OPTIMAL RETROFIT PROGRAM
Figure 6, continued.
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER IV
DEVELOPMENT OF AN EARTHQUAKE LOSS MODEL FOR EXISTING BUILDINGS
An EarthquaKe Loss Model was developed to estimate
earthquaKe losses for existing buildings.
Model inputs are:
1. Building Inventory: ATC-21 Rapid Survey
Data.
2. EarthquaKe Hazard Map indices.
The model outputs are estimates, for 6 values of
PeaK Ground Acceleration (PGA), of the following:
1. Damage Factor.
2. Dollar value of damage.
3. Loss of life & serious injuries to occupants.
The loss model extends current loss estimation
methodologies by including the effect of the following
attributes of each individual building and site in the
inventory:
1. 12 Structural Types, per ATC-21.
2. 10 Structural Attributes, per ATC-21.
3. 3 Soil Attributes, from GMS-79.
4. 9 Occupancy Types, per ATC-21.
5. 7 Occupancy Levels, per PSHS-93.
CHAPTER IV
DEVELOPMENT OF AN EARTHQUAKE LOSS MODEL FOR EXISTING BUILDINGS
An EarthquaKe Loss Model was developed to estimate
earthquaKe losses for existing buildings.
Model inputs are:
1. Building Inventory: ATC-21 Rapid Survey
Data.
2. EarthquaKe Hazard Map indices.
The model outputs are estimates, for 6 values of
PeaK Ground Acceleration (PGA) , of the following:
1. Damage Factor.
2. Dollar value of damage.
3. Loss of life & serious injuries to occupants.
The loss model extends current loss estimation
methodologies by including the effect of the following
attributes of each individual building and Site in the
inventory:
1. 12 Structural Types, per ATC-21.
2. 10 Structural Attributes, per ATC-21.
3. 3 Soil Attributes, from GMS-79.
4. 9 Occupancy Types, per ATC-21.
5. 7 Occupancy Levels, per PSHS-93.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
The Damage Factor estimate is based on correlations
between ATC-21's Structural Score and ATC-13's Mean Best
Damage Factor. Dollar value of damage is estimated from
Damage Factor, building area, and occupancy type. Loss of
life and serious injury estimates were derived from
occupancy level, damage factor, and ATC-13's Table 9. 3.
FORMATS FOR EXPRESSING SEISMIC VULNERABILITY
The expert opinion estimates of Damage Factor
summarized in ATC-13's Appendix G is an expression of
building seismic vulnerability in "damage function"
format, giving damage estimates for discrete steps of
earthquaKe intensity. An alternative expression of
seismic vulnerability is the "fragility curve" format,
wherein the damage factor is modeled as a continuous
function of seismic intensity.
Ground Motion Characterization
In ATC-13, the Damage Factor (DF) is given at seven
discrete steps, for seven levels of Modified Mercalli
Intensity (HHI). The HHI levels summarized in Appendix G
are VI through XII. HHI was used to characterize
earthquaKe intensity in ATC-13 because, as previously
noted, most expert Knowledge of building damage from
earthquaKes exists in that form. However, to relate the
shaKing intensity to currently available seismologic data,
85
The Damage Factor estimate is based on correlations
between ATC-21's Structural Score and ATC-13's Mean Best
Damage Factor. Dollar value of damage is estimated from
Damage Factor, building area, and occupancy type. Loss of
life and serious injury estimates were derived from
occupancy level, damage factor, and ATC-13's Table 9.3.
FORMATS FOR EXPRESSING SEISMIC VULNERABILITY
The expert opinion estimates of Damage Factor
summarized in ATC-13's Appendix G is an expression of
building seismic vulnerability in "damage function"
format, giving damage estimates for discrete steps of
earthquaKe intensity. An alternative expression of
seismic vulnerability is the "fragility curve" format,
wherein the damage factor is modeled as a continuous
function of seismic intensity.
Ground Motion Characterization
In ATC-13, the Damage Factor (DF) is given at seven
discrete steps, for seven levels of Modified Mercalli
Intensity (HHI). The HMI levels summarized in Appendix G
are VI through XII. HMI was used to characterize
earthquaKe intensity in ATC-13 because, as previously
noted, most expert Knowledge of building damage from
earthquaKes exists in that form. However, to relate the
shaKing intensity to currently available seismologic data,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
it will be helpful to base both the damage function and
fragility curves on Peak Ground Acceleration (PGA). ATC-21
suggests the relationship developed by Richter (1958) to
convert HMI to PGA, as follows:
PGA (cm;s2) ~ 1o(MMI-1)/3 (4.1)
More commonly, PGA is expressed as a decimal fraction of
the acceleration of gravity, g(32. 2 ft;s2):
PGA (g) ~ 0.00102 x 1o(MMI-1)/3 (4. 1a)
Damage Function Format
An example of a damage function is shown in Table
XXI for TILT-UP buildings (ATC-13 Facility Class 21). In
Table XXI, the discrete steps of seismic intensity are
given in both MMI and PGA, versus ATC-13's Mean Best
Estimate (MB) of Damage Factor.
TABLE XXI
DAMAGE FUNCTION FOR TILT UP BUILDINGS FROM ATC-13 APPENDIX G
MMI PGA MB
VI . 05 1. 50 VII . 10 4.20 VIII . 22 10.60 IX . 47 18. 50 X L 02 28.70
86
it will be helpful to base both the damage function and
fragility curves on Peak Ground Acceleration (PGA). ATC-21
suggests the relationship developed by Richter (1958) to
convert HMI to PGA, as follows:
PGA (cm/s2) ~ 10(HMI-1)/3 (4. 1)
More commonly, PGA is expressed as a decimal fraction of
the acceleration of gravity, g(32.2 ft/s2):
PGA (g) z 0.00102 x 10(HMI-1)/3 (4. 1 a)
Damage Function Format
An example of a damage function is shown in Table
XXI for TILT-UP buildings (ATC-13 Facility Class 21). In
Table XXI, the discrete steps of seismic intensity are
given in both HMI and PGA, versus ATC-13 J s Mean Best
Estimate (MB) of Damage Factor.
TABLE XXI
DAMAGE FUNCTION FOR TILT UP BUILDINGS FROM ATC-13 APPENDIX G
HMI PGA MB
VI .05 1. 50 VII . 10 4.20 VIII .22 10.60 IX .47 18. 50 X L 02 28.70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
Fragility Curve Format
As an example of a fragility curve, the TILT-UP
damage function in Table XXI is transformed to a
fragility curve by regressing MB against PGA, resulting in
a cubic polynomial,
where b0 , b 1, b2, and b3 are the polynomial coefficients.
This fragility curve and its polynomial coefficients are
shown in Fig. 7.
60 0:::: 0 50 I-u <(
40 LL. w C> 30 <( ~ <( 20 0 z <( 10 w ~
0
'
I
b(O =-2.0~40
I I b(1 =72.3~3
·'""' - .,.,.:,.., I
I
-\~ "'i'"'"
l I
b(3~=35.1i I I
R-sq=10~% I
'
I L----t---r--11 I ---!..........-~~
' l I
~~I I
\ I I I I ' I I I I I I I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 7. Fragility curve for TILT UP buildings derived from Table XXI.
87
Fragility Curve Format
As an example of a fragility curve, the TILT-UP
damage function in Table XXI is transformed to a
fragility curve by regressing HE against PGA, resulting in
a cubic polynomial,
where bO, bt, b2' and b3 are the polynomial coefficients.
This fragility curve and its polynomial coefficients are
shown in Fig. 7.
I I
0::: 60 I: b(O =-2.0940 I b(1 =72.3~3 I 01- 50 .,,.,1 - ..,.., ..,,.,
! "'\~ -, .~.. I U ! I b(3~=35.1~ I ~ 40 --I-l~-+--=-+---:-=-::::l::-:-I--+_-+-_-l---+--!----+---l-j W R-SQ=100l%
~ 30 ~ I I_~I"I ~20 vi 'I « 1 0 -H--+------::;;~! ........... "----+-----+--+--l---+-----l----+---l-jl
~ YI Ii! I \ O~,==~~~~~~~~r,~~~=t I I ! I I I I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 7. Fragility curve for TILT UP buildings derived from Table XXI.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
88
Significance and Limitations
The fragility curve format was chosen for this
research, because it provides a continuous estimate of
damage vs. seismic intensity. This is helpful in maKing
regional damage estimates based on probabilistic estimates
for a range of potential ground motions.
The fragility curve developed here for TILT-UP is
subject to the same serious limitations inherent in all
the ATC-13 damage estimates, as follows:
1) The damage estimates apply to "average" buildings
of the given structural type (in this case, TILT-
UP). The characteristics of design, construction,
and site conditions of specific buildings are not
taKen into account. Such individual variations
can and frequently do result in damages much
different from "average" for the building type.
2) The damage estimates apply to buildings designed
and constructed in California. Since earthquaKes
have been recognized and designed for in
California long in advance of other localities,
it can be expected that the same building type in
a different locality will typically suffer more
damage than a "California" building.
88
Significance and Limitations
The fragility curve format was chosen for this
research, because it provides a continuous estimate of
damage vs. seismic intensity. This is helpful in maKing
regional damage estimates based on probabilistic estimates
for a range of potential ground motions.
The fragility curve developed here for TILT-UP is
subject to the same serious limitations inherent in all
the ATC-13 damage estimates, as follows:
1) The damage estimates apply to "average" buildings
of the given structural type (in this case, TILT-
UP). The characteristics of design, construction,
and site conditions of specific buildings are not
taKen into account. Such individual variations
can and frequently do result in damages much
different from "average" for the building type.
2) The damage estimates apply to buildings designed
and constructed in California. Since earthquaKes
have been recognized and designed for in
California long in advance of other localities,
it can be expected that the same building type in
a different locality will typically suffer more
damage than a "California" building.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
THE BASIS OF THE ATC-21 SCORING SYSTEM
ATC-21 is the only peer-reviewed research effort to
date which utilizes the extensive estimated damage data
from ATC-13, while addressing the two limitations
discussed above: 1) accounting for the impact of estimated
damage caused by characteristics of design, construction,
and site conditions in specific buildings, while 2) taKing
into account the expected increase in damage for
"non-California" buildings.
The purpose of ATC-21 was not to formulate enhanced
fragility curves, but rather to separate the buildings
in a regional inventory into those which are likely to be
a seismic life-safety risk from those which are not. To
accomplish this, ATC-21 formulated the concept of a
Structural Score (S), and then determined a value of S to
use as a "cut-off" between seismic high-risk versus
non-high-risk buildings.
The Structural Score S is defined in ATC-21 as the
negative of the logarithm (base 10) of the probability of
the Damage Factor (DF) exceeding 60-percent (major
damage):
S = -log10 [Pr(DF >= 60Y.)] ( .q., 3)
The Structural Score S has a straightforward
interpretation: if S equals 1, the probability of major
damage is 1 in 10; if S equals 2, the probability of major
89
THE BASIS OF THE ATC-21 SCORING SYSTEM
ATC-21 is the only peer-reviewed research effort to
date which utilizes the extensive estimated damage data
from ATC-13, while addressing the two limitations
discussed above: 1) accounting for the impact of estimated
damage caused by characteristics of design, construction,
and site conditions in specific buildings, while 2) taking
into account the expected increase in damage for
"non-California" buildings.
The purpose of ATC-21 was not to formulate enhanced
fragility curves, but rather to separate the buildings
in a regional inventory into those which are likely to be
a seismic life-safety risk from those which are not. To
accomplish this, ATC-21 formulated the concept of a
Structural Score (S), and then determined a value of S to
use as a "cut-off" between seismic high-risk versus
non-high-risk buildings.
The Structural Score S is defined in ATC-21 as the
negative of the logarithm (base 10) of the probability of
the Damage Factor (DF) exceeding 50-percent (major
damage) :
S = -logiO [Pr(DF >= 50X») (4. 3)
The Structural Score S has a straightforward
interpretation: if S equals 1, the probability of major
damage is i in 10i if S equals 2, the probability of major
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
damage is 1 in 100, and so on. S can also be visualized
as a "performance rating" --- the higher the score, the
better the estimated seismic performance and the less the
damage.
Incorporated within S for a specific building are
the effects of:
1) a specified seismic intensity,
2) the building type, and
3) characteristics of design, construction, and site
conditions of the specific building.
s is determined for a specific building by adding the
Basic Structural Hazard (BSH) score to the sum of the
Performance Modification Factors (PMF, as defined in
Chapter II), i.e.,
S : BSH + E (PMF) ( 4. 4)
The BSH is a generic score for an "average" building
of its type, at a specified seismic intensity. The BSH
score is modified for a specific building by the PMFs,
reflecting the characteristics of design, construction,
and site conditions of that specific building, that cause
its seismic response to differ from "average."
Clearly, E(PMF) = 0 reflects a building that does not
have characteristics that are different from the average
building of its type. This is reflected in Eq. 4.4, which
shows that when E(PMF) = 0, S = BSH. Therefore, the BSH is
90
damage is 1 in 100, and so on. S can also be visualized
as a "performance rating" --- the higher the score, the
better the estimated seismic performance and the less the
damage.
Incorporated within S for a specific building are
the effects of:
1) a specified seismic intensity,
2) the building type, and
3) characteristics of design, construction, and site
conditions of the specific building.
S is determined for a specific building by adding the
Basic Structural Hazard (BSH) score to the sum of the
Performance Modification Factors (PMF, as defined in
Chapter II), i.e.,
S : BSH + E (PMF) (4. 4)
The BSH is a generic score for an "average" building
of its type, at a specified seismic intensity. The BSH
score is modified for a specific building by the PMFs,
reflecting the characteristics of design, construction,
and Site conditions of that specific building, that cause
its seismic response to differ from "average. "
Clearly, E(PMF) : 0 reflects a building that does not
have characteristics that are different from the average
building of its type. This is reflected in Eq. 4.4, which
shows that when E(PMF) : 0, S = BSH. Therefore, the BSH is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
91
the score for an "average" building of its type.
The Basic Structural Hazard Score
ATC-21, recognizing the inherent variability in
structural response to seismic motion, treats the DF as a
random variable. To determine the BSH scores, the
uncertainty in the DF at a given ground motion was
estimated from the expert opinion damage estimates of
ATC-13 (this is appropriate, since both the DF estimates
in ATC-13 and the BSH are for "average" buildings).
Appendix G of ATC-13 contains weighted statistics of the
DF. The weights reflect the experience level and
confidence of the individual experts, who provided low,
best, and high estimates of the DF for the facility
classes at 6 levels of MMI.
The BSH score was developed by taKing ATC-13's mean
low (ML) and mean high (MH) estimates of the DF as the
90-percent probability bounds of the DF distribution. The
mean best (MB) estimate was interpreted as the median DF
(the DF most likely to be observed for a given MHI and
facility class). The uncertainty about the mean was
examined and found to be acceptably modeled by either a
Beta or a lognormal distribution. For convenience, ATC-21
selected the lognormal distribution, primarily because it
offers the advantage of easier calculations using
well-known ploynomial approximations.
91
the score for an "average" building of its type.
The Basic Structural Hazard Score
ATC-2t, recognizing the inherent variability in
structural response to seismic motion, treats the DF as a
random variable. To determine the BSH scores, the
uncertainty in the DF at a given ground motion was
estimated from the expert opinion damage estimates of
ATC-13 (this is appropriate, since both the DF estimates
in ATC-13 and the BSH are for "average" buildings).
Appendix G of ATC-13 contains weighted statistics of the
DF. The weights reflect the experience level and
confidence of the individual experts, who provided low,
best, and high estimates of the DF for the facility
classes at 6 levels of MMI.
The BSH score was developed by taking ATC-13's mean
low (ML) and mean high (MH) estimates of the DF as the
90-percent probability bounds of the DF distribution. The
mean best (ME) estimate was interpreted as the median DF
(the DF most likely to be observed for a given HHI and
facility class). The uncertainty about the mean was
examined and found to be acceptably modeled by either a
Beta or a lognormal distribution. For convenience, ATC-21
selected the lognormal distribution, primarily because it
offers the advantage of easier calculations using
well-known ploynomial approximations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
The lognormally distributed random variable DF has a
related random variable, ln(DF), which is normally
distributed. The mean, a, of the normally distributed
random variable is related to the median of the
lognormally distributed random variable by:
a = ln(HB) (4. 5)
The standard deviation, S, of the normally
distributed random variable ln(DF) can be found by taking
HL and MH as the 90-percent probability bounds of the DF
distribution. Thus approximately 95-percent of the
distribution lies below MH. From tables of the cumulative
standard normal distribution, F(z=1.64) = o. 95, where z is
the number of standard deviations from the mean (the
well-known "z-transform") and F(z) is the cumulative
probability distribution below z. The standard deviation
may then be calculated from S = [ln(HH)-a]/1.64.
Similarly, 95Y. of the DF distribution lies above HL, and
F(z=-1.64) = .05, therefore if ln(DF) was perfectly
normally distributed, S = [a-ln(HL)]/1. 64. The average of
these equations was used to approximate the standard
deviation of the normal distribution, as follows:
S = [ln(HH)-ln(HL)] I 3. 28 (4. 6)
Once the parameters, a and s. of the normal
distribution of ln(DF) are found, the probability of the
DF being greater than 60 is calculated from a polynomial
92
The lognormally distributed random variable DF has a
related random variable, In(DF), which is normally
distributed. The mean, a, of the normally distributed
random variable is related to the median of the
lognormally distributed random variable by:
a : In(HB) (4. 5)
The standard deviation, S, of the normally
distributed random variable In(DF) can be found by taking
HL and MH as the 90-percent probability bounds of the DF
distribution. Thus approximately 95-percent of the
distribution lies below MH. From tables of the cumulative
standard normal distribution, F(z:1.64) : 0.95, where z is
the number of standard deviations from the mean (the
well-known Hz-transform") and F(z) is the cumulative
probability distribution below z. The standard deviation
may then be calculated from S : [In(MH)-a)ll.64.
Similarly, 95Y. of the DF distribution lies above HL, and
F(z:-1.64) : .05, therefore if In(DF) was perfectly
normally distributed, S : [a-ln(HL)1/1.64. The average of
these equations was used to approximate the standard
deviation of the normal distribution, as follows:
S : [In(MH)-ln(HL)) I 3.28 (4.6)
Once the parameters, a and S, of the normal
distribution of In(DF) are found, the probability of the
DF being greater than 60 is calculated from a polynomial
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
approximation. This is necessary because frequently the
probabilities are very low, and beyond the range of
published cumulative standard normal distribution tables.
A polynomial approximation from NBS 55 (1964) was used in
ATC-21, as follows, where Q(z) = Pr(DF>60):
Q(Z) = Z(Z) * [b1t + b2t2 + b3t3 + b4t4 + b5t5J
where: Z(Z): (21T)-.5*eXp(-z2/2) z = [ln(60)-a:]/13 t : 1/(i+PZ) p = . 2316419 b1 = . 319381530 b2 = -. 356563782 b3 = 1. 781477937 b4 = -1. 821255978 b5 = 1. 330274429
To illustrate the calculation of BSH, in the
following example the BSH score is calculated for
(4. 7)
California TILT-UP buildings at PGA : . 22 (HMI : VIII).
The weighted statistics of the damage factor from ATC-13
for TILT-UP buildings are shown here in Table XXII.
TABLE XXII
WEIGHTED STATISTICS OF THE DAMAGE FACTOR FOR TILT-UP BUILDINGS FROM
ATC-13, APPEHDIX G
HMI PGA ML MB MH
VI . 05 0.40 1. 50 4. 20 VII . 10 1. 80 4. 20 9. 60 VIII . 22 4.00 10.60 18. 20 IX . 47 9. 10 18. 50 31. 60 X 1. 02 15. 20 28.70 49. 20
93
approximation, This is necessary because frequently the
probabilities are very low, and beyond the range of
published cumulative standard normal distribution tables.
A polynomial approximation from NBS 55 (1964) was used in
ATC-21, as follows, where G(z) = Pr(DF>60):
G(z) = Z(z) * [b1t + b2t2 + b3t3 + b4t4 + b5t5]
where: Z(z) = (21T) -.5 * exp(-z2/2) z = [In(60)-a)/j3 t = 1/(1+pz) p = .2316419 b1 = .319381530 b2 : -. 356563782 b3 = 1.781477937 b4 : -1. 821255978 b5 : 1.330274429
To illustrate the calculation of BSH, in the
following example the BSH score is calculated for
(4.7)
California TILT-UP buildings at PGA : .22 (MMI : VIII).
The weighted statistics of the damage factor from ATC-13
for TILT-UP buildings are shown here in Table XXII.
TABLE XXII
WEIGHTED STATISTICS OF THE DAMAGE FACTOR FOR TILT-UP BUILDINGS FROM
ATC-13, APPENDIX G
MMI PGA HI. HE HH
VI .05 0.40 1. 50 4. 20 VII . 10 1. 80 4. 20 9. 60 VIII .22 4.00 10.60 18. 20 IX .47 9.10 18, 50 31. 60 X 1. 02 15.20 28.70 49. 20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
The lognormal distribution of the estimated DF at
PGA = . 22 for TILT-UP buildings is illustrated in Fig. 8.,
following.
5%
ML 4.00
MB 10.60
MH 18.20
Figure 8. Lognormal distribution of estimated DF at PGA = . 22 for average California TILT-UP bldgs.
Assuming the DF is lognormally distributed, the
standard deviation ~. and the mean a, of the related
normally distrubuted ln(DF), are:
~ = [ln(18. 20) - ln(4. 00)] 1 3. 28 = 0. 461929
a= ln(10.60) = 2. 3608540
The standard normal variate, z, is:
z = [ln(60) - 2. 3608540] 1 0.461929 = 3.75272
From the polynomial approximation, the probability of the
DF exceeding 60-percent, Q(z), is:
Q(Z) : .00008749054
The Basic Structural Hazard score is:
BSH = -log10(.00008749054) = 4.05
94
The lognormal distribution of the estimated DF at
PGA = .22 for TILT-UP buildings is illustrated in Fig. 8.,
following.
5%
ML 4.00
MB 10.60
MH 18.20
Figure 8. Lognormal distribution of estimated DF at PGA = . 22 for average California TILT-UP bldgs.
Assuming the DF is lognormally distributed, the
standard deviation ~J and the mean a. of the related
normally distrubuted In(DF). are:
~ = [In(18.20) - In(4. 00)] / 3.28 = 0.461929
a = In(10.60) = 2.3608540
The standard normal variate. z, is:
z = [In(60) - 2.3608540] / 0.461929 = 3.75272
From the polynomial approximation. the probability of the
DF exceeding 60-percent, G(z), is:
G(z) = .00008749054
The Basic Structural Hazard score is:
BSH = -log10(.00008749054) = 4.05
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
95
The normally distributed ln(DF) is illustrated in Fig. 9,
following.
/Q(z)=0.000087 49 I
ln(MB) ln(60) 2.361 4.094
f<Ets z ~
Figure 9. Normal distribution of ln(DF) at PGA = .22 for average California TILT-UP buildings.
Extension to non-California Buildings
As previously noted, ATC-21 addresses damage in
regions where building design and construction practices
differ from California. It is expected that in regions
where seismic loading has historically had less influence
in structural design, damage would be greater (and
Structural Score lower) for a given seismic intensity.
ATC-21 solicited expert opinions from experienced
engineers in NEHRP Map Areas 1 to 6, asking them to
compare the performance of specific building types in
their regions to "California" buildings of the same type.
As expected, for the same level of seismic intensity the
95
The normally distributed In(DF) is illustrated in Fig. 9,
following.
In(MB) 2.361
k-tS z
/Q(Z)=0.00008749 I
In(60) 4.094 ~
Figure 9. Normal distribution of In(DF) at PGA = .22 for average California TILT-UP buildings.
Extension to non-California Buildings
As previously noted, ATC-21 addresses damage in
regions where building deSign and construction practices
differ from California. It is expected that in regions
where seismic loading has historically had less influence
in structural deSign, damage would be greater (and
Structural Score lower) for a given seismic intensity.
ATC-21 solicited expert opinions from experienced
engineers in NEHRP Map Areas 1 to 6, asking them to
compare the performance of specific building types in
their regions to "California" buildings of the same type.
As expected, for the same level of seismic intenSity the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
96
experts predicted higher damage for buildings in their
regions than for similar California buildings.
From the results of the expert opinions, ATC-21
determined that the non-California DF differed from the
California buildings DF by a constant multiple, and
developed a table of Modification Constants (MC), which
vary with NEHRP Map Area and Building Type. The values of
MC are summarized in Table XXIII, following, taken from
Table B3 in ATC-21-1.
w S1 S2,S4 S3 C3/S5 C1 C2 PC1 PC2 RM URM
TABLE XXIII
DAMAGE FACTOR MODIFICATION CONSTANTS FOR NOH-CALIFORNIA BUILDINGS
FROM ATC-21-1 TABLE B3
NEHRP Map Area
1, 2 3 4 5
1.0 1.3 1.3 1.2 1.9 1.2 1.4 1.3 1.9 1.2 1.4 1.1 1. 1 1. 1 1.3 1.3 1.2 1.2 1.3 1.3 2. 2 1.3 1.5 1.2 1.7 1.3 1.5 1.1 2. 0 1.2 1.5 1.3 2. 9 1. 1 1.8 1.2 2. 9 1. 1 1.3 1. 1 1. 1 1.2 1.0 1.0
6
1.0 1.0 1. 1 1.2 1.2 1.0 1.0 1.4 1.3 1.0 1.0
In developing the BSH scores for non-California
regions, the MC was used to change the value MB from
96
experts predicted higher damage for buildings in their
regions than for similar California buildings.
From the results of the expert opinions, ATC-21
determined that the non-California DF differed from the
California buildings DF by a constant multiple, and
developed a table of Modification Constants (MC), which
vary with NEHRP Map Area and Building Type. The values of
MC are summarized in Table XXIII, following, taken from
Table B3 in ATC-21-1.
W S1 S2,Sl!-S3 C3/S5 Cl C2 PCl PC2 RM URM
TABLE XXIII
DAMAGE FACTOR MODIFICATION CONSTANTS FOR NON-CALIFORNIA BUILDINGS
FROM ATC-21-1 TABLE B3
NEHRP Map Area
1,2 3 l!- 5
1.0 1.3 1.3 1.2 1.9 1.2 1.l!- 1.3 1.9 1.2 1.l!- 1.1 1. 1 1.1 1.3 1.3 1.2 1.2 1.3 1.3 2.2 1.3 1.5 1.2 1.7 1.3 1.5 1.1 2.0 1.2 1.5 1.3 2. 9 1.1 1.8 1.2 2. 9 1. 1 1.3 1.1 1. 1 1.2 1.0 1.0
6
1.0 1.0 1. 1 1.2 1.2 1.0 1.0 1.l!-1.3 1.0 1.0
In developing the BSH scores for non-California
regions, the MC was used to change the value MB from
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
ATC-13 to a Best Estimate for each NEHRP Map Area (BENA)
according to the following:
BENA = MC * MB (4. 8)
The procedure used by ATC-21 to calculate the BSH
score for non-California buildings is to hold the standard
deviation S constant, and calculate the mean of ln(DF) by
using BENA in lieu of MB in Eq. 4. 5, i.e.,
ex = ln (BENA) (4. 5a)
A comparison of resulting scores for different NEHRP
Map Areas showed little difference in BSH scores for
certain levels of seismicity. Therefore, the BSH scores
were simplified by grouping and averaging the calculated
scores for three levels of "Seismicity:" High, Moderate,
and Low. The NEHRP Map Areas grouped together and the
corresponding seismic intensities are shown in Table XXIV.
TABLE XXIV
ATC-21 SEISMICITY BY NEHRP MAP AREA
Seismicity Map Area EPA PGA
LOW 1 . 05 . 07 LOW 2 . 05 . 07
MODERATE 3 .10 . 13 MODERATE 4 . 15 . 20
HIGH 5 . 20 . 27 HIGH 6 . 30 .40 HIGH 7 .40 . 53
97
ATC-13 to a Best Estimate for each NEHRP Map Area (BENA)
according to the following:
BENA = MC * ME (4. 8)
The procedure used by ATC-21 to calculate the BSH
score for non-California buildings is to hold the standard
deviation S constant, and calculate the mean of In(DF) by
using BENA in lieu of MB in Eq. 4. 5, i. e. ,
ex : In (BENA) (4. 5a)
A comparison of resulting scores for different NEHRP
Map Areas showed little difference in BSH scores for
certain levels of seismicity. Therefore, the BSH scores
were simplified by grouping and averaging the calculated
scores for three levels of "Seismicity:" High, Moderate,
and Low. The NEHRP Map Areas grouped together and the
corresponding seismic intensities are shown in Table XXIV.
TABLE XXIV
ATC-21 SEISMICITY BY NEHRP HAP AREA
Seismicity Hap Area EPA PGA
LOW 1 .05 .07 LOW 2 .05 .07
MODERATE 3 .10 . 13 MODERATE 4 . 15 .20
HIGH 5 .20 .27 HIGH 6 .30 . 40 HIGH 7 . 40 .53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
By grouping and averaging the BSH scores into the
three seismicities, ATC-21 produced 3 sets of BSH scores.
For the example of TILT-UP (PC1), after rounding, the BSH
scores used in ATC-21 were:
LOW 3. 5 MODERATE 3. 5 HIGH 2.0
To illustrate the effect of the HC on BSH score, in
the following example the BSH score will be re-calculated
for a TILT-UP building in a MODERATE Seismicity zone. From
Table XXIV, the HEHRP Hap Areas corresponding to MODERATE
are Areas 3 and 4. From Table XXIII, the HC values for
PC1 for Hap Areas 3 and 4 are 1.2 and 1. 5, respectively;
the average HC is 1. 35. BEHA is calculated from Eq. 4. 8:
BEHA : 1. 35 M (10. 60) : 14.30
The rest of the calculations are adjusted as follows:
a= ln(14. 30) = 2.6602595
The standard normal variate, z, is:
z = [ln(60) - 2.6602595] 1 0.4619291 = 3.1045566
From the polynomial approximation, the probability o£ the
DF exceeding 60-percent, Q(z), is:
Q(z) = .000952892
The Basic Structural Hazard score is:
BSH = -log10(.000952892) = 3.02
As expected, the increase in damage from the HC results in
a reduced BSH score (now 3. 02, reduced from 4.05 using HB).
98
By grouping and averaging the BSH scores into the
three seismicities, ATC-21 produced 3 sets of BSH scores.
For the example of TILT-UP (PC1), after rounding, the BSH
scores used in ATC-21 were:
LOW 3. 5 MODERATE 3. 5 HIGH 2.0
To illustrate the effect of the MC on BSH score, in
the following example the BSH score will be re-calculated
for a TILT-UP building in a MODERATE Seismicity zone. From
Table XXIV, the HEHRP Map Areas corresponding to MODERATE
are Areas 3 and 4. From Table XXIII, the MC values for
PCl for Map Areas 3 and 4 are 1.2 and 1.5, respectively;
the average MC is 1.35. BEHA is calculated from Eq. 4.8:
BEHA = 1.35 w (10.60) = 14.30
The rest of the calculations are adjusted as follows:
a = In(14.30) = 2.6602595
The standard normal variate, z, is:
z = [In(60) - 2.6602595] / 0.4619291 = 3.1045566
From the polynomial approximation, the probability of the
DF exceeding 60-percent, Q(z), is:
Q(z) = .000952892
The Basic Structural Hazard score is:
BSH = -log10(.000952892) = 3.02
As expected, the increase in damage from the MC results in
a reduced BSH score (now 3.02, reduced from 4.05 using MB).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
99
In computing the BSH score for a different PGA, the
ATC-21 procedure described above re-starts with new ML,
MB, and MH values corresponding to that level of ground
motion. Additionally, values of the MC consistent with the
HEHRP Map Areas corresponding to the PGA being considered
are used. In the preceding example of TILT-UP, the BSH
score for MODERATE seismicity used a MC = 1. 35. In
computing the BSH score for TILT-UP for a LOW seismicity
Map Area, a MC = 2.0 would be used; for HIGH seismicity,
the average for Map Areas 5, 6 again results in MC = 1. 35.
The MC is used differently in developing fragility
curves, described in the next section.
Performance Modification Factors
The BSH score was not created to stand alone, but
was designed to be modifed for specific buildings. A
number of characteristics of design, construction, and
site conditions can modify the seismic performance of a
specific building, causing the damage to differ from
"average."
Performance Modification Factors (PMFs) were
developed by the ATC-21 experts to be added to the BSH
score, so the resulting Structural Score (S) would
approximately reflect the probability of major damage,
given the presence of that factor. Reversing Eq. 4. 3, it
can be shown that for the Structural Score s, the
99
In computing the BSH score for a different PGA, the
ATC-21 procedure described above re-starts with new ML,
MB, and MH values corresponding to that level of ground
motion. Additionally, values of the MC consistent with the
NEHRP Map Areas corresponding to the PGA being considered
are used. In the preceding example of TILT-UP, the BSH
score for MODERATE seismicity used a MC = 1.35. In
computing the BSH score for TILT-UP for a LOW seismicity
Map Area, a MC = 2.0 would be usedj for HIGH seismicity,
the average for Map Areas 5, 6 again results in MC = 1.35.
The MC is used differently in developing fragility
curves, described in the next section.
Performance Modification Factors
The BSH score was not created to stand alone, but
was designed to be modifed for specific buildings. A
number of characteristics of design, construction, and
site conditions can modify the seismic performance of a
specific building, causing the damage to differ from
"average. "
Performance Modification Factors (PMFs) were
developed by the ATC-21 experts to be added to the BSH
score, so the resulting Structural Score (S) would
approximately reflect the probability of major damage,
given the presence of that factor. Reversing Eq. 4.3, it
can be shown that for the Structural Score S, the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
probability of the DF exceeding 60-percent is:
[Pr(DF>:60/.)] : 10-S (4. 9)
Continuing with our illustration using TILT-UP,
suppose that at PGA = . 22, a building in a MODERATE
Seismicity zone (for which we calculated a probability of
major damage of .000952892, and a BSH score of 3.02) had
the performance modifier of "Torsion. " The PMF for
torsion is -1.0; Eq. 4.9 gives us the Structural Score:
s = 3.02- 1.0 = 2.02
Applying Eq. 4.8 provides the revised probability of major
damage:
[Pr(DF>=60/.)] : 1o-(2.02) : .0095499
This 10-fold increase in the liKlihood of major
damage is very significant. Torsion is one of the most
common performance modifiers, hence this example
illustrates that many buildings will differ significantly
in their seismic performance from the "average" reflected
by the BSH score.
DEVELOPMEHT OF A TECHHIQUE TO DERIVE DAMAGE FACTOR FROM STRUCUTRAL SCORE
Given that individual characteristics of design and
construction for specific buildings may cause significant
differences in seismic damage compared to "average"
buildings, it is clear that regional damage estimates
would be improved if they reflected the characteristics of
100
probability of the DF exceeding 60-percent is:
[Pr(DF>=60t.») = 10-S (4.9)
Continuing with our illustration using TILT-UP,
suppose that at PGA = . 22, a building in a MODERATE
Seismicity zone (for which we calculated a probability of
major damage of .000952892, and a BSH score of 3.02) had
the performance modifier of "Torsion." The PMF for
torsion is -1.0j Eq. 4.9 gives us the structural Score:
s = 3.02 - 1.0 = 2.02
Applying Eq. 4.8 provides the revised probability of major
damage:
[Pr(DF>=60Y.») = 10-(2.02) = .0095499
This 10-fold increase in the liKlihood of major
damage is very significant. Torsion is one of the most
common performance modifiers, hence this example
illustrates that many buildings will differ significantly
in their seismic performance from the "average" reflected
by the BSH score.
DEVELOPMENT OF A TECHNIQUE TO DERIVE DAMAGE FACTOR FROM STRUCUTRAL SCORE
Given that individual characteristics of design and
construction for specific buildings may cause significant
differences in seismic damage compared to "average"
buildings, it is clear that regional damage estimates
would be improved if they reflected the characteristics of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
specific buildings. However, although ATC-21 presents a
methodology for inventorying and scoring buildings based
on specific characteristics, it does not provide a method
to modify the DF itself. Clearly, if the probability of
major damage (DF = 60) increases, the value of the median
damage should also increase.
Development of modified damage estimates reflecting
the presence of Performance Modifiers is one of the
primary objectives of the present research.
The Mean Damage Factor (MDF) is the DF with a
50-percent probability of exceedance, the DF most likely
to be observed. The ln(MDF) is, in general, the mean a of
the normally distributed ln(DF). The MDF = BEHA only for
"average" buildings, wherein E(PMF) = 0 and s = BSH. When
s differs from BSH (due to the presence of PMFs), then the
MDF differs from BEHA, and a differs from ln(BEHA). The
MDF is related to s by the standard deviation of the
normally distributed ln(DF), ~. which is determined from
the ML and MH estimates of the DF, by Eq. 4.6.
For any value of Structural Score S, the probability
that the DF exceeds 60-percent, Q(z), can be determined
from:
G(z) = 10-S (4. 9a)
The polynomial Eq. 4.7 was "solved" in reverse (the
standard normal variate z was determined for specified
101
specific buildings. However, although ATC-21 presents a
methodology for inventorying and scoring buildings based
on specific characteristics, it does not provide a method
to modify the DF itself. Clearly, if the probability of
major damage (DF : 60) increases, the value of the median
damage should also increase.
Development of modified damage estimates reflecting
the presence of Performance Modifiers is one of the
primary objectives of the present research.
The Mean Damage Factor (MDF) is the DF with a
50-percent probability of exceedance, the DF most likely
to be observed. The In(MDF) is, in general, the mean a of
the normally distributed In(DF). The MDF : BEMA only for
"average" buildings, wherein E(PMF) : 0 and S : BSH. When
S differs from BSH (due to the presence of PMFs), then the
MDF differs from BEMA, and a differs from In(BEHA). The
MDF is related to S by the standard deviation of the
normally distributed In(DF), ~, which is determined from
the ML and MH estimates of the DF, by Eq. 4.6.
For any value of Structural Score S, the probability
that the DF exceeds 60-percent, G(z), can be determined
from:
G(z) : 10-S (4. 9a)
The polynomial Eq. 4.7 was "solved" in reverse (the
standard normal variate z was determined for specified
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
values of Q(z)) with a BASIC computer program which solved
the equation in small increments of z, and tabulated z vs.
Q(z). Part of this solution is presented in Table XXV,
following.
TABLE XXV
THE STANDARD NORMAL VARIATE VS. STRUCTURAL SCORE
s Q (Z) z
0. 30 . 50 0 1 . 10 1. 281 2 . 01 2. 326 3 . 001 3.090 4 . 0001 3. 719 5 . 00001 4.265 6 . 000001 4. 753 7 . 0000001 5. 199 8 . 00000001 5. 612
With the standard normal variate z determined from
Table XXV, and the standard deviation a of the normal
distribution of ln(DF), the Mean Damage Factor (HDF)
corresponding to any Structural Score (S} can be
determined with Eqs. 4. 10 and 4. 10a, as shown in Fig. 10.
ln(HDF) = a = ln(60} - z•a ( 4. 10)
HDF = exp (a) = exp (ln(60) - z•a> (4. 10a)
Hote that in Eqs. 4. 10 that a is, in general, not the
ln(BEHA).
102
values of Q(z)) with a BASIC computer program which solved
the equation in small increments of z, and tabulated z vs.
Q(z). Part of this solution is presented in Table XXV,
following.
TABLE XXV
THE STANDARD HORMAL VARIATE VS. STRUCTURAL SCORE
S Q (z) z
o. 30 .50 0 1 .10 1. 281 2 .01 2. 326 3 .001 3.090 4 .0001 3.719 5 .00001 4.265 6 .000001 4. 753 7 .0000001 5.199 8 .00000001 5. 612
With the standard normal variate z determined from
Table XXV, and the standard deviation a of the normal
distribution of In(DF), the Mean Damage Factor (HDF)
corresponding to any Structural Score (8) can be
determined with Eqs. 4.10 and 4. 10a, as shown in Fig. 10.
In(HDF) = a = In(60) - z*a (4. 10)
HDF = exp (a) = exp (In(60) - z*a) (4. 10a)
Hote that in Eqs. 4. 10 that a is, in general, not the
In(BEHA) .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
o( ln(60) 1.........-,8Z ~ ~I _..,...-1
103
Figure 10. Determining the Mean Damage Factor (MDF) from the Structural Score (S) and the normally distributed ln(DF).
Hext, a continuous relationship between MDF and S
was determined by regressing MDF as determined from Eqs.
4. 10, corresponding to values of of S = 0. 3, 1, 2, 3, and
4. After experimenting with several forms of regression, a
good fit was found by using a Power Function, of the form:
MDF = bo • sbi (4. 11)
where bo and b1 are the regression coefficients.
Returning to the example of the TILT-UP building to
illustrate this methodology, Table XXV is utilized to
calculate discrete values of S vs. MDF, with a= .461929
as previously determined. The results of this step are
shown in Table XXVI, following:
103
0( In(60)
~,&Z ~
Figure 10. Determining the Mean Damage Factor (HDF) from the Structural Score (S) and the normally distributed In(DF).
Next, a continuous relationship between HDF and S
was determined by regressing HDF as determined from Eqs.
4. 10, corresponding to values of of S = 0.3, 1, 2, 3, and
4. After experimenting with several forms of regression, a
good fit was found by using a Power Function, of the form:
(4. 11)
where bO and b1 are the regression coefficients.
Returning to the example of the TILT-UP building to
illustrate this methodology, Table XXV is utilized to
calculate discrete values of S vs. HDF, with a = .461929
as previously determined. The results of this step are
shown in Table XXVI, following:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
s
TABLE XXVI
STRUCTURAL SCORE VS. MEAN DAMAGE FACTOR FOR TILT-UP BUILDINGS (PC1) AT PGA : 0. 22
Q ( Z) z ex
o. 30 . 50 0 4.09434 1 . 10 1. 281 3. 50261 2 . 01 2. 326 3.01990 3 . 001 3.090 2.66698 4 . 0001 3. 719 2. 37643
104
MDF
60.00 33. 20 20.49 14.40 10. 77
As expected, as the Score (S) increases, the mean DF
decreases; this is consistent with the design of S as a
"seismic performance" score.
Next, the five values of MDF were regressed against
S from Table XXVI. The resulting Power Function
coefficients, which produce a coefficient of correlation
of 0. 986, are:
bo = 29.66478 b1 : -. 652952
This procedure was followed for a total of five
levels of ground acceleration: PGA = .05, . 10, .22, .47,
and 1.02 (corresponding to MMI from VI to X). This
resulted in 5 regressed curves of MDF on S, as shown in
Fig. 11 to 15, following,
This procedure was repeated for each of the other
twelve ATC-21 Structure Types, resulting in a total of 60
curves.
S
TABLE XXVI
STRUCTURAL SCORE VS. MEAN DAMAGE FACTOR FOR TILT-UP BUILDINGS (PC1) AT PGA : 0.22
Q (z) z ex
O. 30 .50 0 4.09434 1 . 10 1. 281 3. 50261 2 .01 2. 326 3.01990 3 .001 3.090 2.66698 4 .0001 3.719 2. 37643
104
MDF
60.00 33. 20 20.49 14.40 10. 77
As expected, as the Score (S) increases, the mean DF
decreases; this is consistent with the design of S as a
"seismic performance" score.
Next, the five values of MDF were regressed against
S from Table XXVI. The resulting Power Function
coefficients, which produce a coefficient of correlation
of O. 986, are:
bO : 29.66478 b1 : -.652952
This procedure was followed for a total of five
levels of ground acceleration: PGA: .05, .10, .22, .47,
and 1.02 (corresponding to MHI from VI to X). This
resulted in 5 regressed curves of MDF on S, as shown in
Fig. 11 to 15, following.
This procedure was repeated for each of the other
twelve ATC-21 Structure Types, resulting in a total of 60
curves.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0::
~ (.)
i£ w (!) <( ~ <( 0 z <( w ~
100
80
60
40
20
0
0
\ \·
PC1 - TILT UP PGA = 0.05 BSH = 5.98
b(O =20.101 24 hf1 =-1 01~ FiR
R-!: 1:QR no
~ ~ ...._
2 4 6 STRUCTURAL SCORE
105
8
Figure 11. Mean Damage Factor vs. Structural Score for TILT UP (PC1) buildings, for PGA = .05.
0::: ~ u it w (!) <C ~ <C o Z <C w ~
100
80
60
40
20
o
o
\ \.
PC1 - TILT UP PGA = 0.05 BSH = 5.98
b(O =20.101 24 ht1 =.1 n1~ ~R
R.!l:1 ,=QR no
~ ~ ~
246 STRUCTURAL SCORE
105
8
Figure 11. Mean Damage Factor vs. Structural Score for TILT UP (PCt) buildings, for PGA = .05.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0:: 0 1-0 <( u. w (!) <( ::2 <( c z <( w ::2
100
80
60
40 \ \
20 ~
0
0
PC1 - TILT UP PGA = 0.10 BSH = 5.7
b(O =27.55 86 b(1 --0.721 51
R-s' !:1=98.0°
~ ~
2 4 6 STRUCTURAL SCORE
106
i I 8
Figure 12. Mean Damage Factor vs. Structural Score for TILT UP (PC1) buildings, for PGA : . 10.
0:: 0 l-t) <t: u. w (!) <t: ~ <t: C Z <t: W ~
100
80
60
40 \ \
20 ~
o
o
PC1 - TILT UP PGA = 0.10 BSH = 5.7
b(O =27.55 86 ~1 -..QJ21 51
~, !l::;98~OO
,. ~ r--
246 STRUCTURAL SCORE
106
1 I 8
Figure 12. Mean Damage Factor vs. Structural Score for TILT UP (pe1) buildings, for PGA : .10.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0:: 0 1-(..) < u.. w (!) < ~ < 0 z < w ~
100
80
60
\
40 \ \
20
0
0
PC1 - TILT UP PGA = 0.22 BSH = 3.02
~
b(O =29.66« 77 b(1 -.n R~? ~~
R-s :QB_QO
~ ~ -
2 4 6 STRUCTURAL SCORE
107
8
Figure 13. Mean Damage Factor vs. Structural Score for TILT UP (PC1) buildings, for PGA : . 22.
0:: 0 .... u < u.. w (!) < :2E < 0 Z < w :2E
100
80
60
40 \ \
20
o
o
PC1 - TILT UP PGA = 0.22 BSH = 3.02
~
b(O =29.664 77 b(1 -.n R~' ~~
R·s tJ=9ROO
~ r---;--
246 STRUCTURAL SCORE
107
8
Figure 13. Mean Damage Factor vs. Structural Score for TILT UP (PC1) buildings, for PGA = .22.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0:: 0 1-(.)
it w C> <( ~ <( 0 z <( w ~
100
80
60
\
40 \ ~
1\
20
0
0
PC1 - TILT UP PGA = 0.47 BSH = 1.97
b(O =33.63E 10 b(f -.n ~~A l:i.4
R-~' ~=QROO
~. r---..._ r----
2 4 6
STRUCTURAL SCORE
108
8
Figure 14. Mean Damage Factor vs. Structural Score for TILT UP (PC1) buildings, for PGA: .47.
0:::
~ u it w (!) « ~ « 0 z « w ~
100
80
60
1 40 \
'" I\. 20
o
o
PC1 - TILT UP PGA = 0.47 BSH = 1.97
b(O =33.631 10 b(1 -on ~~R ~
R-s 1=98_0°
"". f'..... ...,~
j ~
246
STRUCTURAL SCORE
108
8
Figure 14. Mean Damage Factor vs. Structural Score for TILT UP (pel) buildings, for PGA : .47.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0::: 0 1-u LE w (!) <( ~ <( 0 z <( w ~
100
80
60 \
40 \ ~
\.
20
0
0
PC1 - TILT UP PGA = 1.02 BSH = 0.95
b(O =34.75 06 b(1 --n ~nfl bQ
R-SI !1=98.0°
~ ~ 1--.....__
2 4 6
STRUCTURAL SCORE
109
8
Figure 15. Mean Damage Factor vs. Structural Score for TILT UP (PC1) buildings, for PGA = 1.02.
0::: 0 l-t)
LE w (!) « ~ « 0 Z « w ~
100
80
60
\ 40 \
~ \.
20
o
o
PC1 - TILT UP PGA = 1.02 BSH = 0.95
b(O =34.75 06 b(1 -.0.508 t2Q
R.!':I IoJ=QROo
~ :::--....... ,---~ 246 STRUCTURAL SCORE
109
8
Figure 15. Mean Damage Factor vs. Structural Score for TILT UP (PCl) buildings, for PGA = 1.02.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEVELOPMENT OF FRAGILITY CURVES BASED ON PERFORMANCE MODIFIERS
Modified Basic Structural Hazard Score
110
The Basic Structural Hazard (BSH) score is shown at
the top of each of the HDF vs. S curves in Figs. 11 to 15.
The BSH for each Structure Type and each PGA was
determined as described previously, with ~ derived from HL
and HH, a derived from BENA, where BENA = HB * HC.
The BSH score has been modified from ATC-21 in two
ways. First, ATC-21 converts PGA to EPA, then regresses ~
against EPA, and uses this "smoothed" version of ~ to
calculate the BSH at even increments of EPA. In the
present research the BSH is computed for discrete values
of PGA only, without smoothing.
Second, the present study uses the Modification
Constant (HC) in a different way than in ATC-21. In
ATC-21, the HC's "step up" along with the EPA values; each
HEHRP Hap Area is assigned both an EPA and HC value, and
these are kept consistent so the resulting BSH will
reflect the expected seismicity and the construction
standard of the Hap Area. However, the objective of the
present research is to develop fragility curves which
express damage across the entire range of possible PGAs,
for a single Hap Area. Therefore, as the PGA takes five
steps up, from .05 to 1.02, a single HC value is used, to
DEVELOPMENT OF FRAGILITY CURVES BASED ON PERFORMANCE MODIFIERS
Modified Basic Structural Hazard Score
110
The Basic Structural Hazard (BSH) score is shown at
the top of each of the MDF vs. S curves in Figs. 11 to 15.
The BSH for each Structure Type and each PGA was
determined as described previously. with ~ derived from HL
and HR. a derived from BENA. where BENA = HB * MC.
The BSH score has been modified from ATC-21 in two
ways. First. ATC-21 converts PGA to EPA. then regresses ~
against EPA. and uses this "smoothed" version of ~ to
calculate the BSH at even increments of EPA. In the
present research the BSH is computed for discrete values
of PGA only. without smoothing.
Second. the present study uses the Modification
Constant (MC) in a different way than in ATC-21. In
ATC-21. the MC's "step up" along with the EPA values; each
HEHRP Map Area is assigned both an EPA and MC value. and
these are kept consistent so the resulting BSH will
reflect the expected seismicity and the construction
standard of the Map Area. However. the objective of the
present research is to develop fragility curves which
express damage across the entire range of possible PGAs.
for a single Map Area. Therefore. as the PGA takes five
steps uP. from .05 to 1.02. a single MC value is used. to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
111
reflect the design and construction standards of one area.
All the BSH values developed herein use the average values
of MC for NEHRP Map Areas 3 and 4, for use in Portland.
Mean Damage Factor vs. ECPMF)
The BSH score, modified as described above,
represents an "average" building for its Structure Type.
That is, when there are no characteristics of design,
construction, nor site conditions that would alter the
seismic performance of the building from "average, " there
are no Performance Modifiersj E(PMF) = 0, and S = BSH.
The value of MDF corresponding to this average building
can be found from the MDF vs. S curves, using S = BSH.
For the TILT-UP (PGA = . 22) illustration, the
coordinates on the MDF vs. S curve (Fig. 13) for
E(PMF) = 0 are (3.02, 14.42).
When Performance Modifiers are present, the value of
E(PMF) moves these coordinates along the curve. For
example, if Torsion is present as the only Performance
Modifier, then E(PMF) = -1, and:
S = BSH + E(PMF) = 3.02- 1 = 2.02
From the power formula regression equation, the MDF
for S = 2. 02 is found to be 18. 74, and the coordinates
corresponding to E(PMF) = -1 are (2.02, 18. 74). This is
illustrated on Fig. 16, following.
111
reflect the design and construction standards of one area.
All the BSH values developed herein use the average values
of Me for NEHRP Map Areas 3 and 4, for use in Portland.
Mean Damage Factor vs. E(PMF)
The BSH score, modified as described above,
represents an "average" building for its Structure Type.
That is, when there are no characteristics of design,
construction, nor site conditions that would alter the
seismic performance of the building from "average," there
are no Performance Modifiersj E(PMF) = 0, and S = BSH.
The value of MDF corresponding to this average building
can be found from the MDF vs. S curves, using S = BSH.
For the TILT-UP (PGA = .22) illustration, the
coordinates on the MDF vs. S curve (Fig. 13) for
E(PMF) = 0 are (3.02,14.42).
When Performance Modifiers are present, the value of
E(PMF) moves these coordinates along the curve. For
example, if Torsion is present as the only Performance
Modifier, then E(PMF) = -1, and:
S = BSH + E(PMF) = 3.02 - 1 = 2.02
From the power formula regression equation, the MDF
for S = 2. 02 is found to be 18. 74, and the coordinates
corresponding to E(PMF) = -1 are (2.02,18.74). This is
illustrated on Fig. 16, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
u. 0 ::iE ~ 0 1-(.) <( u. w (!) <( ::iE <( 0 z <( w ::iE
60
50
40
30
20
10
0
0 1
(2.02, 18.74) FOR SUM(PMF) = -1 I (3.02, 14.42) FOR SUM(PMF) = 0
i 2 3 4 5 6 STRUCTURALSCORE,S
7
112
8
Figure 16. Mean Damage Factor vs. Structural Score for TILT-UP at PGA = .22, showing coordinates for E(PMF) : 0 and E(PMF) = -1.
From the foregoing, it is clear that the BSH score
becomes the "starting point" on the MDF vs. S curve. The
value of E(PMF) "moves" the building along the curve,
increasing damage for negative and decreasing damage for
positive values of E(PMF).
Therefore, the MDF may be calculated for any value of
E(PMF). For the TILT-UP example, this has been done for
E(PMF) = -2 to +2, and is tabulated in Table XXVII,
following.
u. 0 ~
Ei 0 l-t) « u. w (!) « ~ « 0 z « w ~
60 112
50
40
30
20
(2.02,18.74) FOR SUM(PMF) =-1 / (3.02. 14.42) FOR SUM(PMF) = 0
10
0 j
0 1 2 345 6 7 8 STRUCTURALSCORE,S
Figure 16. Mean Damage Factor vs. Structural Score for TILT-UP at PGA = .22, showing coordinates for E(PHF) = 0 and E(PHF) = -1.
From the foregoing, it is clear that the BSH score
becomes the "starting pOint" on the HDF vs. S curve. The
value of E(PHF) "moves" the building along the curve,
increasing damage for negative and decreasing damage for
positive values of E(PHF).
Therefore, the HDF may be calculated for any value of
E(PHF). For the TILT-UP example, this has been done for
E(PHF) = -2 to +2, and is tabulated in Table XXVII,
following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XXVII
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA: .22 WITH MC = 1. 35
E(PMF) s MDF
+2 5.02 10. 34 +1 4.02 11.96 0 3.02 14.42
-1 2.02 18. 74 -2 1. 02 29. 28 -3 < 0 > 60
Fragility Curves based on ECPMF)
To create a family of fragility curves for each
building type, the preceding steps were taken for five
113
intensities (PGA = .05, . 10, . 22, .47, and 1.02), creating
five Tables of E(PMF) vs. MDF, similar to Table XXVII. For
TILT-UP, these are shown in Tables XXVIII to XXXI.
TABLE XXVIII
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA = .05 WITH MC: 1. 35
E(PMF) s MDF
+2 7.98 2.45 +1 6.98 2.81
0 5.98 3. 28 -1 4.98 3.95 -2 3.98 4.96 -3 2.98 6. 65
TABLE XXVII
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA = .22 WITH MC = 1.35
E(PMF) S MDF
+2 5.02 10. 34 +1 4.02 11.96
0 3.02 14.42 -1 2.02 18. 74 -2 1. 02 29. 28 -3 < 0 > 60
Fragility Curves based on E(PMF)
To create a family of fragility curves for each
building type, the preceding steps were taken for five
113
intensities (PGA = .05, .10, .22, .47, and 1.02), creating
five Tables of E(PMF) vs. MDF, similar to Table XXVII. For
TILT-UP, these are shown in Tables XXVIII to XXXI.
TABLE XXVIII
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA = .05 WITH Me = 1.35
E(PMF) S MDF
+2 7.98 2.45 +1 6.98 2.81
0 5.98 3. 28 -1 4.98 3.95 -2 3.98 4.96 -3 2.98 6. 65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XXIX
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA : . 10 WITH MC : 1. 35
E(PMF) s MDF
+2 7.70 6. 32 +1 6.70 6. 98
0 5.70 7.85 -1 4.70 9.02 -2 3.70 10.72 -3 2.70 13.46
TABLE XXX
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA: .47 WITH MC: 1. 35
E(PMF) s MDF
+2 3. 97 16.05 +1 2.97 18.76
0 1. 97 23. 38 -1 o. 97 34. 19 -2 < 0 > 60 -3 < 0 > 60
114
TABLE XXIX
E(PHF) VS. HEAN DAMAGE FACTOR FOR TILT-UP AT PGA : . 10 WITH HC : 1.35
E(PHF) S HDF
+2 7.70 6. 32 +1 6.70 6. 98
0 5.70 7.85 -1 4.70 9.02 -2 3.70 10.72 -3 2.70 13.46
TABLE XXX
E(PHF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA : .47 WITH HC : 1.35
E(PHF) S HDF
+2 3. 97 16.05 +1 2.97 18.76
0 1. 97 23. 38 -1 O. 97 34. 19 -2 < 0 > 60 -3 < 0 > 60
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XXXI
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA = 1.02 WITH MC = 1. 35
E(PMF) s MDF
+2 2.95 20. 10 +1 1. 95 24. 78
0 0.95 35. 67 -1 < 0 > 60 -2 < 0 > 60 -3 < 0 > 60
115
The next step in the methodology was to "dismantle"
these five tables of E(PMF) vs. MDF (one for each PGA
intensity) and "reassemble" them into six tables of MDF vs.
PGA (one for each integer value of E(PMF): -3, -2, -1, 0,
+1, +2). For example, the values of MDF vs PGA for E(PMF)
= 0 for TILT UP are shown in Table XXXII, following (five
more tables, for other E(PMF) values, were also developed).
TABLE XXXII
PGA VS. MEAN DAMAGE FACTOR FOR TILT-UP AT E(PMF) = 0 WITH MC = 1. 35
PGA MDF
. 05 3. 28
. 10 7. 85
. 22 14.42
. 47 23. 38 1. 02 35. 67
TABLE XXXI
E(PMF) VS. MEAN DAMAGE FACTOR FOR TILT-UP AT PGA = 1.02 WITH MC = 1.35
E(PMF) S MDF
+2 2.95 20. 10 +1 1. 95 24. 78
0 0.95 35. 67 -1 < 0 > 60 -2 < 0 > 60 -3 < 0 > 60
115
The next step in the methodology was to "dismantle"
these five tables of E(PMF) vs. MDF (one for each PGA
intensity) and "reassemble" them into six tables of MDF vs.
PGA (one for each integer value of E(PMF): -3, -2, -1, 0,
+1, +2). For example, the values of MDF vs PGA for E(PMF)
= 0 for TILT UP are shown in Table XXXII, following (five
more tables, for other E(PMF) values, were also developed).
TABLE XXXII
PGA VS. MEAN DAMAGE FACTOR FOR TILT-UP AT E(PMF) = 0 WITH MC = 1.35
PGA MDF
.05 3.28
.10 7. 85
.22 14.42
. 47 23. 38 1. 02 35.67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
116
Each of the six tables of PGA vs. MDF constitutes a
"damage function" for an integer value of E(PMF). To
create a fragility curve, a regression of HDF against PGA
was made for each damage function. The best fits were
attained with a third-degree polynomial, as follows:
MDF = bo + b 1 (PGA) + b2 (PGA) 2 + b 3 (PGA) 3 ( 4. 12)
where b0 , b 1, b2, and b3 are the regression coefficients.
The resulting fragility curve for E(PMF) = 0 for TILT-UP,
along with the polynomial coefficients, is shown in Fig.
17, following.
Similar curves were established for E(PMF) = +2, +1,
-1, -2, and -3, resulting in a family of fragility curves
based on E(PMF) for the TILT UP (PC1) Structure Type, as
shown in Fig. 18.
This methodology was repeated for the other eleven
building types of ATC-21. The resulting fragility curve
families are presented in Figs. 19 to 29.
To provide more resolution of MDF, another set of
"half-step" fragility curves was developed, for E(PMF) =
+1. 5, +. 5, -. 5, -1. 5, and -2. 5. These families of fragility
curves are presented in the Appendix.
In developing these curves from the ATC-13 expert
opinion damage estimates, where several ATC-13 facility
116
Each of the six tables of PGA vs. MDF constitutes a
"damage function" for an integer value of E(PMF). To
create a fragility curve, a regression of HDF against PGA
was made for each damage function. The best fits were
attained with a third-degree polynomial, as follows:
MDF = bO + bl(PGA) + b2(PGA)2 + b3(PGA)3 (4.12)
where bO' bl' b2' and b3 are the regression coefficients.
The resulting fragility curve for E(PMF) = 0 for TILT-UP,
along with the polynomial coefficients, is shown in Fig.
17, following.
Similar curves were established for E(PMF) = +2, +1,
-1, -2, and -3, resulting in a family of fragility curves
based on E(PMF) for the TILT UP (PC1) Structure Type, as
shown in Fig. 18.
This methodology was repeated for the other eleven
building types of ATC-21. The resulting fragility curve
families are presented in Figs. 19 to 29.
To provide more resolution of HOF, another set of
"half-step" fragility curves was developed, for E(PMF) =
+1. 5, +.5, -.5, -1. 5, and -2.5. These families of fragility
curves are presented in the Appendix.
In developing these curves from the ATC-13 expert
opinion damage estimates, where several ATC-13 facility
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
~ 0 1-() <( u. w (!) <( ~ <( 0 z <( w ~
PC1 - TILT UP PMF = 0
100 I b(C )=-0.2 218 b£1 1\=86.~ ~57 b(2 )=-99. ~8
80 ht:= 1\:.dR 1.1
R- tn=QQ I
60 I
40
--20
~ ~
l---r !---__.
v v L
/ v
0
0.0 0.2 0.4 0.6 0.8 PGAON ROCK
Figure 17. PC1 - TILT UP fragility curve for E (PMF) = 0.
A%
!---'" 1-e
1.0
117
a:: 0 I-U <C u. w (!) <C ::2! <C 0 Z <C W ::2!
PC1 - TILT UP PMF = 0
100 I b(C )=-0.2 bf1 \=86_
b(2 )=-99.
80 bf~ 1\=48_~
R-! ~a=99. I
60 I
40
--20
I--~ L.---~ ---
V v
/
/ r'"
o
0.0 0.2 0.4 0.6 0.8 PGAON ROCK
Figure 17. PC1 - TILT UP fragility curve for E (PHF) = O.
117
'18 !57
38 14
9%
l---'" I-e
1.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
118
classes correspond to one ATC-21 building type, the values
of ML, MB, and MH were averaged at the beginning of the
methodology. Where an ATC-13 facility class designated HR
(high-rise) was given, this was not included in the damage
averages, because ATC-21 includes a PMF for high-rise.
Application of Fragility Curves to NEHRP Areas 3,4
The fragility curve families presented in Figures 18
to 29 reflect a "non-California" standard of design and
construction found in NEHRP Map Areas 3 and 4 (by use of
the Modification Constant, as discussed on pp. 110-111).
In applying these curves, it is important to note
that this reference to NEHRP Areas refers to the quality
of seismic resistance of the buildings, and not to the
level of seismicity.
Therefore, a different set of fragility curve
families would be required for NEHRP Map Areas 1, 2, 5, 6,
or 7, to reflect the Modification Constant (MC)
appropriate for those areas.
118
classes correspond to one ATC-21 building type, the values
of ML, ME, and MH were averaged at the beginning of the
methodology. Where an ATC-13 facility class designated HR
(high-rise) was given, this was not included in the damage
averages, because ATC-21 includes a PMF for high-rise.
Application of Fragility Curves to NEHRP Areas 3,4
The fragility curve families presented in Figures 18
to 29 reflect a "non-California" standard of design and
construction found in NEHRP Map Areas 3 and 4 (by use of
the Modification Constant, as discussed on pp. 110-111).
In applying these curves, it is important to note
that this reference to NEHRP Areas refers to the quality
of seismic resistance of the buildings, and not to the
level of seismicity.
Therefore, a different set of fragility curve
families would be required for NEHRP Map Areas 1, 2, 5, 6,
or 7, to reflect the Modification Constant (MC)
appropriate for those areas.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
PC1 - TILT UP NEHRP AREAS 3,4
-3 -2 100 -l+===t=~~:;::: ==~~==t===t====+===+==+.=l-1
I / I i_v \
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
119
Figure 18. PC1 - TILT UP fragility curve for integer E (PMF) steps.
100
80 c::: 0 l-t) « 60 I.L W C> « ~ « 40 0 Z « w ~
20
o
I I
0.0
I I
J
PC1 - TILT UP NEHRP AREAS 3,4
-3 -2 :
11 1 I
I II
J
I / I ;/ !
Y V )I L
/ / v
// v V ~
/' v-v--)~ ~ k~ ~~
~ ~ I
, I VI I / I
I /1 I 1/' i
I v I /
V I
..--~~ ~
1----1-
I I
0.2 0.4 0.6 0.8 1.0 PGAON ROCK
119
-1
o
+1 +2
Figure 18. pe1 - TILT UP fragility curve for integer E (PMF) steps.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.0
WOOD - WOOD FRAME NEHRP AREAS 3,4
0.2 0.4 0.6 0.8 PGAON ROCK
Figure 19. W - WOOD fragility curves for integer E (PMF) steps.
120
WOOD - WOOD FRAME NEHRP AREAS 3,4
120
-5 -4 -3 1 00 ~I+==-=+=I ~,==+==+=, ==;:==+i ==4'=1 ====+!==::J:
1 ===P
V '1-2
I I I I I I
0.0 0.2 0.4 0.6 0.8 PGAON ROCK
Figure 19. W - WOOD fragility curves for integer E (PMF') steps.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
80 0::: 0 I-(.) <( 60 u.. w (!) <( ::?! <(
40 Cl z <( w ::?!
20
0
S1 - STEEL MRF NEHRP AREAS 3,4
-6 '
-5
I I i
I I 1/ I
i
I /' I I I
I I /I /I I 1/ I I
I
I /I I I I I I v !
I / v I /
I / v v / / ~
v ~
/ _.........~
I / ---- _.., -~ --t ~ / ~ ~ ~ ---= ---= ~ -~ -:..--::
~ ~ ~ -~ -
I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 20. S1 - STEEL MRF fragility curves for integer E(PMF) steps.
I
121
-4
-3
-2
-1 0 +1
+2
100
80 c:: 0 .... () « 60 u.. w C9 « ~ « 40 0 z « w ~
20
o
S1 - STEEL MRF NEHRP AREAS 3,4
-6 , -5 !
I I i I I / Vi
I i J
I /1 I / I
I I I /1 II I 1/ / I
I
I I I I v V I !/ V
I V / / /
/ / V V V
V ~ ,.../ / ~
~
/ / L.-- ~ v--- - ..... ----l V: v
~ ~ ~ --~ f..-- - -~ ~
~ ~ ;:;-~~
~ p-
I I 0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
Figure 20. S1 - STEEL MRF fragility curves for integer E(PHF) steps.
, I
121
-4
-3
-2
-1 o
+1 +2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
80 0:: 0 I-(.) <( 60 11.. w (!) <( :2 <(
40 0 z <( w :2
20
0
S2 - BRACED STEEL FRAME NEHRP AREAS 3,4
-5 -4 '
lL I I 1 I
I I
I i
I
j_l j
I I Ll 1/ I
1/ I
I I I I v v I ..1.
II v v v L
I
I
I J
I !
v I I
I ~ v v v v r--
/ ,.....,.... .,. I v: 1.---~ -+---r- 1----1 ~
~ .----~ ~ ~ -
I I I 0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
-3
-2
-1 0
+1 +2
122
Figure 21. S2 - BR STL FRAME fragility curves for integer E(PMF) steps.
100
80 0::: 0 l-t) « 60 LL. W C> « :E « 40 0 Z « w :E
20
o
S2 - BRACED STEEL FRAME NEHRP AREAS 3,4
-5 -4 , I
II I I 1 I I
I I i I If
\ 1 I II I
/ ! Lt I !
II I il I I
I / v / I
/ I v V / II v v /V
I ~ v
V v
V I-'
./ ~ ...... / ~
~ -f----- I----i ~ I---~ ~ ~ !-::::::::
~ ~ :....:::::::;
~
I I I 0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
-3
-2
-1 o +1 +2
122
Figure 21. S2 - BR STL FRAME fragility curves for integer E(PHF) steps.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
80 0::: 0 1-(.) <( 60 LL w (!) <( ~ <(
40 Q
z <( w ~
20
0
0.0
I
S3 - LIGHT METAL NEHRP AREAS 3,4
-8 -7 -6 -5
I I I I I I I I ,
' I I I I 1/ I
I I
'
1
' /I /1 }'
/ I I I I I
I v I I I /; I /
v / ./
i/ ~ / v v ..---~~ ,i~~ ---== f ~~~
I I I I
-4
I I
1/
I I
I I v /I
k.....-1--- e:-: ~
0.2 0.4 0.6 0.8 PGAON ROCK
! I I I
v 1/ I
~ ...
1.-----~
~
~
1.0
123
-3
-2
-1 0 !~
Figure 22. S3 - LIGHT METAL fragility curves for integer E(PMF) steps.
100
80 0:: 0 I-0 <C 60 LL W (!) <C ~ <C 40 Q
Z <C w ~
20
o
0.0
I i
S3 - LIGHT METAL NEHRP AREAS 3,4
-8 -7 6 5 - -I i
I I I I I I I '
, I I I I I I
J:
II /1 /1 , I'
/ / / / / /
/ V
/ I
I II I I- V / / ./
/j ~ 1/ k ,,/ .---~ ~ --== ::.-:::
~ ~ ~ ~
I ---
-4
/ !
I I
I i/
/ /
I 1/ / V
J
V /'
~ ....
~ I--' [:::: ~i.-~
f....-- _ -
0.2 0.4 0.6 0.8 1.0 PGAON ROCK
123
-3
-2
-1 o !~
Figure 22. S3 - LIGHT METAL fragility curves for integer E(PMF) steps.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
84 - STEEL FRAME W/ CONC SHEARWALLS NEHRP AREAS 3,4
100
80 ~ 0 1-(.) <( 60 u. w C) <( ~ <(
40 0 z <( w ~
20
0
-5 -4 -3 -2 ' '
I I I
i I, I I~~ I 71 I I I I I
I I ! I n.J I
I I I/ I I y lj[l
I I I I I ~
' 1/ 1/1 v J
I ;i I 71 I I I I
IJ I /I I I /I /I
I I v / /
I v v / ~ "' v
It I / v l---l--/ l.--- L-----_.;-
t ~ & ~ ~ I
~
~ ~ ......
' I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 23. S4 - STL FRAME W/ COHC SHEARWALLS fragility curves for integer E(PMF) steps.
-1
0 +1 +2
124
S4 - STEEL FRAME WI CONC SHEARWALLS NEHRP AREAS 3,4
100
80 ~ 0 I-U « 60 u. w C) « ~ « 40 0 z « w ~
20
o
-5 -4 -3 -2 ,
I I I
i /, I I Ii I YI I
I I I I I I ! I LLJ i
I I II I
I Y 1/1 1 I I I I I ~
, 1/ 1/1 l J
/ Ii I /1 I I I
\
II V /, I I II /1 I / 7 /
v
I 'j V v ~ '" V
'/ / v l------./ --- !---~ ~
/; ~ ~ ~ ~ I
~ -~ ~
, I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 23. S4 - STL FRAME wI CONC SHEARWALLS fragility curves for integer E(PHF) steps.
-1
o +1 +2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.0
C1 - REINF CONC MRF NEHRP AREAS 3,4
0.2 0.4 0.6 0.8 PGAON ROCK
I
1.0
Figure 24. C1 - REIHF COHC MRF fragility curves for integer E(PMF) steps.
125
0.0
C1 - REINF CONC MRF NEHRP AREAS 3,4
0.2 0.4 0.6 O.B
PGAON ROCK
I
1.0
Figure 24. C1 - REIHF CONC HRF fragility curves for integer E(PMF) steps.
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
80 0::: 0 1-(.) <( 60 LL w (!) <( ~ <(
40 Cl z <( w ::2
20
0
C2 - REINF CONC SHEARWALL NEHRP AREAS 3,4
-4 -3 -2 _j_ I
I I
11 I
I I l I
I I
I I l ' I I !
' 11 1/ I
I l I I
I I i
i I I I l I I
I I I j
I I I I
!I
I 1/ iII ~ J i I
I !
I I VI
I /! I
I I I _/I I L v
I
1/ i/ I v v ...,1,...---!-'
/ / v I--" r-/ ~
~ +---~f"" ,_- ~~
I ~ ~ ~ k ~ ~ j..----r"' ~~
~
~I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 25. C2 - REIHF COHC SHEARWALL fragility curves for integer ~(PMF) steps.
I
I
126
-1
0
+1 +2
0::: 0 ~ () « LL W (!) « ~ « Q Z « w ~
C2 - REINF CONC SHEARWALL NEHRP AREAS 3,4
-4 -3 -2 100 1
80
60
40
20
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 25. C2 - REINF CONC SHEARWALL fragility curves for integer E(PHF) steps.
126
-1
0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
80 0:: 0 1-(.) <( 60 LL w C> <( ~ <(
40 0 z <( w ~
20
0
0.0
-
I I
I
C3/S5 - URM INFILL NEHRP AREAS 3,4
2 -1
l I I
I I I I
I I
II
J I I / vr--
1/ v v
L
I v !----~ / ..- l---" 1/ v ~ ....... ~ ~
./
j J / ~ ~ ~ ~ ~
I
~
0.2 0.4 0.6 0.8 PGAON ROCK
127
I
I
I I I
~ 0
+1
+2
1.0
Figure 26. C3/S5 - URM IHFILL fragility curves for integer E(PMF) steps.
100
80 ~ 0 ~ () <{ 60 LL W C> <{ ~ <{
40 Q Z <{ W ~
20
o 0.0
-
I I
I
C3/S5 - URM INFILL NEHRP AREAS 3,4
2 -1
/ I I
I I I I
I I
il
/ I I /" V"
I / V V
L
/ V L---~ / -~ V/ V r:: I"'" ~ r---- ;
I J v:: ~ ~
I I
~
0.2 0.4 0.6 0.8 PGAON ROCK
127
i I
I I I
~ o
+1
+2
1.0
Figure 26. C3/S5 - URM IHFILL fragility curves for integer E(PHF) steps.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
a:: 0 I-(.) <( u.. w C) <( :2 <( 0 z <( w :2
PC2 - PRECAST CONCRETE FRAME NEHRP AREAS 3,4
-3 2 - -100 I I
80 I I I I I
I 60 I
I / /
I I I I /
40 I I I v I /
I I v L_ v --
20 1/ v /
v ~
..........-I / ---r---I t ~ ~ ::,.-
0 ~ ~~
I
I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 27. PC2 - PRECAST COHC FRAME fragility curves for integer E(PMF) steps.
128
I 0
+1
+2
a: 0 l-t) « u.. w G « :2 « Cl z « w :2
PC2 - PRECAST CONCRETE FRAME NEHRP AREAS 3,4
-3 2 -1 -100 ,
I I
80 I I
I I I
V 60
I
/ V :,/
I J I I ./
40 I I / / /
I I V L V
-~
20 II V /
V V V--
/ / l.-----/ t ~ ~ ~ ...... o ~ ~
i-'
I
I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 27. PC2 - PRECAST CONC FRAME fragility curves for integer E(PHF) steps.
128
I o
+1
+2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RM - REINFORCED MASONRY NEHRP AREAS 3,4
-3 -2 100 ~==~='~4==+~~~~==+'==~=R
I I /1 80 ---H-----l---JI--t------!-1-/--+---+-~-f-----+--t
I
0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
Figure 28. RM - REINF MASONRY fragility curves for integer E(PMF) steps.
129
100
80 c:: 0 I-0 « 60 u. w (!) « :E « 40 0 z « w :2
20
o
RM - REINFORCED MASONRY NEHRP AREAS 3,4
-3 -2 I I
I
I II I
I I
II I / I j I 1/ I
II / " ./
/ I / /'
I II /' V-/' ---~ VI.", ~ ----~ ~ ~ ~ ~I -~
~ ~ :::::::-~ P""
I
0.0 0.2 0.4 0.6 0.8 PGAON ROCK
I
/ ~
~ ...
.... ... V ...
1.0
Figure 28. RM - REINF MASOHRY fragility curves for integer E(PMF) steps.
129
-1
o +1 +2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0:: 0 I-(..) <( LL w (9 <( :E <( 0 z <( w :E
130
URM - UNREINFORCED MASONRY NEHRP AREAS 3,4
100
80
60
40
20
0
-2 -I
I v 0
I
1/ /1 I I
1/ / /
1/ I v
I I I
~ / I I v
/
I / v
+1
I v ~ --/ J - +2
1/ / / v v ~ ---/
lj~ v v
~v
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 29. URM - UHREIHF MASONRY fragility curves for integer E(PMF) steps.
0::: 0 l-t) « u. w C) « ::2! « Cl z « w ::2!
130
URM - UNREINFORCED MASONRY NEHRP AREAS 3,4
100
80
60
40
20
o
-2 -1 I ,
I I \/ o
I
II I VI I
II / Y
1/ I / i I I ~ /
I I V /
I / V
+1
I V ~ ~
V J - +2
II / / V V ~ ~
V I/~ V V
~V
I I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 29. URH - UHREINF MASONRY fragility curves for integer E(PHF) steps.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
DEVELOPMENT OF A SOIL EFFECTS MODEL
A Soil Effects Model was developed to quantify the
effect of the site soil profiles on earthquaKe damage to
buildings.
Model input is:
1. DOGAMI's EarthquaKe Hazard Map Series, GMS-79.
Model output is:
1. A Soil Modification Factor (SMF) to be included
in E(PMF) in the EarthquaKe Loss Model.
The soil effects model extends current soil effects
modeling by including three types of geologic seismic
hazard:
1. Ground Motion Amplification.
2. Soil Liquefaction/Lateral Spread.
3. Dynamic Slope Instability.
As previously discussed, the Basic Structural Hazard
score (BSH) is modifed by adding applicable Performance
Modification Factors (PMF) to arrive at the final
Structural Score (S). The PMF's, ranging from -2.5
(detractors, reducing S) to +2.0 (enhancers, increasing S),
modify the BSH score to relect deviations from "average"
structural practice or conditions. The PMF values were
assigned by the ATC-21 project engineering panel expert
opinion, so that the resulting Structural Score (S = BSH +
E(PMF)) would approximately reflect the probablity of
131
DEVELOPMENT OF A SOIL EFFECTS MODEL
A Soil Effects Model was developed to quantify the
effect of the site soil profiles on earthquake damage to
buildings.
Model input is:
1. DOGAMI's Earthquake Hazard Map Series, GMS-79.
Model output is:
1. A Soil Modification Factor (SMF) to be included
in E(PMF) in the Earthquake Loss Model.
The soil effects model extends current soil effects
modeling by including three types of geologic seismic
hazard:
1. Ground Motion Amplification.
2. Soil Liquefaction/Lateral Spread.
3. Dynamic Slope Instability.
As previously discussed, the Basic Structural Hazard
score (BSH) is modifed by adding applicable Performance
Modification Factors (PMF) to arrive at the final
Structural Score (S). The PMF's, ranging from -2.5
(detractors, reducing S) to +2.0 (enhancers, increasing S),
modify the BSH score to relect deviations from "average"
structural practice or conditions. The PMF values were
assigned by the ATC-21 project engineering panel expert
opinion, so that the resulting Structural Score (S = BSH +
E(PMF» would approximately reflect the probablity of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
132
major damage for a building, given the presence of the
specific PMFs.
ATC-21 includes a Performance Modifier for Soil
Profile, SL. In this way, ATC-21 models soil profile
effects in the same manner as structural effects (such as
torsion, plan irregularity, etc.): an SL factor for less
than ideal soil conditions contributes to reducing the BSH
score and increasing the probability of major damage.
In order to provide a consistent approach, the
present study follows suit by developing a new Soil
Modification Factor (SMF), to be applied in a similar
manner to the ATC-21 Soil Profile factor, SL. However,
the SMF extends ATC-21's SL factor by including additional
potential soil hazards.
The ATC-21 Soil Profile Modifier
The ATC-21 SL modifier is based on the UBC and NEHRP
classifications of "standard" soil profiles, as follows:
SL1 (PMF = 0): RocKi or sand, gravel, or stiff clay deposits less than 200-feet thicK.
SL2 (PMF = -. 3): Sand, gravel, or stiff clay deposits greater than 200-feet thicK.
SL3 (PMF = -. 6i -.8 for buildings 8-20 stories): Soft or medium stiff clay deposits greater than 30-feet thicK.
The seismic hazard addressed by this modifier is
ground motion amplification, wherein the rocK PGA is
amplified due to the soil column "spring" effect.
132
major damage for a building, given the presence of the
specific PMFs.
ATC-21 includes a Performance Modifier for Soil
Profile, SL. In this way, ATC-21 models soil profile
effects in the same manner as structural effects (such as
torsion, plan irregularity, etc.): an SL factor for less
than ideal soil conditions contributes to reducing the BSH
score and increasing the probability of major damage.
In order to provide a consistent approach, the
present study follows suit by developing a new Soil
Modification Factor (SMF) , to be applied in a similar
manner to the ATC-21 Soil Profile factor, SL. However,
the SMF extends ATC-21's SL factor by including additional
potential soil hazards.
The ATC-21 Soil Profile Modifier
The ATC-21 SL modifier is based on the UBC and NEHRP
classifications of "standard" soil profiles, as follows:
SL1 (PMF = 0): RocKj or sand, gravel, or stiff clay deposits less than 200-feet thicK.
SL2 (PMF = -.3): Sand, gravel, or stiff clay deposits greater than 200-feet thicK.
SL3 (PMF = -.6j -.8 for buildings 8-20 stories): Soft or medium stiff clay deposits greater than 30-feet thicK.
The seismic hazard addressed by this modifier is
ground motion amplification, wherein the rocK PGA is
amplified due to the soil column "spring" effect.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
133
ATC-21 follows building code practice, which is to
amplify lateral loads by 20X for SL2 profiles, and 50X for
SL3 profiles. The SL profile, amplification modeled, and
PMF assigned are summarized in Table XXXIII, following.
TABLE XXXIII
ATC-21 SOIL PROFILE AND GROUND AMPLIFICATION
PROFILE
SL1
SL2
SL3
Amplification Factor
AMPLIF FACTOR
1.0
1.2
1. 5
PMF
0
-. 3
-. 6/-. 8
The Amplification Factor (AF) developed here
reflects ground motion amplification (similar to the SL
factor in ATC-21), using DOGAMI's GHS-79 EarthquaKe Hazard
Maps. GMS-79/Plate 2 provides a map of the amplification
hazard for the Portland Quadrangle, color coding
amplifications ranging from 1.0 or less (no amplification)
to 2. 5 or more (150X).
Since the Ground Motion Amplification from GMS-79
Plate 2 is a multiplier of bedrock seismic intensity (Rock
PGA), AF was designed to reduce the Structural ScoreS at
the same rate that the Basic Structural Hazard Score BSH is
133
ATC-21 follows building code practice, which is to
amplify lateral loads by 20X for SL2 profiles, and 50X for
SL3 profiles. The SL profile, amplification modeled, and
PMF assigned are summarized in Table XXXIII, following.
TABLE XXXIII
ATC-21 SOIL PROFILE AND GROUND AMPLIFICATION
PROFILE
SLl
SL2
SL3
Amplification Factor
AHPLIF FACTOR
1.0
1.2
1.5
PMF
o
-. 3
-. 6/-. 8
The Amplification Factor (AF) developed here
reflects ground motion amplification (similar to the SL
factor in ATC-21), using DOGAMI's GHS-79 EarthquaKe Hazard
Maps. GMS-79/Plate 2 provides a map of the amplification
hazard for the Portland Quadrangle, color coding
amplifications ranging from 1.0 or less (no amplification)
to 2.5 or more (150X).
Since the Ground Motion Amplification from GMS-79
Plate 2 is a multiplier of bedrock seismic intensity (Rock
PGA) , AF was designed to reduce the Structural Score S at
the same rate that the Basic Structural Hazard Score BSH is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
134
reduced for increasing PGA. An increase in ATC-21
Seismicity of one "level," from Low to Moderate or from
Moderate to High, roughly doubles the PGA. The average
decrease in BSH from Moderate to High Seismicity is 0.625.
Therefore a ground motion amplification of 2 (Yellow on
GMS-79, Plate 2) correlates to an AF of -0. 6. The AF was
stepped down and up for lesser and higher ground motion
amplifications than 2. GMS-79/Plate 2 map colors,
corresponding amplification, and AF developed herein, are
summarized on Table XXXIV, following.
TABLE XXXIV
AMPLIFICATION FACTOR (AF)
OREGON DOGAMI GMS-79, PLATE 2
Map Color Amplification AF
Dark Blue 1 or less 0 Dark Green 1 to 1.4 - . 2 Light Green 1.4 to 1.8 - .4 Yellow 1. 8 to 2.2 - • 6 Orange 2.2 to 2.5 - .8 Red 2. 5 or more - 1.0
It can be seen from Table IV-XIV that the range of
AF, from 0 to -1.0, approximates the analogous ATC-21 SL,
which ranges from o to -0.8.
134
reduced for increasing PGA. An increase in ATC-21
Seismicity of one "level," from Low to Moderate or from
Moderate to High, roughly doubles the PGA. The average
decrease in BSH from Moderate to High Seismicity is 0.625.
Therefore a ground motion amplification of 2 (Yellow on
GMS-79, Plate 2) correlates to an AF of -0. 6. The AF was
stepped down and up for lesser and higher ground motion
amplifications than 2. GMS-79/Plate 2 map colors,
corresponding amplification, and AF developed herein, are
summarized on Table XXXIV, following.
TABLE XXXIV
AMPLIFICATION FACTOR (AF)
OREGON DOGAMI GHS-79, PLATE 2
Map Color Amplification AF
Dark Blue 1 or less 0 Dark Green 1 to 1.4 - . 2 Light Green 1.4 to 1.8 - .4 Yellow 1.8 to 2.2 - .6 Orange 2.2 to 2.5 - .8 Red 2. 5 or more - 1.0
It can be seen from Table IV-XIV that the range of
AF, from 0 to -1.0, approximates the analogous ATC-21 SL,
which ranges from 0 to -0.8.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
135
Liquefaction Factor
The Liquefaction Factor (LF) developed here was
designed to model the contribution to building damage from
earthquaKe induced soil liquefaction and/or lateral spread.
It was developed to be used as another Performance
Modification Factor.
The DOGAMI GMS-79 maps depict soil liquefaction and
lateral spread displacement on Plates 1 and 3,
respectively. Plates 1 and 3 clearly show these two related
phenomena occurring together, in the saturated Columbia and
Willamette River floodplains. Lateral spread displacement
from Plate 3 was selected as the index to model the
liquefaction hazard. This selection was made because the
lateral spread hazard map depicts finer gradations in
severity than are shown on the liquefaction map.
Additionally, the lateral spread severity is quantified in
feet of lateral displacement, an index relevant to
building performance.
The values of LF were established from correlations
with the ground displacement shown on GMS-79 Plate 3.
Severe liquefaction or lateral spread can have a
devastating effect on building structures. Therefore, the
most severe category of ground displacement was reflected
by a modification factor comparable to the most severe PMF
135
Liquefaction Factor
The Liquefaction Factor (LF) developed here was
designed to model the contribution to building damage from
earthquake induced soil liquefaction and/or lateral spread.
It was developed to be used as another Performance
Modification Factor.
The DOGAMI GMS-79 maps depict soil liquefaction and
lateral spread displacement on Plates 1 and 3,
respectively. Plates 1 and 3 clearly show these two related
phenomena occurring together, in the saturated Columbia and
Willamette River floodplains. Lateral spread displacement
from Plate 3 was selected as the index to model the
liquefaction hazard. This selection was made because the
lateral spread hazard map depicts finer gradations in
severity than are shown on the liquefaction map.
Additionally, the lateral spread severity is quantified in
feet of lateral displacement, an index relevant to
building performance.
The values of LF were established from correlations
with the ground displacement shown on GMS-79 Plate 3.
Severe liquefaction or lateral spread can have a
devastating effect on building structures. Therefore, the
most severe category of ground displacement was reflected
by a modification factor comparable to the most severe PMF
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
136
recommended in ATC-21. In ATC-21, the most severe PMF is
Soft Story SS, which for Moderate Seismicity ranges from
-1.0 to -2.0, averaging -1. 5. Based on this correlation,
-1. 5 was selected as the LF for the most severe category
(more than 4 feet), and stepped down for less severe
ground displacements.
The values of the Liquefaction Factor developed are
shown in Table XXXV, following.
TABLE XXXV
LIQUEFACTION FACTOR
OREGON DOGAMI GMS-79, PLATE 3
Map Color
White Light Green Yellow Orange Red
Slope Factor
Ground Displacement
0 1 to 2 feet 2 to 3 feet 3 to 4 feet more than 4 feet
LF
0 - . 3 - . 6
- 1. 0 - 1. 5
The Slope Factor (SF) developed here was designed to
model the contribution to building damage from earthquaKe
induced dynamic slope instability. It was developed to be
used as another Performance Modification Factor.
Dynamic Slope Instability categories are depicted
GMS-79, Plate 3. The Slope Instability was mapped on the
136
recommended in ATC-21. In ATC-21, the most severe PMF is
Soft Story SS, which for Moderate Seismicity ranges from
-1.0 to -2.0, averaging -1.5. Based on this correlation,
-1.5 was selected as the LF for the most severe category
(more than 4 feet), and stepped down for less severe
ground displacements.
The values of the Liquefaction Factor developed are
shown in Table XXXV, following.
TABLE XXXV
LIQUEFACTION FACTOR
OREGON DOGAMI GMS-79, PLATE 3
Map Color
White Light Green Yellow Orange Red
Slope Factor
Ground Displacement
o 1 to 2 feet 2 to 3 feet 3 to 4 feet more than 4 feet
LF
o - .3 - .6
- 1. 0 - 1. 5
The Slope Factor (SF) developed here was designed to
model the contribution to building damage from earthquaKe
induced dynamic slope instability. It was developed to be
used as another Performance Modification Factor.
Dynamic Slope Instability categories are depicted
GMS-79, Plate 3. The Slope Instability was mapped on the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
137
same plate as Lateral Spread Displacement, since these 2
hazards occur on different types of ground.
The SF was developed using a correlation similar to
that used in developing the Liquefaction Factor. The most
severe category depicted on the slope instability maps is
"existing landslide," which has an effect on building
damage comparable in seriousness to severe lateral spread.
Therefore, the existing landslide category was assigned an
SF of -1. 5, and the SF was stepped down for less severe
categories.
The values of SF developed are shown in Table XXXVI.
TABLE XXXVI
SLOPE FACTOR
OREGOH DOGAMI GMS-79, PLATE 3
Map Color
White Dark Green Blue Dark Purple Light Purple
Relative Instability
Hone Slope >= 15Y. FS 1. 25 to 2 FS less than 1. 25 Existing Landslide
Soil Modification Factor
SF
0 - . 3 - . 6
- 1. 0 - 1. 5
The AF, LF, and SF are combined into a single
performance modification factor, the Soil Modification
Factor (SMF), for inclusion in the Earthquake Loss Model.
137
same plate as Lateral Spread Displacement, since these 2
hazards occur on different types of ground.
The SF was developed using a correlation similar to
that used in developing the Liquefaction Factor. The most
severe category depicted on the slope instability maps is
"existing landSlide, II which has an effect on building
damage comparable in seriousness to severe lateral spread.
Therefore, the existing landslide category was assigned an
SF of -1.5, and the SF was stepped down for less severe
categories.
The values of SF developed are shown in Table XXXVI.
TABLE XXXVI
SLOPE FACTOR
OREGON DOGAHI GMS-79, PLATE 3
Map Color
White Dark Green Blue Dark Purpl e Light Purple
Relative Instability
None Slope >= 15Y. FS 1. 25 to 2 FS less than 1. 25 Existing Landslide
Soil Modification Factor
SF
o - .3 - .6
- 1. 0 - 1. 5
The AF, LF, and SF are combined into a single
performance modification factor, the Soil Modification
Factor (SMF) , for inclusion in the Earthquake Loss Model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
138
The procedure developed for combining the component
factors into the SMF is summarized as follows:
1. Determine the AF and LF for the building site
from the GMS-79, Plates 2 and 3.
2. Use the larger of AF and LF; discard the smaller.
3. Add the SF.
This procedure was developed because liquefaction/
lateral spread and ground motion amplification hazards are
unli~ely to occur concurrently. If the ground liquefies,
its shear strength is lost, and the liquefied soil does
not have the ability to amplify shaKing. However, slope
instability and ground motion amplification could occur
simultaneously, hence SF is added to AF (SF will
automatically not be added to LF, since the mapping shows
these 2 hazards to be mutually exclusive).
COMPARISON TO GOETTEL AND HORNER STUDY
A comparison of the expected damage predicted by the
EarthquaKe Loss Model and the EarthquaKe Ris~ Analysis of
Portland made by Goettel and Horner (1995) was made.
In G&H Volume Two, damage functions are presented, in
the form of DF vs. discrete values of PGA, for "average"
buildings on roc~, "firm" soil, and "soft" soil.
Two Structure Types which are expected to suffer
high damage in Portland were selected for comparison:
138
The procedure developed for combining the component
factors into the SMF is summarized as follows:
1. Determine the AF and LF for the building site
from the GMS-79, Plates 2 and 3.
2. Use the larger of AF and LFj discard the smaller.
3. Add the SF.
This procedure was developed because liquefaction/
lateral spread and ground motion amplification hazards are
unli~ely to occur concurrently. If the ground liquefies,
its shear strength is lost, and the liquefied soil does
not have the ability to amplify shaking. However, slope
instability and ground motion amplification could occur
simultaneously, hence SF is added to AF (SF will
automatically not be added to LF, since the mapping shows
these 2 hazards to be mutually exclusive).
COMPARISON TO GOETTEL AND HORNER STUDY
A comparison of the expected damage predicted by the
Earthquake Loss Model and the Earthquake Ris~ Analysis of
Portland made by Goettel and Horner (1995) was made.
In G&H Volume Two, damage functions are presented, in
the form of DF vs. discrete values of PGA, for "average"
buildings on roc~, "firm" soil, and "soft" soil.
Two Structure Types which are expected to suffer
high damage in Portland were selected for comparison:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
139
1. Unreinforced Masonry (URM), and
2. Steel or Concrete Frames with URM infill (C3/S5).
"Average" buildings are represented in the present
study by the E(PMF) = 0 fragility curve. In Figures 30 and
31, following, fragility curve ranges are shown depicting
the increased damage to average buildings for moderate
ground motion amplification (1.0 to 1.8), high
amplification (1.8 to 2. 5), and severe lateral spread or
slope instability. Shown on the same graphs are
bar-charts depicting the estimated damage factor for
"firm" and "soft" soils predicted by G&H.
It can be observed that the moderate amplification
range (1. 0 to 1. 8) of the present study and the "firm"
soils of the G&H study predict comparable levels of
damage, especially in the . 2 to . 3 PGA range, Comparable
levels of damage are also predicted for the high
amplification range (1.8 to 2. 5) of the present study and
the "soft" soils of the G&H study.
139
1. Unreinforced Masonry (URM) , and
2. Steel or Concrete Frames with URM infill (C3/S5).
"Average" buildings are represented in the present
study by the E(PMF) = 0 fragility curve. In Figures 30 and
31, following, fragility curve ranges are shown depicting
the increased damage to average buildings for moderate
ground motion amplification (1.0 to 1.8), high
amplification (1.8 to 2.5), and severe lateral spread or
slope instability. Shown on the same graphs are
bar-charts depicting the estimated damage factor for
"firm" and "soft" soils predicted by G&H.
It can be observed that the moderate amplification
range (1.0 to 1.8) of the present study and the "firm"
soils of the G&H study predict comparable levels of
damage, especially in the. 2 to .3 PGA range. Comparable
levels of damage are also predicted for the high
amplification range (1.8 to 2.5) of the present study and
the "soft" soils of the G&H study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
URM - UNREINFORCED MASONRY
100 0
~ 80 ~
~ 0 -1-(.) I
<( 60 I LL I w I
I (.!) I
<( --1
::! I <(
40 I
0 -j
z j <( w I ::!
20
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
DAMAGE FACTORS FROM G&H REPORT
.FIRM SOIL
EFFECT OF SMF ON BASIC DAMAGE FACTOR
OsoFTSOIL D GROUND MOTION AMPLIFICATION 1.0TO 1.8
II GROUND MOTION AMPLIFICATION 1.8T02.5 m SEVERE LATERAL SPREAD OR
~SLOPE INSTABILITY POTENTIAL
Fieure 30. Comparison of URH fragility curve for "average" buildings with Goettel & Horner (1995) Damage Factor estimates.
140
URM - UNREINFORCED MASONRY
o
80 0:: 0 I-0 « 60 LL W (!) « ~ « 40 0 z J « w
I ~
20
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
DAMAGE FACTORS FROM G&H REPORT
• FIRM SOIL
EFFECT OF SMF ON BASIC DAMAGE FACTOR
o SOFT SOIL D GROUND MOTION AMPLIFICATION 1.0TO 1.8
II GROUND MOTION AMPLIFICATION 1.BT02.S m SEVERE LATERAL SPREAD OR
~ SLOPE INSTABILITY POTENTIAL
Figure 30. Comparison of URH fragility curve for "average" buildings with Goettel & Horner (1995) Damage Factor estimates.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
141
C3/S5 - URM INFILL
80 ~ 0 1-(.) <( 60 0 u. w (!) <( :E <(
40 0 z <( w :E
20
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
DAMAGE FACTORS FROM G&H REPORT
.FIRM SOIL
EFFECT OF SMF ON BASIC DAMAGE FACTOR
D GROUND MOTION AMPLIFICATION 1.0TO 1.8 OsoFTsoiL II GROUND MOTION AMPLIFICATION 1.8 TO 2.5
~ SEVERE LATERAL SPREAD OR ~SLOPE INSTABILITY POTENTIAL
Figure 31. Comparison of C3/S5 fragility curve for "average" buildings with Goettel & Horner (1995) Damage Factor estimates.
100
I I
80 ~ ~ I
~ -I ~ 60 ~ W I
~ l ~ 40 ~ Z « w ~
20
o 0.0
141
C3/S5 - URM INFILL
o
0.2 0.4 0.6 0.8 1.0 PGAON ROCK
DAMAGE FACTORS FROM G&H REPORT
• FIRM SOIL
EFFECT OF SMF ON BASIC DAMAGE FACTOR
o SOFT SOIL D GROUND MOTION AMPLIFICATION 1.0 TO 1.8
II GROUND MOTION AMPLIFICATION 1.8 TO 2.5
~ SEVERE LATERAL SPREAD OR ~ SLOPE INSTABILITY POTENTIAL
Figure 31. Comparison of C3/S5 fragility curve for "average" buildings with Goettel & Horner (1995) Damage Factor estimates.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
142
EARTHQUAKE LOSSES ESTIMATION
The result of the preceding sections of the
EarthquaKe Loss Model, (based on selected PGA, ATC-21
survey data, and earthquaKe hazard maps) is a Damage
Factor (DF) for each building in the inventory.
The final phase of the model is to estimate 2 types
of loss for each building in the inventory.
1. Dollar value of building damage, and
2. Number of lives lost and serious injuries.
Dollar Value of Building Damage
Damage Factor has been previously defined as:
DF : Dollar Loss Replacement Value
The Dollar Loss to buildings is therefore:
Dollar Loss = DF * (Replacement Value) (4. 12)
The Replacement Value of a building is not the same
thing as the buildings market value. Nor is market value
constant in relation to Replacement Value. Market value
is affected by lot location, age, and architectural style,
and may be below or above Replacement Vvlue. Furthermore,
Replacement Value should include additional post-
earthquaKe costs including cleanup and demolition.
142
EARTHQUAKE LOSSES ESTIMATION
The result of the preceding sections of the
EarthquaKe Loss Model, (based on selected PGA, ATC-21
survey data, and earthquaKe hazard maps) is a Damage
Factor (DF) for each building in the inventory.
The final phase of the model is to estimate 2 types
of loss for each building in the inventory.
1. Dollar value of building damage, and
2. Number of lives lost and serious injuries.
Dollar Value of Building Damage
Damage Factor has been previously defined as:
DF = Dollar Loss Replacement Value
The Dollar Loss to buildings is therefore:
Dollar Loss = DF * (Replacement Value) (4. 12)
The Replacement Value of a building is not the same
thing as the buildings market value. Nor is market value
constant in relation to Replacement Value. Market value
is affected by lot location, age, and architectural style,
and may be below or above Replacement Vvlue. Furthermore,
Replacement Value should include additional post-
earthquaKe costs including cleanup and demolition.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
143
Replacement Values are determined for this model
from the product of building square footage (SF) and Unit
Replacement Value (URV), as follows:
Replacement Value = SF * URV (4. 13)
The SF values are the AREA values on the building
inventory. For URV values, Appendix 15C of the NIBS
Earthquake Loss Estimation Methodology was used (National
Institute of Building Sciences 1995).
NIBS Appendix 15C presents Unit Replacement Values
as a function of Occupancy Classes. A total of 28
occupancy classes are presented. The unit replacement
costs were derived from Means Square Foot Cost 1994. NIBS
multiplied the Means "Total Building Cost" by 1. 35 to
account for contractor's overhead and profit, design fees,
and for additional post-earthquake costs including cleanup
and demolition.
The URV values used herein were determined by
grouping the 28 NIBS Occupancy Classes into the 9 ATC-21
Occupancy Types used in the inventory, and averaging each
group, then rounding the final value, based on engineering
.~udgement. The URV values used in the present mode 1 are
presented in Table XXXVII, following.
143
Replacement Values are determined for this model
from the product of building square footage (SF) and Unit
Replacement Value (URV) , as follows:
Replacement Value : SF * URV (4.13)
The SF values are the AREA values on the building
inventory. For URV values, Appendix 15C of the NIBS
Earthquake Loss Estimation Methodology was used (National
Institute of Building SCiences 1995).
NIBS Appendix 15C presents Unit Replacement Values
as a function of Occupancy Classes. A total of 28
occupancy classes are presented. The unit replacement
costs were derived from Means Square Foot Cost 1994. NIBS
multiplied the Means "Total Building Cost" by 1.35 to
account for contractor's overhead and profit, design fees,
and for additional post-earthquake costs including cleanup
and demolition.
The URV values used herein were determined by
grouping the 28 NIBS Occupancy Classes into the 9 ATC-21
Occupancy Types used in the inventory, and averaging each
group, then rounding the final value, based on engineering
,~udgement. The URV values used in the present mode I are
presented in Table XXXVII, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XXXVII
UNIT REPLACEMENT VALUES (URV)
ATC-21 OCCUPANCY TYPE
Residential Commercial Office Industrial Public Assembly School Government Building Emergency Services Historic Building
URV ($/sq. ft. )
80. 75. 75. 60. 90. 85. 75.
120. 100.
144
The Earthquake Loss Model determines the Dollar Loss
for each building in the inventory as follows:
1. Unit Replacement Value is selected from the
Occupancy Type in the inventory, and Table XXXVII
2. The Replacement Value is determined from Eq.
4. 13, using the AREA from the inventory.
3. The Dollar Loss is determined from Eq. 4. 12, using
the DF previously found in the model.
Loss of Life and Serious Iniuries
As reported in Chapter 2 herein, ATC-13 (1965)
presents in its Table 9. 3 a recommended table for
estimating casualties, which includes Fraction Dead (FD)
vs. Central Damage Factor (CDF). These values are shown
in Table XXXVIII, following.
TABLE XXXVII
UNIT REPLACEMENT VALUES (URV)
ATC-21 OCCUPANCY TYPE
Residential Commercial Office Industrial Public Assembly School Government Building Emergency Services Historic Building
URV (t/sq. ft. )
80. 75. 75. 60. 90. 85. 75.
120. 100.
144
The Earthquake Loss Model determines the Dollar Loss
for each building in the inventory as follows:
1. Unit Replacement Value is selected from the
Occupancy Type in the inventory, and Table XXXVII
2. The Replacement Value is determined from Eq.
4.13, using the AREA from the inventory.
3. The Dollar Loss is determined from Eq. 4.12, using
the DF previously found in the model.
Loss of Life and Serious Injuries
As reported in Chapter 2 herein, ATC-13 (1965)
presents in its Table 9.3 a recommended table for
estimating casualties, which includes Fraction Dead (FD)
vs. Central Damage Factor (CDF). These values are shown
in Table XXXVIII, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XXXVIII
FRACTION DEAD VS. CENTRAL DAMAGE FACTOR FROM ATC-13 TABLE 9. 3
DAMAGE CENTRAL FRACTION STATE DAMAGE FACTOR DEAD
1 0 0 2 0. 5 . 000001 3 5 . 00001 4 20 . 0001 5 45 . 001 6 80 . 01 7 100 .2
145
A continuous relationship between Damage Factor (DF)
and Fraction Dead (FD) was developed in the present
research by regressing FD against DF, as shown in Figure
32, resulting in the following exponential equation:
FD = 5. 94E-06 * exp (DF•. 104)
Fraction Dead is defined as:
FD = ...1.k_ NP
(4. 14)
where LL = lives lost, and HP = No. of People in building.
Therefore, the Loss of Life (LL) per building is:
LL = FD * HP (4. 15)
From ATC-13 Table 9, the Serious Injures (SI) are
approximately 4 times the number of deaths. Therefore:
S I = 4 * LL ( 4. 1 6 )
The EarthquaKe Loss Model determines the Loss of Life
TABLE XXXVIII
FRACTION DEAD VS. CENTRAL DAMAGE FACTOR FROM ATC-13 TABLE 9.3
DAMAGE CENTRAL FRACTION STATE DAMAGE FACTOR DEAD
1 0 0 2 O. 5 .000001 3 5 .00001 4 20 .0001 5 45 .001 6 80 .01 7 100 .2
145
A continuous relationship between Damage Factor (DF)
and Fraction Dead (FD) was developed in the present
research by regressing FD against DF, as shown in Figure
32, resulting in the following exponential equation:
FD = 5. 94E-06 * exp (DF*. 104) (4. 14)
Fraction Dead is defined as:
FD = ...1.k. NP
where LL = lives lost, and NP = No. of People in building.
Therefore, the Loss of Life (LL) per building is:
LL = FD * NP (4.15)
From ATC-13 Table 9, the Serious Injures (SI) are
approximately 4 times the number of deaths. Therefore:
S I = 4 * LL ( .q.. 16)
The EarthquaKe Loss Model determines the Loss of Life
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
-Cl u. -Cl <( w Cl z 0 I-(.) <( 0::: u.
146
FRACTION DEAD vs. DAMAGE FACTOR
0.20 I !
I I
I
• F1 pmATC-13 T, ble 9.3 I 0.15
I _1
0.10
I IT I v F)/
v 1/ F p = (5. 4E-06 exp(O. 04x [
0.05 I /
0.00 l.---' /. 0 20 40 60 80 100
DAMAGE FACTOR (DF)
Figure 32. Fraction Dead vs. Damage Factor relationship derived from ATC-13 (1985) Table 9. 3
146
FRACTION DEAD VS. DAMAGE FACTOR
0.20 I !
I I
I eF pm ATC-13 To ble 9.3 I
0.15 -o u.. - 1 I I I -
o « w o Z 0.10 V
V ~ II o
I-U « c::: u..
FP = (5. 94E-06 exp(O. 04x [ F)/
0.05 I l[
0.00 l--" ,/
I
o 20 40 60 80 DAMAGE FACTOR (OF)
Figure 32. Fraction Dead vs. Damage Factor relationship derived from ATC-13 (1985) Table 9.3
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and Serious Injuries for each inventory in the building
as follows:
1. The Fraction Dead is determined from Eq. 4. 14,
using the DF previously determined for the
building.
2. The Number of People (NP) is taKen as the
Midpoint of the Range of Number of People from
the inventory data (alternatively, LL is
determined with the Lowpoint of the range of
number of people).
3. Loss of Life is determined from Eq. 4. 15
147
4. Serious Injuries are determined from Eq. 4. 16.
Time of Day. Clearly, the Loss of Life is directly
proportional to the Number of People (NP), from Eq. 4. 15.
The number of people can be expected to vary with the time
of day. Generally, two times of day may be considered:
during business hours (Monday - Friday, 8:00 to 5:00), and
during non-business hours.
and Serious Injuries for each inventory in the building
as follows:
1. The Fraction Dead is determined from Eq. 4. 14,
using the DF previously determined for the
building.
2. The Number of People (NP) is taKen as the
Midpoint of the Range of Number of People from
the inventory data (alternatively, LL is
determined with the Lowpoint of the range of
number of people).
3. Loss of Life is determined from Eq. 4.15
147
4. Serious Injuries are determined from Eq. 4. 16.
Time of Day. Clearly, the Loss of Life is directly
proportional to the Number of People (NP), from Eq. 4.15.
The number of people can be expected to vary with the time
of day. Generally, two times of day may be considered:
during business hours (Monday - Friday, 8:00 to 5:00), and
during non-business hours.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER V
DEVELOPMENT OF A RETROFIT DIRECT BENEFIT AND COST MODEL
A model was developed to estimate the direct
benefits and costs resulting from seismic retrofit, for
each building in the regional inventory.
Retrofit Direct Benefits are defined as:
1. Dollar value of avoided building damage, and
2. Lives saved and serious injuries avoided.
Indirect benefits are not included in the model.
Indirect benefits include various items of economic loss,
including rental losses, relocation expenses, personal
income loss, business inventory loss, and personal
property loss.
Retrofit costs are defined as the total cost to
complete a life-safety retrofit, including structural
costs, restoration of architectural finishes, "light"
non-structural retrofit, design fees, permits, management,
insurance, and relocation during retrofit construction.
Retrofit is defined for this study as "life-safety"
retrofit, in which the retrofit criteria is designed to
reduce building collapse or partial collapse, without
attempting to strengthen the building to the degree
necessary to preserve building serviceability. Guidelines
CHAPTER V
DEVELOPMENT OF A RETROFIT DIRECT BENEFIT AND COST HODEL
A model was developed to estimate the direct
benefits and costs resulting from seismic retrofit. for
each building in the regional inventory.
Retrofit Direct Benefits are defined as:
1. Dollar value of avoided building damage. and
2. Lives saved and serious injuries avoided.
Indirect benefits are not included in the model.
Indirect benefits include various items of economic loss.
including rental losses. relocation expenses. personal
income loss. business inventory loss. and personal
property loss.
Retrofit costs are defined as the total cost to
complete a life-safety retrofit. including structural
costs. restoration of architectural finishes. "light"
non-structural retrofit. design fees. permits. management,
insurance. and relocation during retrofit construction.
Retrofit is defined for this study as "life-safety"
retrofit. in which the retrofit criteria is designed to
reduce building collapse or partial collapse. without
attempting to strengthen the building to the degree
necessary to preserve bUilding serviceability. Guidelines
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.1!9
for analyzing and retrofitting buildings can be found in
FEMA-178 (1992): NEHRP Handbook for the Seismic Evaluation
of Existing Buildings, and FEMA-172 (1989): Techniques for
Seismically Rehabilitating Existing Buildings.
Retrofit Effectiveness Model
First, a model was developed to estimate the
effectiveness of seismic retrofit for each building in the
inventory.
Model inputs are:
1. Input and Output from Earthquake Loss Model.
Model output is:
1. A "post-retrofit" Damage Factor (RDF) for each
building.
Retrofit Effectiveness is defined as the percentage
reduction in expected damages in the post-retrofit
facility, compared to expected damages prior to retrofit.
FEMA-227 (1992) provides estimated values of
Expected Retrofit Effectiveness (ERE) in its Tables 3-6a
and 3-6b. These estimates are for life-safety retrofit,
based on engineering experience and judgement. The
estimates are tabulated in FEMA-227 by ATC-13 Facility
Class and MMI intensity.
For use in the Retrofit Effectiveness Model herein,
facility classes in FEMA-227 Table 3-6b have been combined
to ATC-21 Structure Types, and MMI intensities converted
1J!.9
for analyzing and retrofitting buildings can be found in
FEHA-178 (1992): NEHRP Handbook for the Seismic Evaluation
of Existing Buildings, and FEMA-172 (1989): Techniques for
Seismically Rehabilitating Existing Buildings.
Retrofit Effectiveness Model
First, a model was developed to estimate the
effectiveness of seismic retrofit for each building in the
inventory.
Hodel inputs are:
1. Input and Output from Earthquake Loss Hodel.
Hodel output is:
1. A "post-retrofit" Damage Factor (RDF) for each
building.
Retrofit Effectiveness is defined as the percentage
reduction in expected damages in the post-retrofit
facility, compared to expected damages prior to retrofit.
FEHA-227 (1992) provides estimated values of
Expected Retrofit Effectiveness (ERE) in its Tables 3-6a
and 3-6b. These estimates are for life-safety retrofit,
based on engineering experience and judgement. The
estimates are tabulated in FEMA-227 by ATC-13 Facility
Class and HHI intenSity.
For use in the Retrofit Effectiveness Hodel herein,
facility classes in FEHA-227 Table 3-6b have been combined
to ATC-21 Structure Types, and HHI intensities converted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to PGA. Tables XXXIX through XLIII were developed, as
follows:
TABLE XXXIX
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR W, S3 BUILDINGS
PGA ERE(/.)
. 05 50
. 10 50
. 22 43
. 47 35 1. 02 28
TABLE XL
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR C1,S1,S2,S4 BUILDINGS
PGA ERE (I.)
. 05 35
. 10 35
. 22 31
. 47 28 1. 02 24
150
to PGA. Tables XXXIX through XLIII were developed, as
follows:
TABLE XXXIX
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR 'W, S3 BUILDINGS
PGA ERE(I.)
.05 50
.10 50
.22 43
.47 35 1. 02 28
TABLE XL
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR C1,S1,S2,S4 BUILDINGS
PGA ERE(I.)
.05 35
. 10 35
.22 31
.47 28 1. 02 24
150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XLI
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR C3/S5,C2,PC2 BUILDINGS
PGA ERE(:I.)
. 05 40
.10 40
. 22 36
. 47 33 1. 02 29
TABLE XLII
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR RM BUILDINGS
PGA ERE(:I.)
. 05 30
. 10 30
. 22 26
. 47 23 1. 02 19
151
TABLE XLI
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR C3/S5,C2,PC2 BUILDINGS
PGA ERE ( ;1,)
.05 40
.10 40
.22 36
.47 33 1. 02 29
TABLE XLII
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR RM BUILDINGS
PGA ERE(X)
.05 30
. 10 30
.22 26
.47 23 1. 02 19
151
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE XLIII
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR URM, PC1 BUILDINGS
PGA ERE(~)
. 05 50
. 10 50
. 22 45
. 47 40 1. 02 35
152
According to FEMA-227, these estimates for reduction
in damage correspond to the minimum retrofit needed to
address the life safety criteria in each facility class.
These values would not be appropriate for retrofits
designed to preserve the function of the buildings.
It can be seen from the preceding Tables that the
effectiveness of retrofit gradually decreases with
increasing seismic intensity, because stronger shaking
begins to exceed the strength of the retrofit.
The ERE values were regressed against PGA for each
of Tables XXXIX through XLIII. The result was a quadratic
relationship between ERE and PGA, of the form:
where bo, b1, and b2 are the regression coefficients.
The five regression curves and coefficients are shown in
Figs. 33 through 37, following.
TABLE XLIII
EXPECTED RETROFIT EFFECTIVENESS (ERE) FOR URM, PC1 BUILDINGS
PGA ERE (Yo)
.05 50
. 10 50
.22 45
.47 40 1. 02 35
152
According to FEMA-227, these estimates for reduction
in damage correspond to the minimum retrofit needed to
address the life safety criteria in each facility class.
These values would not be appropriate for retrofits
designed to preserve the function of the buildings.
It can be seen from the preceding Tables that the
effectiveness of retrofit gradually decreases with
increasing seismic intensity, because stronger shaking
begins to exceed the strength of the retrofit.
The ERE values were regressed against PGA for each
of Tables XXXIX through XLIII. The result was a quadratic
relationship between ERE and PGA, of the form:
where bO' b1' and b2 are the regression coefficients.
The five regression curves and coefficients are shown in
Figs. 33 through 37, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
w 50 C) <( ~ <( 0 40 z z 0 I- 30 (.) :::> 0 w 0:: I- 20 z w (.) 0:: w 10 a.
0
I
! I
0.0
RETROFIT EFFECTIVENESS W,S3
'
I I I b(O ~=534f~
~I I up __ _,I. I
I i I b(2 ~=25.9 4
"""' I
I R-~ q=99.2% I ~ I I
' ~~ I I '
.......__ I I
I ~ ! ~I i ---- I
I -
I
I
0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 33. Retrofit Effectiveness for W and S3 buildings.
153
60
w 50 C) « :2 « 0 40 Z Z 0 ..... 30 () :::J 0 w 0:: ..... 20 z w () 0:: w 10 a.
o
1 I
! I
0.0
RETROFIT EFFECTIVENESS W,S3
,
I I I b(O ~::;4r7-t-1-1 ~I i I
up
I b(2 ~=25.9 4
"'" I I R-~ q=99.2%
i ! " I I
~ I \ ~I i i , i'-.. I I
I -~b---...I 1 I --- I
I
I
I 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
Figure 33. Retrofit Effectiveness for W and S3 buildings.
153
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
w 50 C) <( ~ <( 0 z 40 z 0 I- 30 t) :::::> 0 w ~ I- 20 z w t) 0:: w 10 a.
0
I
0.0
RETROFIT EFFECTIVENESS C1 ,51 ,S2,S4
I b(O =36.4 ~93
I Ull --..:;"t. , .... u
b(2 =11.7P2
I R-s q=98.i%
I
I I i -........:
.............. !e-..... -- --....... __ --
0.2 0.4 0.6 0.8 PGAON ROCK
1.0
Ficure 34. Retrofit Effectiveness for C1, S1, S2, and S4 bui 1 dings.
154
60
w 50 CD « ~ « 0 z 40 Z 0 ..... 30 t) :::::> 0 w ~ ..... 20 z w t) 0:: W 10 a.
o
I
0.0
RETROFIT EFFECTIVENESS C1 ,S1 ,S2,S4
,
I b(O =36.4 ~93
I Dl' --£"t. ,OU
b(2 =11.7p2
I R-s Q=9S.j%
I
I , i
~ Ie--. ----..... ----
0.2 0.4 0.6 0.8 PGAON ROCK
-
1.0
Figure 34. Retrofit Effectiveness for Ci, Si, S2, and S4 bui 1 ding s.
15l!-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
w 50 (!) <( :2 <( Cl z 40 z 0 1- 30 (.) ::J Cl w n:: 1- 20 z w (.) n:: w 10 c..
0
i I
I
i
'
0.0
RETROFIT EFFECTIVENESS C3/S5,C2,PC2
I I
!
b(O iJ=41.4B93
I
U~l --.:."f. ,~u
b(2 ~=11.7 p2 j R-~ q=98.$%
'
: I
~I I I --I !
I r-----I
I
0.2 0.4 0.6 0.8
PGAON ROCK
I
I
I !
I ~
1.0
Figure 35. Retrofit Effectiveness for C3/S5, C2, and PC2 buildings
155
60
w 50 (!) « :2 « Cl Z 40 Z 0 ~ 30 () ::J Cl w n:: ~ 20 z w () n:: w 10 0...
o
I I
I
I
,
0.0
RETROFIT EFFECTIVENESS C3/S5,C2,PC2
I I b(C =41.4693
I D~ -.:;'t. I~U
b(2 =11.7 02
i R-s q=98.5% ,
I ,
"'~I I I ----I I
I --r--I
I
0.2 0.4 0.6 0.8 PGAON ROCK
I
I
I !
I
1.0
Figure 35. Retrofit Effectiveness for C3/S5, C2, and PC2 buildings
155
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
w 50 (!) <( ~ <( 0 z 40 z 0 I- 30 (.) ::> 0 w 0::: I- 20 z w (.) 0::: w 10 Q.
0
'
0.0
RETROFIT EFFECTIVENESS RM
:
I I I
I b(0
1
=31.4~93 ! I
I
I I !
! U\ I --, ... f'-IV
I 1 b(2 I =11.7 P2
I R-l q=98l~'o
I
I I I
I i I
I I
i I I I I
I I I !
i i I
I '"'~ I
~ I
---1--e.. r--!---
I
I
0.2 0.4 0.6 0.8 PGAON ROCK
156
I
I !
!
I
i I
I I I I
I ~
I
I
1.0
Figure 36. Retrofit Effectiveness for RM buildings.
RETROFIT EFFECTIVENESS RM
60 ~==+==r==~~~==~=4==~=+'==~ , I I I i b(C~=31.4~93 I J I I I ! U\ I --Lot. I iJV I
~ 50 -H __ ~I __ ~I __ -r __ r--T __ TI~b(2TO=_1=1.~7P~:2~ __ ~! ~ I R-! Q=98'1% I I
~ I i I I I Ii ~ 4°~~I--+I~i--+I~I--T-~I--+I~I--~~
~ 30 ~~~~~--+'--rl--ri' --r-~I--~~--+-~ ::> I -........ I o ~ I
W ---r---.-, 0::: !---r--~ 20~-:--II-:--!-J--J~--~~~_ w U 0::: ~ 10~~--~~~~--~~~~--~1
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
156
Figure 36. Retrofit Effectiveness for RM buildings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RETROFIT EFFECTIVENESS URM,PC1
60 ~==~=c~==+=~==~~=+==~=R I I b(0,)=52.2~68 I
I II LJ~ I --,J"t ~ t"t.f LU - I b(2 =17.QD4 rn 50 -H
1
-~-~~--~-T--T--+~~~~~~--H ~ ~ I R-!q=99.1% :2 I ..........._~ <( .......... r-......_ 0 I I --...r---... z 40 -+1-----+---1---+----t-=....r-r-.._---+---t---+--l---+-ll z I I I --r--~--- I I I
~ 30 -41----+---1----rl--+--+---+---t--+---1---h-:::> 0 LU 0::: ~ 20 -H--~--r--r--+--T--~~-~-r-~ z w {.) 0::: LU 10 -H--~--1---T--+--+---+-_,---+---1--~ a.
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Figure 37. Retrofit Effectiveness for URM and PC1 buildings.
157
60
LU 50 (!) <C :2 <C 0 Z 40 Z 0 ~ 30 (J => 0 LU 0::: ~ 20 z W (J 0::: LU 10 a.
o
I
,
I
0.0
RETROFIT EFFECTIVENESS URM,PC1
I I b(O =52.2668 I I
I I Lip
:~~~~t: I b(2
~ I R-~ q=99.1 % r:..
........... ~ r---....
I I--r-- I I I--
I I ! J
I I I
0.2 0.4 0.6 0.8 PGAON ROCK
I
I I I
1.0
F1gure 37. Retrofit Effect1veness for URH and PC1 bu1ld1ngs.
157
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
158
The RDF from the regressions is checKed against a
Minimum Damage Factor (RDF-HIN), wherein RDF is limited so
that the Damage Factor in the post-retrofit building is
not less than the DF for a building with E(PHF) equal to
+2. A E(PHF) of +2 corresponds to a modern (Post
BenchmarK) building without any detracting Performance
Modifiers; judgement dictates that a building
post-retrofit to life-safety standards should not be
considered "fitter" than this.
The Damage Factor for post-retrofit buildings (RDF)
was determined with the following steps:
1. Determine the Damage Factor (DF) for the
building prior to retrofit, based on
earthquaKe PGA, E(PHF), and the fragility
curves presented in the EarthquaKe Loss Hodel.
2. Based on Structure Type and earthquaKe PGA,
calculate the Expected Retrofit Effectiveness
(ERE) from the appropriate regression
equation.
3. Calculate RDF = DF * (1-ERE).
4. Determine a Minimum Damage Factor (RDF-HIN)
equal to the DF from the fragility curves
corresponding to E(PHF) = +2.
5. RDF equals the greater of RDF from Step 3, or
RDF-HIN from step 4.
158
The RDF from the regressions is checKed against a
Minimum Damage Factor (RDF-MIH), wherein RDF is limited so
that the Damage Factor in the post-retrofit building is
not less than the DF for a building with E(PHF) equal to
+2. A E(PHF) of +2 corresponds to a modern (Post
BenchmarK) building without any detracting Performance
MOdifiersj judgement dictates that a building
post-retrofit to life-safety standards should not be
considered "fitter" than this.
The Damage Factor for post-retrofit buildings (RDF)
was determined with the following steps:
1. Determine the Damage Factor (DF) for the
building prior to retrofit, based on
earthquaKe PGA, E(PHF) , and the fragility
curves presented in the EarthquaKe Loss Model.
2. Based on Structure Type and earthquaKe PGA,
calculate the Expected Retrofit Effectiveness
(ERE) from the appropriate regression
equation.
3. Calculate RDF = DF * (i-ERE).
4. Determine a Minimum Damage Factor (RDF-MIH)
equal to the DF from the fragility curves
corresponding to E(PHF) = +2.
5. RDF equals the greater of RDF from Step 3, or
RDF-MIH from step 4.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Retrofit Direct Benefit Hodel
Hodel inputs are:
1. Input and output £rom the retrofit
effectiveness model.
Hodel outputs are:
1. Dollar value o£ avoided building damage.
2. Humber o£ lives saved and injuries avoided.
159
The dollar value o£ avoided building damage is
determined £or each building in the inventory as follows:
1. The Replacement Value o£ the building is
determined as described in chapter IV.
2. The dollar loss o£ the post-retrofit building
is determined by multiplying the Replacement
Value by RDF.
3. The dollar value o£ avoided building damage
equals the difference between the dollar loss
o£ the existing building (determined in the
Earthquake Loss Hodel) and the dollar loss o£
the post-retrofit building.
The number o£ lives saved is determined £or each
building in the inventory as follows:
1. The Fraction Dead is determined £rom Eq. 4. 14,
using RDF in place o£ DF.
2. Loss o£ life in the post-retrofit building is
determined £rom Eq. 4. 15.
Retrofit Direct Benefit Hodel
Hodel inputs are:
1. Input and output from the retrofit
effectiveness model.
Hodel outputs are:
1. Dollar value of avoided building damage.
2. Humber of lives saved and injuries avoided.
159
The dollar value of avoided building damage is
determined for each building in the inventory as follows:
1. The Replacement Value of the building is
determined as described in chapter IV.
2. The dollar loss of the post-retrofit building
is determined by multiplying the Replacement
Value by RDF.
3. The dollar value of avoided building damage
equals the difference between the dollar loss
of the existing building (determined in the
Earthquake Loss Hodel) and the dollar loss of
the post-retrofit building.
The number of lives saved is determined for each
building in the inventory as follows:
1. The Fraction Dead is determined from Eq. 4. 14,
using RDF in place of DF.
2. Loss of life in the post-retrofit building is
determined from Eq. 4. 15.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
160
3. Number of Lives saved equals the difference
between Loss of Life in the existing building
(determined in the EarthquaKe Loss Model) and
the loss of life in the post-retrofit building.
Retrofit Cost Model
A dollar value of retrofit cost was determined for
each building in the inventory. This was determined from
the building AREA (square feet) from the inventory,
multiplied by total unit cost of retrofit (dollars per
square foot).
The unit structural retrofit costs are determined
from the published values in FEMA-156 (1994), as presented
in Chapter II herein. The values used were for the high
seismicity and life-safety performance objective
categories (Goettel and Horner (1995) recommends lower
values for west-coast RM and PC1 buildings, and their
recommendations were followed). The structural costs vary
with Structure Type and Building Area, as described in
Chapter II.
Total retrofit costs include, in addition to
structural costs, the following additional costs, as
presented in FEMA-156, Volume 2. :
1. Restoration of architectural finishes: $1 per
sq. ft. for industrial buildings, $4 for
institutional, and $3 for commercial and other.
160
3. Number of Lives saved equals the difference
between Loss of Life in the existing building
(determined in the Earthquake Loss Model) and
the loss of life in the post-retrofit building.
Retrofit Cost Model
A dollar value of retrofit cost was determined for
each building in the inventory. This was determined from
the building AREA (square feet) from the inventory,
multiplied by total unit cost of retrofit (dollars per
square foot).
The unit structural retrofit costs are determined
from the published values in FEMA-156 (1994), as presented
in Chapter II herein. The values used were for the high
seismicity and life-safety performance objective
categories (Goettel and Horner (1995) recommends lower
values for west-coast RM and PC1 buildings, and their
recommendations were followed). The structural costs vary
with Structure Type and Building Area, as described in
Chapter II.
Total retrofit costs include, in addition to
structural costs, the following additional costs, as
presented in FEMA-156, Volume 2. :
1. Restoration of architectural finishes: $1 per
sq. ft. for industrial buildings, $4 for
institutional, and $3 for commercial and other.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
161
2. Non-structural retrofit costs: $2 per sq. ft.
for industrial buildings, $~ for institutional
buildings, and $3 for commercial and other
buildings.
3. Other costs, including design fees, permits,
management, and insurance, were taken as 30/. of
the subtotal of the structural, restoration,
and non-structural costs.
~. Occupant relocation costs of $9 per sq. ft. ,
except $1.50 for RM and PC1 buildings, and
Industrial occupancies.
Because the restoration and non-structural costs
depend on the 3 broad occupancy groups of industrial,
institutional, and commercial/other, it was necessary to
group the nine ATC-21 occupancy types into one of the
three, in order to assign retrofit costs to the building
inventory. This assignment was made as follows:
1. INDUSTRIAL: Industrial
2. INSTITUTIONAL: Public Assembly, School, Emergency Services, Historic.
3. COMMERCIAL/OTHER: Residential, Commercial, Office, Government Bldg.
The total unit cost of retrofit values used are
shown in Tables XLIV through XLVI, following.
161
2. Non-structural retrofit costs: $2 per sq. ft.
for industrial buildings, $~ for institutional
buildings, and $3 for commercial and other
buildings.
3. Other costs, including design fees, permits,
management, and insurance, were taken as 30Y. of
the subtotal of the structural, restoration,
and non-structural costs.
~. Occupant relocation costs of $9 per sq. ft.,
except $1.50 for RH and PC1 buildings, and
Industrial occupancies.
Because the restoration and non-structural costs
depend on the 3 broad occupancy groups of industrial,
institutional, and commercial/other, it was necessary to
group the nine ATC-21 occupancy types into one of the
three, in order to assign retrofit costs to the building
inventory. This assignment was made as follows:
1. INDUSTRIAL: Industrial
2. INSTITUTIONAL: Public Assembly, School, Emergency Services, Historic.
3. COMMERCIAL/OTHER: Residential, Commercial, Office, Government Bldg.
The total unit cost of retrofit values used are
shown in Tables XLIV through XLVI, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
STR. TYP.
URM w PC1, RM C1,C3/S5 S1 S2,S3 S4,C2,PC2
STR. TYP.
URM w PC1, RM C1,C3/S5 S1 S2,S3 S4,C2,PC2
TABLE XLIV
TOTAL UNIT RETROFIT COSTS FOR COMMERCIAL BUILDINGS, $/SF
SMALL MEDIUM LARGE
34.66 34.49 33. 61 30. 59 31. 31 35.00 16.45 14.83 12,55 42.05 41. 36 40.20 42. 11 41.68 40. 59 26.67 26. 17 24. 33 41.73 41. 13 39.92
TABLE XLV
TOTAL UNIT RETROFIT COSTS FOR INSTITUTIONAL BUILDINGS, $/SF
SMALL MEDIUM LARGE
37.26 37.09 36.21 33. 19 33.91 37.60 19.05 17.43 15. 15 44.65 43.96 42.80 44.71 44.28 43. 19 29. 27 28.77 26.93 44. 33 43. 73 42. 52
162
V.LARGE
30. 96 40. 12 11. 25 36. 26 34. 91 21. 06 36. 13
V.LARGE
33. 56 42. 72 13. 85 38. 86 37. 51 23. 66 38. 73
STR. TYP.
URM W PC1, RM Cl,C3/S5 Sl S2,S3 S4,C2,PC2
STR. TYP.
URM W PC1, RM Cl,C3/S5 SI S2,S3 S4,C2,PC2
TABLE XLIV
TOTAL UNIT RETROFIT COSTS FOR COMMERCIAL BUILDINGS, $/SF
SMALL MEDIUM LARGE
34.66 34.49 33. 61 30. 59 31. 31 35.00 16.45 14.83 12,55 42.05 41. 36 40.20 42. 11 41.68 40. 59 26.67 26.17 24. 33 41.73 41. 13 39.92
TABLE XLV
TOTAL UNIT RETROFIT COSTS FOR INSTITUTIONAL BUILDINGS, $/SF
SMALL MEDIUM LARGE
37.26 37.09 36.21 33.19 33.91 37.60 19.05 17.43 15. 15 44.65 43.96 42.80 44.71 44.28 43. 19 29. 27 28.77 26.93 44. 33 43. 73 42. 52
162
V.LARGE
30. 96 40.12 11. 25 36. 26 34. 91 21. 06 36.13
V.LARGE
33. 56 42. 72 13.85 38. 86 37.51 23. 66 36. 73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
STR. TYP.
URM w PC1,RM C1,C3/S5 S1 S2,S3 S4,C2,PC2
TABLE XLVI
TOTAL UNIT RETROFIT COSTS FOR INDUSTRIAL BUILDINGS, $/SF
SHALL MEDIUM LARGE
23. 26 23.09 22. 21 19. 19 19. 91 23. 60 12. 55 10.93 8.65 30.65 29.96 28.80 30. 71 30. 28 29. 19 15.27 14.77 12.93 30. 33 29.73 28. 52
163
V.LARGE
19. 56 28.72
7. 35 24.86 23. 51 9.66
24.73
STR. TYP.
URM W PC1,RM C1,C3/S5 S1 S2,S3 S4,C2,PC2
TABLE XLVI
TOTAL UNIT RETROFIT COSTS FOR INDUSTRIAL BUILDINGS, $/SF
SHALL MEDIUM LARGE
23. 26 23.09 22. 21 19. 19 19.91 23. 60 12.55 10.93 8.65 30.65 29.96 28.80 30. 71 30. 28 29.19 15.27 14.77 12.93 30. 33 29.73 28. 52
163
V.LARGE
19. 56 26. 72 7.35
24.66 23. 51 9.66
24.73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER VI
DEVELOPMENT OF A REGIONAL RETROFIT PLANNING METHODOLOGY
BUILDING CLASSIFICATION SYSTEM
A system was developed to classify buildings, based
on the results of the Earthquake Loss Model and Retrofit
Direct Benefit and Cost Model. The objective of this
Building Classification System (BCS) was to prioritize
classes of buildings in a region for a cost-effective
retrofit program.
Current methods used to classify buildings for
earthquake retrofit priority are based only on structure
type (Goettel and Horner 1995). In addition to structure
type, other indices which may be useful are: site-specific
soil conditions, individual characteristics of design and
construction, occupancy type, and building size. How
important any one of these indices becomes is dependent on
the specific regional building inventory being classified,
but clearly any or all may be important. The BCS was
designed with the dual objectives of sufficient detail to
capture buildings with similar magnitudes of retrofit
benefits, while being simple enough for practical
implementation in a real-world retrofit program.
CHAPTER VI
DEVELOPMENT OF A REGIONAL RETROFIT PLANNING METHODOLOGY
BUILDING CLASSIFICATION SYSTEM
A system was developed to classify buildings, based
on the results of the Earthquake Loss Model and Retrofit
Direct Benefit and Cost Hodel. The objective of this
Building Classification System (BCS) was to prioritize
classes of buildings in a region for a cost-effective
retrofit program.
Current methods used to classify buildings for
earthquake retrofit priority are based only on structure
type (Goettel and Horner 1995). In addition to structure
type, other indices which may be useful are: site-specific
soil conditions, individual characteristics of design and
construction, occupancy type, and building size. How
important anyone of these indices becomes is dependent on
the specific regional building inventory being classified,
but clearly any or all may be important. The BCS was
designed with the dual objectives of sufficient detail to
capture buildings with similar magnitudes of retrofit
benefits, while being simple enough for practical
implementation in a real-world retrofit program.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The BCS is based on 5 indices:
1. Structural Type Group (STG)
2. Soil Hazard Zone (SHZ)
3. Structural Score Range (SSR)
4. Occupancy Type Group (OTG)
5. Building Size Group (BSG)
Hine indices of Structural Type Group are used,
based on similarities in ATC-21 structural types,
earthquake risk, and unit retrofit costs.
165
Two indices of Soil Hazard Zonee are used, based on
the range of the Soil Modification Factor (SMF), to
correlate roughly with the "firm" soils and "soft" soil
categories used in Goettel & Horner (1995), and others.
Two indices of Structural Score Range are used.
Structural Score (S = BSH + E(PMF) + SMF) captures the
effect of structure type, characteristics of design and
construction, and soil profile effects, and provides a
measure of their aggregate effect. The index was designed
to separate buildings with a significant collapse (and
life-sa£ety) risk £rom those with a lower risk.
Five indices o£ Occupancy Type Group are used,
combining ATC-21 Occupancy Types based on similarities in
occupant density, importance, and retrofit cost.
Finally, three indices of Building Size Group are
used, based on FEMA-156 (1994) categories, to help group
The BCS is based on 5 indices:
1. Structural Type Group (STG)
2. Soil Hazard Zone (SHZ)
3. Structural Score Range (SSR)
4. Occupancy Type Group (OTG)
5. Building Size Group (BSG)
Nine indices of Structural Type Group are used,
based on similarities in ATC-21 structural types,
earthquake risk, and unit retrofit costs.
165
Two indices of Soil Hazard Zonee are used, based on
the range of the Soil Modification Factor (SMF) , to
correlate roughly with the "firm" soils and "soft" soil
categories used in Goettel & Horner (1995), and others.
Two indices of Structural Score Range are used.
Structural Score (S = BSH + E(PMF) + SMF) captures the
effect of structure type, characteristics of design and
construction, and soil profile effects, and provides a
measure of their aggregate effect. The index was designed
to separate buildings with a significant collapse (and
life-safety) risk from those with a lower risk.
Five indices of Occupancy Type Group are used,
combining ATC-21 Occupancy Types based on similarities in
occupant density, importance, and retrofit cost.
Finally, three indices of Building Size Group are
used, based on FEMA-156 (1994) categories, to help group
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
166
similar occupant loads, bldg values, and retrofit cost.
The Building Classification System (BCS) indices are
as follows:
STRUCTURAL TYPE GROUP (STG):
1) S1, S2, S4 (Steel frame buildings)
2) C1 (Concrete moment frame)
3) C2 (Concrete shearwall)
4) C3/S5 (Steel or Cone frame with URM infill)
5) PC1 (Tilt-up concrete)
6) PC2 (Precast concrete frame)
7) RM (Reinforced masonry)
8) URM (Unreinforced masonry)
9) W, S3 (Wood & light metal; low-ris~ bldgs.)
SOIL HAZARD ZONE (SHZ):
1) SMF lower than -.6 (Higher soil hazard)
2) SMF equal to -. 6 or higher (Lower hazard)
STRUCTURAL SCORE RANGE (SSR):
1) Less than +1.0 (higher ris~)
2) Equal to +1.0 or greater (lower risk)
OCCUPANCY TYPE GROUP (OTG):
1) School, Public Assembly
2) Government, Emergency Service
3) Office, Commercial, Historical
4) Residential
5) Industrial
166
similar occupant loads, bldg values, and retrofit cost.
The Building Classification System (BCS) indices are
as follows:
STRUCTURAL TYPE GROUP (STG):
1) Sl, S2, S4 (Steel frame buildings)
2) Cl (Concrete moment frame)
3) C2 (Concrete shearwall)
4) C3jS5 (Steel or Conc frame with URM infill)
5) PCl (Tilt-up concrete)
6) PC2 (Precast concrete frame)
7) RM (Reinforced masonry)
8) URM (Unreinforced masonry)
9) W, S3 (Wood & light metal; low-ris~ bldgs. )
SOIL HAZARD ZONE (SHZ):
1) SMF lower than -.6 (Higher soil hazard)
2) SMF equal to -. 6 or higher (Lower hazard)
STRUCTURAL SCORE RANGE (SSR):
1) Less than +1.0 (higher ris~)
2) Equal to +1.0 or greater (lower risk)
OCCUPANCY TYPE GROUP (OTG):
1) School, Public Assembly
2) Government, Emergency Service
3) Office, Commercial, Historical
4) Residential
5) Industrial
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
167
BUILDING SIZE GROUP (BSG):
1) Small, less than 10,000 sq. ft.
2) Medium, 10,000 to 50,000 sq. ft.
3) Large and Very Large, more than 50,000 sq. ft.
In the Building Classification System, each building
in the regional inventory will fall into a Building Class
(BC) specified by five indices. For example: (2, 1, 1,4, 3}
is a concrete moment frame (C1) building type, in the
higher soil hazard zone, with a structural score less than
1.0, residential occupancy, large or very large.
The total number of possible Building Classes is:
9 X 2 X 2 X 5 X 3 : 540
However, because not all combinations of indices
occur, the number of indices used in an actual regional
classification will be less (as one example, there is
really no such thing as a light metal residence, and a
wood building in the higher risk score range is unlikely).
RETROFIT COST-BENEFIT ANALYSIS
Two approaches to a retrofit cost-benefit analysis
for a regional building inventory were taken.
First, a "Scenario Earthquake Event" was analyzed,
wherein the benefits from retrofitting that would be
realized against the full damage from a single earthquake
PGA magnitude, the Design Basis Earthquake (DBE),
167
BUILDING SIZE GROUP (BSG):
1) Small, less than 10,000 sq. ft.
2) Medium, 10,000 to 50,000 sq. ft.
3) Large and Very Large, more than 50,000 sq. ft.
In the Building Classification System, each building
in the regional inventory will fall into a Building Class
(Be) specified by five indices. For example: (2,1,1,4,3)
is a concrete moment frame (C1) building type, in the
higher solI hazard zone, with a structural score less than
1.0, residential occupancy, large or very large.
The total number of possible Building Classes is:
9 x 2 x 2 x 5 x 3 = 540
However, because not all combinations of indices
occur, the number of indices used in an actual regional
classification will be less (as one example, there is
really no such thing as a light metal reSidence, and a
wood building in the higher risk score range is unlikely).
RETROFIT COST-BEHEFIT ANALYSIS
Two approaches to a retrofit cost-benefit analysis
for a regional building inventory were taken.
First, a "Scenario Earthquake Event" was analyzed,
wherein the benefits from retrofitting that would be
realized against the full damage from a single earthquake
PGA magnitude, the Design BaSis Earthquake (DBE) ,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
168
were determined.
Second, an Expected Value analysis was made, wherein
the expected present value of retrofit benefit against the
probability of damages for a range of earthquake
magnitudes was determined over a planning horizon.
In both approaches, the results were determined by
Building Class, utilizing the outputs from the Retrofit
Direct Benefit and Retrofit Cost Models, as presented in
Chapter V.
Dollar Value of Life-Safety Benefits
In order to make a cost-benefit analysis, it is
necessary to assign a dollar value to loss of life and
serious injuries sustained.
For this study, we have used the values recommended
by Goettel & Horner (these values have also been adopted
by FEMA for valuing casualties avoided by FEMA-funded
hazard mitigation projects), as follows:
Value of Human Life: $ 2,200,000.
Value of Serious Injury: $ 12, 500.
Since, from Eq. 4. 16, the number of serious injuries
is modeled as 4 times the loss of life, the dollar value
of serious injuries are combined with loss of life, as
follows:
VOL=$ 2,200,000. + 4(12, 500.)
= $ 2, 250, 000. ( 6. 1)
168
were determined.
Second, an Expected Value analysis was made, wherein
the expected present value of retrofit benefit against the
probability of damages for a range of earthquake
magnitudes was determined over a planning horizon.
In both approaches, the results were determined by
Building Class, utilizing the outputs from the Retrofit
Direct Benefit and Retrofit Cost Models, as presented in
Chapter V.
Dollar Value of Life-Safety Benefits
In order to make a cost-benefit analysis, it is
necessary to assign a dollar value to loss of life and
serious injuries sustained.
For this study, we have used the values recommended
by Goettel & Horner (these values have also been adopted
by FEMA for valuing casualties avoided by FEHA-funded
hazard mitigation projects), as follows:
Value of Human Life: $ 2,200,000.
Value of Serious Injury: $ 12,500.
Since, from Eq. 4. 16, the number of serious injuries
is modeled as 4 times the loss of life, the dollar value
of serious injuries are combined with loss of life, as
follows:
VOL = $ 2,200,000. + 4(12,500.)
= $ 2,250,000. (6. 1)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
169
The dollar value of lives saved and serious injuries
avoided is determined by multiplying the number of lives
saved (from the Retrofit Direct Benefit Model) by VOL,
from Eq. 6. 1.
Scenario EarthquaKe Event Analysis
In maKing a cost-benefit analysis of a Scenario
EarthquaKe Event, 100Y. of the retrofit benefits and costs
are used from a selected earthquaKe PGA. The scenario
earthquaKe event used was a Design Basis Earthquake (DBE),
defined as a PGA with a 10Y. probability of occurring
within 50 years (500 year return). From the Geomatrix
(1995) report, for Portland this is a PGA of 0.20 g on
rocL
From the Retrofit Direct Benefit Model and Retrofit
Cost Model, the following were determined, and summed by
Building Class:
1. The dollar value of avoided building damage.
2. The number of lives saved.
3. The dollar value of retrofit Benefit (B),
equal to the dollar value of avoided building
damage plus the dollar value of lives saved.
4. The dollar value of the Net Benefit (HB) of
retrofit, equal to the retrofit Benefit minus
the Retrofit Cost (HB = B - C),
5. The B/C ratio.
169
The dollar value of lives saved and serious injuries
avoided is determined by multiplying the number of lives
saved (from the Retrofit Direct Benefit Model) by VOL,
from Eq. 6. 1.
Scenario Earthquake Event Analysis
In making a cost-benefit analysis of a Scenario
Earthquake Event, 100Y. of the retrofit benefits and costs
are used from a selected earthquake PGA. The scenario
earthquake event used was a Design Basis Earthquake (DBE) ,
defined as a PGA with a 10Y. probability of occurring
within 50 years (500 year return). From the Geomatrix
(1995) report, for Portland this is a PGA of 0.20 g on
rocK.
From the Retrofit Direct Benefit Model and Retrofit
Cost Model, the following were determined, and summed by
Building Class:
1. The dollar value of avoided building damage.
2. The number of lives saved.
3. The dollar value of retrofit Benefit (B),
equal to the dollar value of avoided building
damage plus the dollar value of lives saved.
4. The dollar value of the Net Benefit (HB) of
retrofit, equal to the retrofit Benefit minus
the Retrofit Cost (HB : B - C).
5. The BIC ratio.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
170
Expected Value Analysis
In maKing a cost-benefit analysis using an Expected
Value approach, the expected Net Present Value (NPV) of
retrofit benefit was determined from the sum of the
present value of benefits expected to accrue each year
over a selected planning horizon (PV), minus the initial
cost (C) of retrofit (NPV = PV - C).
The annual value of retrofit benefits are taKen as
the dollar value of Benefit (B) of a scenario earthquake
(PGA), multiplied by the annual exceedance probability of
that earthquake (p(PGA)), summed over a range of possible
earthquaKes. The range of earthquaKe scenarios and annual
exceedance probabilities used were as follows:
PGA p(PGA)
. 08 . 01
. 20 . 002
. 27 . 001
. 39 . 0004
To determine the PV of the retrofit benefit, the
annual benefits are multiplied by a Uniform Series/Present
Worth Factor (PWF), based on a selected discount rate and
planning horizon. The economic parameters used were:
Discount rate: 7Y. Planning Horizon: 50 years PWF: 13. 801
These earthquaKe scenarios, exceedance probabilities,
and economic parameters, were applied to the results of the
Retrofit Direct Benefit Model and Retrofit Cost Model. The
170
Expected Value Analysis
In maKing a cost-benefit analysis using an Expected
Value approach, the expected Net Present Value (NPV) of
retrofit benefit was determined from the sum of the
present value of benefits expected to accrue each year
over a selected planning horizon (PV) , minus the initial
cost (C) of retrofit (NPV : PV - C).
The annual value of retrofit benefits are taKen as
the dollar value of Benefit (B) of a scenario earthquaKe
(PGA) , multiplied by the annual exceedance probability of
that earthquaKe (p(PGA», summed over a range of possible
earthquaKes. The range of earthquaKe scenarios and annual
exceedance probabilities used were as follows:
PGA p(PGA)
.08 .01
.20 .002
.27 .001
.39 .0004
To determine the PV of the retrofit benefit, the
annual benefits are multiplied by a Uniform Series/Present
Worth Factor (PWF) , based on a selected discount rate and
planning horizon. The economic parameters used were:
Discount rate: 7Y. Planning Horizon: 50 years PWF: 13.801
These earthquaKe scenariOS, exceedance probabilities,
and economic parameters, were applied to the results of the
Retrofit Direct Benefit Hodel and Retrofit Cost Hodel. The
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
171
following were determined, and summed by Building Class:
1. Net Present Value (NPV = PV - C) of dollar
value of total retrofit benefit.
2. PV ;c ratio.
RETROFIT PRIORITIZATION AND SYSTEM OPTIMIZATION
Each Building Class forms an alternative solution
set of buildings for retrofit. When considering a group of
more than one Building Class for a retrofit program, that
group of Building Classes becomes a retrofit "system."
This phase of the methodology determined:
1. Building Class retrofit priority.
2. The point of maximum system efficiency for a
retrofit program.
3. The optimum retrofit budget, where Net System
Benefit is maximized in a retrofit program.
Using the results of the cost-benefit analysis, the
Building Classes were prioritized for retrofit by sorting
them from highest to lowest B/C ratios. Thereby, any
retrofit program performed in priority order will attain
the maximum net system benefit per dollar spent.
A key decision making principle in allocating
resources is the "principle of diminishing returns", which
states that eventually a point is encountered where the
return from additional resources invested begins to
171
following were determined, and summed by Building Class:
1. Net Present Value (NPV = PV - C) of dollar
value of total retrofit benefit.
2. PV/C ratio.
RETROFIT PRIORITIZATION AND SYSTEM OPTIMIZATION
Each Building Class forms an alternative solution
set of buildings for retrofit. When considering a group of
more than one Building Class for a retrofit program, that
group of Building Classes becomes a retrofit "system."
This phase of the methodology determined:
1. Building Class retrofit priority.
2. The point of maximum system efficiency for a
retrofit program.
3. The optimum retrofit budget, where Net System
Benefit is maximized in a retrofit program.
Using the results of the cost-benefit analysis, the
Building Classes were prioritized for retrofit by sorting
them from highest to lowest BIC ratiOS. Thereby, any
retrofit program performed in priority order will attain
the maximum net system benefit per dollar spent.
A key decision making principle in allocating
resources is the "principle of diminishing returns", which
states that eventually a point is encountered where the
return from additional resources invested begins to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
172
decline. This is the point of "maximum system
efficiency", beyond which additional resource investment
causes a decrease in system net benefit.
Once the Building Classes were prioritized from
highest to lowest B/C ratio, the first B/C ratio
encountered less than 1.0 marKed the point of maximum
system efficiency (this coincides with the first negative
value of Het Benefit).
A "running sum" of Retrofit Cost (C) was made of the
prioritized Building Classes. This sum is a budget for a
retrofit program consisting of all preceding Building
Classes. The optimum retrofit budget --- where System Het
Benefit is maximized --- is the sum of all Retrofit Cost
up to the point of maximum system efficiency.
A running sum of Direct Benefits --- dollar value of
building damage avoided and lives saved --- was also made
of the prioritized Building Class sequence. This may be
used to determine the direct benefits attained for various
retrofit budgets.
172
decline. This is the point of "maximum system
efficiency", beyond which additional resource investment
causes a decrease in system net benefit.
Once the Building Classes were prioritized from
highest to lowest B/C ratiO, the first B/C ratio
encountered less than 1.0 marKed the point of maximum
system efficiency (thiS coincides with the first negative
value of Ret Benefit).
A "running sum" of Retrofit Cost (C) was made of the
prioritized Building Classes. This sum is a budget for a
retrofit program consisting of all preceding Building
Classes. The optimum retrofit budget --- where System Ret
Benefit is maximized --- is the sum of all Retrofit Cost
up to the point of maximum system efficiency.
A running sum of Direct Benefits --- dollar value of
building damage avoided and lives saved --- was also made
of the prioritized Building Class sequence. This may be
used to determine the direct benefits attained for various
retrofit budgets.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER VII
DEVELOPMENT OF REGIONAL EARTHQUAKE LOSS AND RETROFIT ANALYSIS SOFTWARE
Computer software was developed to perform the
earthquake loss modeling, loss estimation, retrofit
benefit modeling, cost-benefit analysis, building
classification and prioritization, and system efficiency
analysis presented in Chapters IV, v, and VI.
The software has been named REAL-RAP, an acronym for
Regional EArthquake Loss and Retrofit Analysis Program.
REAL-RAP consists of a number of program modules, which
operate sequentially. The sequence of execution of the
modules is as follows:
1. Earthquake loss estimates.
2. Retrofit direct benefits and costs.
3. Regional retrofit classification, prioritization,
and optimization.
EARTHQUAKE LOSS ESTIMATATION MODULES
DAM-FACT Module
Input: The ATC-21 Rapid Survey data. An example from
the input database may be found in Figure 2. The input
CHAPTER VII
DEVELOPMENT OF REGIONAL EARTHQUAKE LOSS AND RETROFIT ANALYSIS SOFTWARE
Computer software was developed to perform the
earthquake loss modeling, loss estimation, retrofit
benefit modeling, cost-benefit analysis, building
classification and prioritization, and system efficiency
analysis presented in Chapters IV, V, and VI.
The software has been named REAL-RAP, an acronym for
Regional EArthquake Loss and Retrofit Analysis Program.
REAL-RAP conSists of a number of program modules, which
operate sequentially. The sequence of execution of the
modules is as follows:
1. Earthquake loss estimates.
2. Retrofit direct benefits and costs.
3. Regional retrofit classification, prioritization,
and optimization.
EARTHQUAKE LOSS ESTlHATATION MODULES
DAM-FACT Module
Input: The ATC-21 Rapid Survey data. An example from
the input database may be found in Figure 2. The input
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
174
format is a comma-delimited text file, which permits the
use of exported data from any spreadsheet or database
software. For the Portland Quad, the survey data was
grouped into files by survey section (square mile).
Processing: The DAM-FACT module calculates the
damage factor for each building in the inventory. The
module contains twelve subroutines, one for each ATC-21
Building Type. Based on the Performance Modifiers and
Soil Modifiers for each building, a value of E(PMF) is
determined, and the DF is calculated from the regression
equations of the fragility curves presented in Chapter IV.
This is done for a range of six PGA values, which may
adjusted by the user.
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, structural score, soil modification factor,
and the Damage Factor for six PGA values. A sample of the
output is shown on Figure 38.
DAMAGE-$ Module
Input: Two data files are input: the Rapid Survey
data and the output from the DAM-FACT module, by section.
Processing: Based on the Occupancy Type and building
area, a replacement value is determined for each building,
as described in Chapter IV. The dollar value of damage is
determined from the damage factor, for six PGA values.
174
format is a comma-delimited text file, which permits the
use of exported data from any spreadsheet or database
software. For the Portland Quad, the survey data was
grouped into files by survey section (square mile).
Processing: The DAM-FACT module calculates the
damage factor for each building in the inventory. The
module contains twelve subroutines, one for each ATC-21
Building Type. Based on the Performance Modifiers and
Soil Modifiers for each building, a value of E(PMF) is
determined, and the DF is calculated from the regression
equations of the fragility curves presented in Chapter IV.
This is done for a range of six PGA values, which may
adjusted by the user.
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, structural score, soil modification factor,
and the Damage Factor for six PGA values. A sample of the
output is shown on Figure 38.
DAMAGE-$ Module
Input: Two data files are input: the Rapid Survey
data and the output from the DAM-FACT module, by section.
Processing: Based on the Occupancy Type and building
area, a replacement value is determined for each building,
as described in Chapter IV. The dollar value of damage is
determined from the damage factor, for six PGA values.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
175
Output: A data file, by section, of one line for
each building, consisting of sequence number, building
type, area, occupancy type, and the dollar loss for six
PGA values. A sample of the output is shown on Figure 39.
LIFELOSS Module
Input: Two data files are input: the Rapid Survey
data and the output from the DAM-FACT module, by section.
Processing: Based on the range of estimated Humber
of People and the Damage Factor, the probabilistic loss of
life is determined for six PGA values for each building,
in accordance with the model presented in Chapter IV. The
Midpoint of the range of the number of people is used
(this module may be adjusted to compute the loss of life
based on other points, such as lowpoint, in the range).
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, range of number of people, occupancy type,
and the loss of life for six PGA values. A sample of the
output is shown on Figure 40.
DAM-SUM Module
Input: The input is the output from the DAM-FACT,
DAMAGE-$, and LIFELOSS modules, for one section.
Processing: The loss estimation results, which were
made previously building-by-building, are summed to
175
Output: A data file, by section, of one line for
each building, consisting of sequence number, building
type, area, occupancy type, and the dollar loss for six
PGA values. A sample of the output is shown on Figure 39.
LIFELOSS Module
Input: Two data files are input: the Rapid Survey
data and the output from the DAM-FACT module, by section.
Processing: Based on the range of estimated Humber
of People and the Damage Factor, the probabilistic loss of
life is determined for six PGA values for each building,
in accordance with the model presented in Chapter IV. The
Midpoint of the range of the number of people is used
(thiS module may be adjusted to compute the loss of life
based on other pOints, such as lowpoint, in the range).
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, range of number of people, occupancy type,
and the loss of life for six PGA values. A sample of the
output is shown on Figure 40.
DAM-SUM Module
Input: The input is the output from the DAM-FACT,
DAMAGE-$, and LIFELOSS modules, for one section.
Processing: The loss estimation results, which were
made previously bui I ding-by-bui 1 ding, are summed to
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited without perm
ission.
~ t%J M
~ .... ID
~ I
~ ~
3: 0
~ .... ID
0 !=! r+
'8 r+ ..., ..... .... ID
Section 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08
Sequence Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 26 29
Structure Structure Soil Factor Type ScoreS SMF w 4.1 -0.4 C1 1.1 -0.4 w 5.1 -0.4 w 3.6 -0.4 w 5.6 -0.4 w 5.6 -0.4 w 4.6 -0.4
URM 0.1 -0.4 RM 2.1 -0.4 RM 2.1 -0.4 w 4.1 -0.4 RM 2.1 -0.4 RM 1.6 -0.4
URM 0.1 -0.4 C2 2.1 -0.4 w 4.6 -0.4 w 4.6 -0.4 RM 2.1 -0.4 w 4.8 -0.2 w 5.8 -0.2 w 4.8 -0.2 C2 1.8 -0.2 w 5.8 -0.2 w 5.6 -0.4 w 4.6 -0.4 w 4.6 -0.2 C2 1.6 -0.2 w 4.8 -0.2 C2 2.1 -0.4
PGA=.OB PGA=.20 4.811788 10.38851 9.044233 37.7771 3.891901 8.44262 5.458322 11.87867 3.549243 7.772906 3.549243 7.772906 4.264622 9.322571 53.77915 100 4.9232 15.14601 4.9232 15.14601
4.811788 10.38851 4.9232 15.14601
6.038383 19.07551 53.77915 100 6.439631 15.21982 4.264622 9.322571 4.264622 9.322571 4.9232 15.14601
3.891901 8.44262 3.290639 7.202165 3.891901 8.44262 6.439631 15.21982 3.290639 7.202165 3.549243 7.772906 4.264622 9.322571 3.691901 8.44262 6.439631 15.21962 3.891901 8.44262 6.439631 15.21982
DAMAGE FACTOR PGA=.27 PGA-.39 13.03165 17.57891 60.26953 100 10.56485 13.7705 15.04541 21.0445 9.758814 12.60356 9.758814 12.60356 11.72124 15.47527
100 100 22.78669 38.4232 22.78669 38.4232 13.03165 17.57891 22.78669 38.4232 31.33173 65.2707
100 100 20.21264 29.94182 11.72124 15.47527 11.72124 15.47527 22.78669 38.4232 10.56485 13.7705 9.034731 11.5919 10.56485 13.7705 20.21264 29.94182 9.034731 11.5919 9.758814 12.60356 11.72124 15.47527 10.56485 13.7705 20.21264 29.94162 10.56485 13.7705 20.21264 29.94182
PGA=.SO 22.90541
100 16.74474 28.85403 14.86898 14.86898 19.15064
100 55.21106 55.21106 22.90541 55.21106
100 100
41.65246 19.15064 19.15064 55.21106 16.74474 13.48683 16.74474 41.65246 13.48683 14.86898 19.15064 16.74474 41.65246 16.74474 41.65246
PGA=.60 29.7501
100 19.94902 39.40286 16.89887 16.89887 23.26276
100 72.17931 72.17931 29.7501
72.17931 100 100
55.9801 23.26276 23.26276 72.17931 19.94902 15.01868 19.94902 55.9801 15.01868 16.89887 23.26276 19.94902 55.9801 19.94902 55.9801 ....
-1 Ol
Section Sequence Structure Structure Soil Factor DAMAGE FACTOR 1N1E08 Number Type ScoreS SMF PGA=.08 PGA=.20 PGA=.27 PGA-.39 PGA=.SO PGA=.60 1N1E08 1 W 4.1 -0.4 4.811788 10.38851 13.03165 17.57891 22.90541 29.7501 1N1E08 2 C1 1.1 -0.4 9.044233 37.7771 60.26953 100 100 100 1N1E08 3 W 5.1 -0.4 3.891901 8.44262 10.56485 13.7705 16.74474 19.94902 1N1E08 4 W 3.6 -0.4 5.458322 11.87867 15.04541 21.0445 28.85403 39.40286 1N1E08 5 W 5.6 -0.4 3.549243 7.772906 9.758814 12.60356 14.86898 16.89887 1N1E08 6 W 5.6 -0.4 3.549243 7.772906 9.758814 12.60356 14.86898 16.89887 1N1E08 7 W 4.6 -0.4 4.264622 9.322571 11.72124 15.47527 19.15064 23.26276 1N1E08 8 URM 0.1 -0.4 53.77915 100 100 100 100 100 1N1E08 9 RM 2.1 -0.4 4.9232 15.14601 22.78669 38.4232 55.21106 72.17931 1N1E08 10 RM 2.1 -0.4 4.9232 15.14601 22.78669 38.4232 55.21106 72.17931 1N1E08 11 W 4.1 -0.4 4.811788 10.38851 13.03165 17.57891 22.90541 29.7501 1N1E08 12 RM 2.1 -0.4 4.9232 15.14601 22.78669 38.4232 55.21106 72.17931 1N1E08 13 RM 1.6 -0.4 6.038383 19.07551 31.33173 65.2707 100 100 1N1E08 14 URM 0.1 -0.4 53.77915 100 100 100 100 100 1N1E08 15 C2 2.1 -0.4 6.439631 15.21982 20.21264 29.94182 41.65246 55.9801 1N1E08 16 W 4.6 -0.4 4.264622 9.322571 11.72124 15.47527 19.15064 23.26276 1N1E08 17 W 4.6 -0.4 4.264622 9.322571 11.72124 15.47527 19.15064 23.26276 1N1E08 18 RM 2.1 -0.4 4.9232 15.14601 22.78669 38.4232 55.21106 72.17931 1N1E08 19 W 4.8 -0.2 3.891901 8.44262 10.56485 13.7705 16.74474 19.94902 1N1E08 20 W 5.8 -0.2 3.290639 7.202165 9.034731 11.5919 13.48683 15.01868 1N1E08 21 W 4.8 -0.2 3.891901 8.44262 10.56485 13.7705 16.74474 19.94902 1N1E08 22 C2 1.8 -0.2 6.439631 15.21982 20.21264 29.94182 41.65246 55.9801 1N1E08 23 W 5.8 -0.2 3.290639 7.202165 9.034731 11.5919 13.48683 15.01868 1N1E08 24 W 5.6 -0.4 3.549243 7.772906 9.758814 12.60356 14.86898 16.89887 1N1E08 25 W 4.6 -0.4 4.264622 9.322571 11.72124 15.47527 19.15064 23.26276 1N1E08 26 W 4.8 -0.2 3.891901 8.44262 10.56485 13.7705 16.74474 19.94902 1N1E08 27 C2 1.8 -0.2 6.439631 15.21982 20.21264 29.94182 41.65246 55.9801 1N1E08 28 W 4.8 -0.2 3.891901 8.44262 10.56485 13.7705 16.74474 19.94902 1N1E08 29 C2 2.1 -0.4 6.439631 15.21982 20.21264 29.94182 41.65246 55.9801
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited without perm
ission.
! tzj 1111
~ ...... ID
0 ...,
I tzj I
* :::c 0
~ ...... ID
0 J:::l t+
'S t+
.... .... ...... ID
Section 1N1EOB 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 --
Sequence Structure Number Type
1 w 2 C1 3 w 4 w 5 w 6 w 7 w 8 URM 9 RM 10 RM 11 w 12 RM 13 RM 14 URM 15 C2 16 w 17 w 18 RM 19 w 20 w 21 w 22 C2 23 w 24 w 25 w 26 w 27 C2 28 w 29 C2
--~-~-
Area Occupancy Sq-Ft PGA=.OB 2970 Residential 11432 2860 Commercial 19399 630 Commercial 1838 1986 Commercial 8130 2436 Emer. Serv. 10375 6668 Office 17749 2116 Residential 7219 2500 Commercial 100835 6000 Office 22154 4000 Office 14769 3440 Commercial 12414 1288 Commercial 4755 1650 Commercial 7472 1432 Commercial 57758 6230 Commercial 30089 9356 Pub. Assem. 35909 3060 Pub. Assem. 11744 2100 Commercial 7754 1540 Commercial 4495 3772 Pub. Assem. 11171 6596 Commercial 19253 5792 Industrial 22379 936 Commercial 2310 2516 Office 6697 2756 Commercial 8814 10932 Commercial 31909 6037 Commercial 29157 1970 Commercial 5750 12180 Commercial 58826
-- ----
$-DAMAGE PGA=.20 PGA=.27 PGA-.39
24683 30963 41767 81031 129278 214500 3989 4991 6506 17693 22410 31345 22721 28526 36842 38872 48803 63030 15781 19841 26196 187500 187500 187500 68157 102540 172904 45438 68360 115269 26802 33621 45353 14631 22011 37116 23605 38773 80772 107400 107400 107400 71114 94443 139903 78499 98697 130307 25674 32280 42618 23854 35889 60516 9751 12202 15904
24449 30671 39352 41765 52264 68122 52891 70242 104053 5055 6342 8137 14667 18414 23782 19269 24227 31987 69221 86621 112904 68911 91517 135569 12473 15609 20345
139033 184642 273518 -- --
PGA=.50 54423 214500 7911
42978 43465 74359 32418 187500 248449 165633 59095 53333 123750 107400 194621 161256 52740 86957 19340 45785 82836 144750 9467
28057 39584 137290 188591 24740
380495
PGA-.60 70686 214500 9425 58690 49398 84511 39379 187500 324806 216537 76755 69725 123750 107400 261567 195881 64065 113682 23041 50985 98687 194542 10543 31888 48084 163562 253463 29474
511378 ..... ~ ~
Section Sequence Structure Area Occupancy $-DAMAGE 1N1E08 Number Type Sq-Ft PGA=.08 PGA=.20 PGA=.27 PGA-.39 PGA=.50 PGA-.60 1N1E06 1 W 2970 Residential 11432 24683 30963 41767 54423 70666 1N1E06 2 C1 2860 Commercial 19399 81031 129278 214500 214500 214500 1N1E08 3 W 630 Commercial 1838 3989 4991 6506 7911 9425 1N1E08 4 W 1986 Commercial 8130 17693 22410 31345 42978 58690 1N1E06 5 W 2436 Emer. Servo 10375 22721 28526 36842 43465 49398 1N1E08 6 W 6668 Office 17749 36872 48803 63030 74359 84511 1N1E08 7 W 2116 Residential 7219 15781 19841 26196 32418 39379 1N1E06 8 URM 2500 Commercial 100835 187500 187500 187500 187500 187500 1N1E08 9 RM 6000 Office 22154 68157 102540 172904 248449 324806 1N1E08 10 RM 4000 Office 14769 45438 68360 115269 165633 216537 1N1E08 11 W 3440 Commercial 12414 26802 33621 45353 59095 76755 1N1E06 12 RM 1268 Commercial 4755 14631 22011 37116 53333 69725 1N1E08 13 RM 1650 Commercial 7472 23605 38773 80772 123750 123750 1N1E08 14 URM 1432 Commercial 57758 107400 107400 107400 107400 107400 1N1E08 15 C2 6230 Commercial 30089 71114 94443 139903 194621 261567 1N1E08 16 W 9356 PUb. Assem. 35909 78499 98697 130307 161256 195881 1N1E08 17 W 3060 PUb. Assem. 11744 25674 32280 42618 52740 64065 1N1E06 16 RM 2100 Commercial 7754 23654 35889 60516 86957 113682 1N1E06 19 W 1540 Commercial 4495 9751 12202 15904 19340 23041 1N1E08 20 W 3772 Pub. Assem. 11171 24449 30671 39352 45785 50985 1N1E08 21 W 6596 Commercial 19253 41765 52264 68122 82836 98687 1N1E08 22 C2 5792 Industrial 22379 52891 70242 104053 144750 194542 1N1E08 23 W 936 Commercial 2310 5055 6342 8137 9467 10543 1N1E06 24 W 2516 Office 6697 14667 18414 23782 28057 31888 1N1E08 25 W 2756 Commercial 8814 19269 24227 31987 39584 48084 1N1E08 26 W 10932 Commercial 31909 69221 86621 112904 137290 163562 1N1E08 27 C2 6037 Commercial 29157 68911 91517 135569 188591 253463 1N1E08 28 W 1970 Commercial 5750 12473 15609 20345 24740 29474 1N1E08 29 C2 12180 Commercial 58826 139033 184642 273518 380495 511378
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited without perm
ission.
~ tzl M
~ .... ID
0 .... t"' 1-1
~ 0 [I) [I)
3: 0 §' .... ID
0 ~ t+ 'g t+
.... ..... .... ID
Section 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1EOB 1N1E08 1N1EOB 1N1E08 1N1E08 1N1E08 1N1E08 1N1EOB 1N1EOB 1N1E08 1N1E08 1N1E08 1N1E08 1N1EOB
Sequence Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Structure Number Type People w 0-10 C1 0-10 w 11-30 w 11-30 w 0-10 w 11-30 w 11-30
URM 11-30 RM 11-30 RM 30-100 w 0-10
RM 11-30 RM 11-30
URM 11-30 C2 11-30 w 30-100 w 30-100 RM 0-10 w 0-10 w 30-100 w 11-30 C2 11-30 w 0-10 w 0-10 w 0-10 w 30-100 C2 30-100 w 11-30 C2 30-100
Occupancy PGA=.OB PGA=.20
Residential 4.90E-05 8.75E-05 Commercial 7.61E-05 1.51E-03 Commercial 1.78E-04 2.86E-04 Commercial 2.10E-04 4.09E-04 Emer. Serv. 4.30E-05 6.67E-05
Office 1.72E-04 2.67E-04 Residential 1.85E-04 3.13E-04 Commercial 3.19E-02 3.903726
Office 1.98E-04 5.74E-04 Office 6.44E-04 1.87E-03
Commercial 4.90E-05 8.75E-05 Commercial 1.98E-04 5.74E-04 Commercial 2.23E-04 8.64E-04 Commercial 3.19E-02 3.903726 Commercial 2.32E-04 5.78E-04 Pub. Assem. 6.02E-04 1.02E-03 Pub. Assem. 6.02E-04 1.02E-03 Commercial 4.96E-05 1.43E-04 Commercial 4.45E-05 7.15E-05 Pub. Assem. 5.44E-04 8.17E-04 Commercial 1.78E-04 2.86E-04
Industrial 2.32E-04 5.78E-04 Commercial 4.18E-05 6.2BE-05
Office 4.30E-05 6.67E-05 Commercial 4.63E-05 7.83E-05 Commercial 5.79E-04 9.29E-04 Commercial 7.54E-04 1.88E-03 Commercial 1.78E-04 2.86E-04 Commercial 7.54E-04 1.88E-03
----
LOSS OF LIFE PGA=.27 PGA=.39 1.15E-04 1.85E-04 0.015665 0.975931 3.56E-04 4.97E-04 5.68E-04 1.06E-03 8.19E-05 1.10E-04 3.28E-04 4.41E-04 4.02E-04 5.94E-04 3.903726 3.903726 1.27E-03 6.46E-03 4.13E-03 2.10E-02 1.15E-04 1.85E-04 1.27E-03 6.46E-03 3.09E-03 0.105408 3.903726 3.903726 9.72E-04 2.67E-03 1.31E-03 1.93E-03 1.31E-03 1.93E-03 3.18E-04 1.62E-03 8.91E-05 1.24E-04 9.88E-04 1.29E-03 3.56E-04 4.97E-04 9.72E-04 2.67E-03 7.60E-05 9.92E-05 8.19E-05 1.10E-04 1.00E-04 1.48E-04 1.16E-03 1.62E-03 3.16E-03 8.69E-03 3.56E-04 4.97E-04 3.16E-03 8.69E-03
PGA=.SO 3.22E-04 0.975931 6.78E-04 2.39E-03 1.39E-04 5.58E-04 8.71E-04 3.903726 3.70E-02 0.120336 3.22E-04 3.70E-02 3.903726 3.903726 9.04E-03 2.83E-03 2.83E-03 9.26E-03 1.69E-04 1.57E-03 6.78E-04 9.04E-03 1.21E-04 1.39E-04 2.18E-04 2.20E-03 2.94E-02 6.78E-04 2.94E-02
PGA=.60 6.55E-04 0.975931 9.46E-04 7.15E-03 1.72E-04 6.89E-04 1.34E-03 3.903726 0.216228 0.702742 6.55E-04 0.216228 3.903726 3.903726 4.01E-02 4.34E-03 4.34E-03 5.41E-02 2.36E-04 1.84E-03 9.46E-04 4.01E-02 1.42E-04 1.72E-04 3.34E-04 3.07E-03 0.130356 9.46E-04 0.130356 ....
~ Q)
Section Sequence Structure Number Occupancy LOSS OF LIFE 1N1E08 Number Type People PGA=.08 PGA=.20 PGA=.27 PGA=.39 PGA=.50 PGA=.60 1N1EOB 1 W 0-10 Residential 4.90E-OS B.7SE-OS 1.1SE-04 1.BSE-04 3.22E-04 6.SSE-04 1N1EOS 2 C1 0-10 Commercial 7.61E-05 1.51E-03 0.01566S 0.975931 0.97S931 0.975931 1N1EOB 3 W 11-30 Commercial 1.7BE-04 2.B6E-04 3.56E-04 4.97E-04 6.7BE-04 9.46E-04 1N1EOB 4 W 11-30 Commercial 2.10E-04 4.09E-04 5.6BE-04 1.06E-03 2.39E-03 7.1SE-03 1N1EOB 5 W 0-10 Emer. Servo 4.30E-05 6.67E-05 B.19E-OS 1.10E-04 1.39E-04 1.72E-04 1N1EOB 6 W 11-30 Office 1.72E-04 2.67E-04 3.28E-04 4.41E-04 5.SSE-04 6.B9E-04 1N1EOB 7 W 11-30 Residential 1.S5E-04 3.13E-04 4.02E-04 5.94E-04 B.71E-04 1.34E-03 1N1EOS B URM 11-30 Commercial 3.19E-02 3.903726 3.903726 3.903726 3.903726 3.903726 1N1EOB 9 RM 11-30 Office 1.9SE-04 5.74E-04 1.27E-03 6.46E-03 3.70E-02 0.21622B 1N1EOB 10 RM 30-100 Office 6.44E-04 1.S7E-03 4.13E-03 2.10E-02 0.120336 0.702742 1N1EOB 11 W 0-10 Commercial 4.90E-05 B.75E-OS 1.1SE-04 1.BSE-04 3.22E-04 6.SSE-04 1N1E08 12 RM 11-30 Commercial 1.98E-04 5. 74E-04 1.27E-03 6.46E-03 3.70E-02 0.216228 1N1EOS 13 RM 11-30 Commercial 2. 23 E-04 8.64E-04 3.09E-03 0.105408 3.903726 3.903726 1N1E08 14 URM 11-30 Commercial 3.19E-02 3.903726 3.903726 3.903726 3.903726 3.903726 1N1EOB 15 C2 11-30 Commercial 2.32E-04 5.78E-04 9.72E-04 2.67E-03 9.04E-03 4.01E-02 1N1E08 16 W 30-100 Pub. Assem. 6.02E-04 1.02E-03 1.31E-03 1.93E-03 2.83E-03 4.34E-03 1N1EOB 17 W 30-100 PUb. Assem. 6.02E-04 1.02E-03 1.31E-03 1.93E-03 2.S3E-03 4.34E-03 1N1EOB 18 RM 0-10 Commercial 4.96E-OS 1.43E-04 3.18E-04 1.62E-03 9.26E-03 5.41E-02 1N1EOB 19 W 0-10 Commercial 4.45E-05 7.15E-05 B.91E-05 1.24E-04 1.69E-04 2.36E-04 1N1EOB 20 W 30-100 Pub. Assem. 5.44E-04 8.17E-04 9.8BE-04 1.29E-03 1.57E-03 1.B4E-03 1N1EOB 21 W 11-30 Commercial 1.7BE-04 2.B6E-04 3. 56E-04 4.97E-04 6.7BE-04 9.46E-04 1N1EOB 22 C2 11-30 Industrial 2.32E-04 5.7BE-04 9.72E-04 2.67E-03 9.04E-03 4.01E-02 1N1EOB 23 W 0-10 Commercial 4.1BE-05 6.2BE-05 7.60E-OS 9.92E-05 1.21E-04 1.42E-04 1N1EOB 24 W 0-10 Office 4.30E-05 6.67E-05 B.19E-OS 1.10E-04 1.39E-04 1.72E-04 1N1EOS 25 W 0-10 Commercial 4.63E-05 7.B3E-05 1.00E-04 1.4BE-04 2. 1 BE-04 3.34E-04 1N1EOB 26 W 30-100 Commercial 5.79E-04 9.29E-04 1.16E-03 1.62E-03 2.20E-03 3.07E-03 1N1EOB 27 C2 30-100 Commercial 7.54E-04 1.88E-03 3.16E-03 8.69E-03 2.94E-02 0.130356 1N1EOB 2B W 11-30 Commercial 1.7BE-04 2.B6E-04 3.56E-04 4.97E-04 6.7BE-04 9.46E-04 1N1EOB 29 C2 30-100 Commercial 7.S4E-04 1.BBE-03 3.16E-03 B.69E-03 2.94E-02 0.130356
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
179
determine the total losses for one section by Building
Type. Structural Scores and Damage Factors are averaged
by Building Type and for the section.
output: A data file where each line gives the
section totals for each Building Type: number of
buildings, area, average structural score, average damage
factor, total dollar value of damage, and total loss of
life. The TABLE-1 sub-module writes this output file in
table form.
REGION-L Modules
Input: The inputs are the outputs from the DAM-SUM
modules, for a number of sections comprising a region.
Processing: The loss estimation results, which were
summed by section in DAM-SUM, are summed for the region,
by Building Type. Structural Scores and Damage Factors
are averaged by Building Type and for the region.
Output: A data file where each line shows the region
totals for each Building Type: number of buildings, area,
average structural score, average damage factor, total
dollar value of damage, and total loss of life. The TABLE-2
sub-module writes this output in table form; Tables XLVII
and LI were derived from the TABLE-2 output. Six tables
are output, for six PGA values.
179
determine the total losses for one section by Building
Type. Structural Scores and Damage Factors are averaged
by Building Type and for the section.
Output: A data file where each line gives the
section totals for each Building Type: number of
buildings, area, average structural score, average damage
factor, total dollar value of damage, and total loss of
life. The TABLE-1 sub-module writes this output file in
table form.
REGION-L Modules
Input: The inputs are the outputs from the DAM-SUM
modules, for a number of sections comprising a region.
Processing: The loss estimation results, which were
summed by section in DAM-SUM, are summed for the region,
by Building Type. Structural Scores and Damage Factors
are averaged by Building Type and for the region.
Output: A data file where each line shows the region
totals for each Building Type: number of buildings, area,
average structural score, average damage factor, total
dollar value of damage, and total loss of life. The TABLE-2
sub-module writes this output in table formj Tables XLVII
and LI were derived from the TABLE-2 output. Six tables
are output, for six PGA values.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
180
RETROFIT DIRECT BENEFIT AND COST MODULES
RETRO-DF Module
Input: The ATC-21 Rapid Survey data, by section.
Processing: The RETRO-DF module calculates the
post-retrofit d~age factor for each building in the
inventory. The module first calculates the d~age factor
for the building, in a procedure identical to the DAM-FACT
module. The RETRO-DF module then calculates the retrofit
effectiveness and reduces the damage factor, in accordance
with the model presented in Chapter V. This is done for a
range of six PGA values, which may adjusted by the user.
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, structural score, soil modification factor,
and the post-retrofit Damage Factor for six PGA values. A
sample of the output is shown in Figure 41.
RETRO-$D Module
Input: Two data files are input: the Rapid Survey
data and the output from the RETRO-DF module, by section.
Processing: Based on the Occupancy Type and building
area, a replacement value is determined for each building,
as described in Chapter IV. The dollar value of damage is
then determined from the post-retrofit damage factor, for
six PGA values.
180
RETROFIT DIRECT BENEFIT AND COST MODULES
RETRO-DF Module
Input: The ATC-21 Rapid Survey data, by section.
Processing: The RETRO-DF module calculates the
post-retrofit d~age factor for each building in the
inventory. The module first calculates the d~age factor
for the building, in a procedure identical to the DAM-FACT
module. The RETRO-DF module then calculates the retrofit
effectiveness and reduces the damage factor, in accordance
with the model presented in Chapter V. This is done for a
range of six PGA values, which may adjusted by the user.
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, structural score, soil modification factor,
and the post-retrofit Damage Factor for six PGA values. A
sample of the output is shown in Figure 41.
RETRO-$D Module
Input: Two data files are input: the Rapid Survey
data and the output from the RETRO-DF module, by section.
Processing: Based on the Occupancy Type and building
area, a replacement value is determined for each building,
as described in Chapter IV. The dollar value of damage is
then determined from the post-retrofit damage factor, for
six PGA values.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
181
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, area, occupancy type, and the post-retrofit
dollar loss for six PGA values. A sample of the output is
shown in Figure 42.
RETRO-LL Module
Input: Two data files are input: the Rapid Survey
data and the output from the RETRO-LL module, by section.
Processing: The post-retrofit loss of life is
determined for each building in the inventory using the
same procedure as the LIFELOSS module, except that the
post-retrofit damage factor is used. This is done of six
PGA values.
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, range of number of people, occupancy type,
and the post-retrofit loss of life for six PGA values. A
sample of the output is shown on Figure 43.
181
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, area, occupancy type, and the post-retrofit
dollar loss for six PGA values. A sample of the output is
shown in Figure 42.
RETRO-LL Module
Input: Two data files are input: the Rapid Survey
data and the output from the RETRO-LL module, by section.
Processing: The post-retrofit loss of life is
determined for each building in the inventory using the
same procedure as the LIFELOSS module, except that the
post-retrofit damage factor is used. This is done of six
PGA values.
Output: A data file, by section, of one line for
each inventory building, consisting of sequence number,
building type, range of number of people, occupancy type,
and the post-retrofit loss of life for six PGA values. A
sample of the output is shown on Figure 43.
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited without perm
ission.
~ tzl M
~ .... 11)
0 ..... ~
~ ~ 0 I
~ :J: 0 g-.... 11)
0 J;: r+
'S r+ ..... ..... .... 11)
Section 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08
Sequence Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Structure Structure Type ScoreS w 4.1 C1 1.1 w 5.1 w 3.6 w 5.6 w 5.6 w 4.6
URM 0.1 RM 2.1 RM 2.1 w 4.1
RM 2.1 RM 1.6
URM 0.1 C2 2.1 w 4.6 w 4.6 RM 2.1 w 4.8 w 5.8 w 4.8 C2 1.8 w 5.8 w 5.6 w 4.6 w 4.8 C2 1.8 w 4.8 C2 2.1
Soil Factor SMF PGA=.OB -0.4 2.581024 -0.4 5.912015 -0.4 2.581024 -0.4 2.754901 -0.4 2.581024 -0.4 2.581024 -0.4 2.581024 -0.4 27.09475 -0.4 3.464347 -0.4 3.464347 -0.4 2.581024 -0.4 3.464347 -0.4 4.249076 -0.4 27.09475 -0.4 3.887463 -0.4 2.581024 -0.4 2.581024 -0.4 3.464347 -0.2 2.581024 -0.2 2.581024 -0.2 2.581024 -0.2 3.887463 -0.2 2.581024 -0.4 2.581024 -0.4 2.581024 -0.2 2.581024 -0.2 3.887463 -0.2 2.581024 -0.4 3.887463
DAMAGE FACTOR AFTER RETROFIT PGA=.20 PGA=.27 PGA=.39 PGA=.50 PGA-.60 5.793753 7.625787 11.01063 15.06143 20.24399 25.64031 41.69332 71.14931 72.66019 73.78796 5.613468 7.023073 8.918213 11.01049 13.57467 6.624827 8.804189 13.18132 18.97294 26.81239 5.613468 7.023073 8.918213 10.17858 11.49914 5.613468 7.023073 8.918213 10.17858 11.49914 5.613468 7.023073 9.693004 12.59249 15.82956 53.92881 55.77072 58.53855 60.64324 62.19765 11.03729 16.90273 29.25901 42.87702 56.86861 11.03729 16.90273 29.25901 42.87702 56.86861 5.793753 7.625787 11.01063 15.06143 20.24399 11.03729 16.90273 29.25901 42.87702 56.86861 13.90082 23.24128 49.70319 77.66019 78.78797 53.92881 55.77072 58.53855 60.64324 62.19765 9.569098 12.97209 19.80631 28.18213 38.50756 5.613468 7.023073 9.693004 12.59249 15.829561 5.613468 7.023073 9.693004 12.59249 15.82956 I
11.03729 16.90273 29.25901 42.87702 56.86861 5.613468 7.023073 8.918213 11.01049 13.57467 5.613468 7.023073 8.918213 10.17858 11.02138 5.613468 7.023073 8.918213 11.01049 13.57467 9.569098 12.97209 19.80631 28.18213 38.50756 5.613468 7.023073 8.918213 10.17858 11.02138 5.613468 7.023073 8.918213 10.17858 11.49914 5.613468 7.023073 9.693004 12.59249 15.82956 5.613468 7.023073 8.918213 11.01049 13.57467 9.569098 12.97209 19.80631 28.18213 38.50756 5.613468 7.023073 8.918213 11.01049 13.57467 9.569098 12.97209 19.80631 28.1821 ~_l8.507561 ....
()) 1'\)
Section Sequence Structure Structure Soil Factor DAMAGE FACTOR AFTER RETROFIT 1N1EOB Number Type Score S SMF PGA=.08 PGA=.20 PGA=.27 PGA=.39 PGA=.50 PGA-.60 1N1E08 1 W 4.1 -0.4 2.581024 5.793753 7.625787 11.01063 15.06143 20.24399 1N1E08 2 C1 1.1 -0.4 5.912015 25.64031 41.69332 71.14931 72.66019 73.78796 1N1E08 3 W 5.1 -0.4 2.581024 5.613468 7.023073 8.918213 11.01049 13.57467 1N1E08 4 W 3.6 -0.4 2.754901 6.624827 8.804189 13.18132 18.97294 26.81239 1N1E08 5 W 5.6 -0.4 2.581024 5.613468 7.023073 8.918213 10.17858 11.49914 1N1E08 6 W 5.6 -0.4 2.581024 5.613468 7.023073 8.918213 10.17858 11.49914 1N1E08 7 W 4.6 -0.4 2.581024 5.613468 7.023073 9.693004 12.59249 15.82956 1N1E08 8 URM 0.1 -0.4 27.09475 53.92881 55.77072 58.53855 60.64324 62.19765 1N1E08 9 RM 2.1 -0.4 3.464347 11.03729 16.90273 29.25901 42.87702 56.86861 1N1E08 10 RM 2.1 -0.4 3.464347 11.03729 16.90273 29.25901 42.87702 56.86861 1N1E08 11 W 4.1 -0.4 2.581024 5.793753 7.625787 11.01063 15.06143 20.24399 1N1E08 12 RM 2.1 -0.4 3.464347 11.03729 16.90273 29.25901 42.87702 56.86861 1N1E08 13 RM 1.6 -0.4 4.249076 13.90082 23.24128 49.70319 77.66019 78.78797 1N1E08 14 URM 0.1 -0.4 27.09475 53.92881 55.77072 58.53855 60.64324 62.19765 1N1E08 15 C2 2.1 -0.4 3.887463 9.569098 12.97209 19.80631 28.18213 38.50756 1N1E08 16 W 4.6 -0.4 2.581024 5.613468 7.023073 9.693004 12.59249 15.82956 1N1E08 17 W 4.6 -0.4 2.581024 5.613468 7.023073 9.693004 12.59249 15.82956 1N1E08 18 RM 2.1 -0.4 3.464347 11.03729 16.90273 29.25901 42.87702 56.86861 1N1E08 19 W 4.8 -0.2 2.581024 5.613468 7.023073 8.918213 11.01049 13.57467 1N1E08 20 W 5.8 -0.2 2.581024 5.613468 7.023073 8.918213 10.17858 11.02138 1N1E08 21 W 4.8 -0.2 2.581024 5.613468 7.023073 8.918213 11.01049 13.57467 1N1E08 22 C2 1.8 -0.2 3.887463 9.569098 12.97209 19.80631 28.18213 38.50756 1N1E08 23 W 5.8 -0.2 2.581024 5.613468 7.023073 8.918213 10.17858 11.02138 1N1E08 24 W 5.6 -0.4 2.581024 5.613468 7.023073 8.918213 10.17858 11.49914 1N1E08 25 W 4.6 -0.4 2.581024 5.613468 7.023073 9.693004 12.59249 15.82956 1N1E08 26 W 4.8 -0.2 2.581024 5.613468 7.023073 8.918213 11.01049 13.57467 1N1E08 27 C2 1.8 -0.2 3.887463 9.569098 12.97209 19.80631 28.18213 38.50756 1N1E08 28 W 4.8 -0.2 2.581024 5.613468 7.023073 8.918213 11.01049 13.57467 1N1E08 29 C2 2.1 -0.4 3.887463 9.569098 12.97209 19.80631 28.18213 38.50756
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited without perm
ission.
~ tzl M
~ .... 111
0 .... ~
~ ~ 0 I
* 1:::1 :J: 0
~ .... 111
0 ~ t+ 'g t+
.... ..... .... 111
Section 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08 1N1E08
Sequence Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
Structure Area Occupancy Type Sq-Ft w 2970 Residential C1 2860 Commercial w 630 Commercial w 1986 Commercial w 2436 Emer. Serv. w 6668 Office w 2116 Residential
URM 2500 Commercial RM 6000 Office RM 4000 Office w 3440 Commercial
RM 1288 Commercial RM 1650 Commercial
URM 1432 Commercial C2 6230 Commercial w 9356 Pub. Assem. w 3060 Pub. Assem.
RM 2100 Commercial w 1540 Commercial w 3772 Pub. Assem. w 6596 Commercial C2 5792 Industrial w 936 Commercial w 2516 Office w 2756 Commercial w 10932 Commercial C2 6037 Commercial w 1970 Commercial C2 12180 Commercial
PGA=.OB 6132 12681 1219 4103 7544 12907 4369 50802 15589 10393 6659 3346 5258
29099 18164 21733 7108 5456 2981 8762 12768 13509 1811 4870 5334
21161 17601 3813 35511
$-DAMAGE AFTER RETROFIT PGA=.20 PGA=.27 PGA=.39 PGA=.50
13765 18118 26161 35785 54998 89432 152615 155856 2652 3318 4213 5202 9867 13113 19633 28260 16409 20529 26069 29754 28072 35122 44599 50903 9502 11888 16408 21316
101116 104570 109759 113706 49667 76062 131665 192946 33111 50708 87777 128631 14947 19674 28407 38858 10662 16328 28264 41419 17202 28761 61507 96104 57919 59897 62870 65130 44711 60612 92544 131681 47267 59137 81618 106033 15459 19341 26694 34679 17383 26621 46082 67531 6483 8111 10300 12717 19056 23841 30275 34554 27769 34743 44118 54468 33254 45080 68830 97938 3940 4930 6260 7145 10592 13252 16828 19206 11603 14516 20035 26028 46024 57582 73120 90275 43326 58734 89678 127601 8293 10376 13176 16267 87413 118500 180930 257443
PGA=.60 48099 158275 6414 39937 33614 57507 26796 116620 255908 170605 52229 54935 97500 66800 179926 133291 43594 89568 15678 37415 67153 133821 7737 21698 32719 111298 174352 20056 351766 ..
(J) (.1.1
Section Sequence Structure Area Occupancy 1N1E08 Number Type Sq-Ft PGA=.08 1N1E08 1 W 2970 Residential 6132 1N1E08 2 C1 2860 Commercial 12681 1N1E08 3 W 630 Commercial 1219 1N1E08 4 W 1986 Commercial 4103 1N1E08 5 W 2436 Emer. Servo 7544 1N1E08 6 W 6668 Office 12907 1N1E08 7 W 2116 Residential 4369 1N1E08 8 URM 2500 Commercial 50802 1N1E08 9 RM 6000 Office 15589 1N1E08 10 RM 4000 Office 10393 1N1E08 11 W 3440 Commercial 6659 1N1E08 12 RM 1288 Commercial 3346 1N1E08 13 RM 1650 Commercial 5258 1N1E08 14 URM 1432 Commercial 29099 1N1E08 15 C2 6230 Commercial 18164 1N1E08 16 W 9356 Pub. Assem. 21733 1N1E08 17 W 3060 Pub. Assem. 7108 1N1E08 18 RM 2100 Commercial 5456 1N1E08 19 W 1540 Commercial 2981 1N1E08 20 W 3772 Pub. Assem. 8762 1N1E08 21 W 6596 Commercial 12768 1N1E08 22 C2 5792 Industrial 13509 1N1E08 23 W 936 Commercial 1811 1N1EOB 24 W 2516 Office 4870 1N1E08 25 W 2756 Commercial 5334 1N1E08 26 W 10932 Commercial 21161 1N1E08 27 C2 6037 Commercial 17601 1N1E08 28 W 1970 Commercial 3813 1N1E08 29 C2 12180 Commercial 35511
$-DAMAGE AFTER RETROFIT PGA=.20 PGA=.27 PGA=.39 PGA=.50
13765 18118 26161 35785 54998 89432 152615 155856 2652 3318 4213 5202 9867 13113 19633 28260 16409 20529 26069 29754 28072 35122 44599 50903 9502 11888 16408 21316
101116 104570 109759 113706 49667 76062 131665 192946 33111 50708 87777 128631 14947 19674 28407 38858 10662 16328 28264 41419 17202 28761 61507 96104 57919 59897 62870 65130 44711 60612 92544 131681 47267 59137 81618 106033 15459 19341 26694 34679 17383 26621 46082 67531 6483 8111 10300 12717 19056 23841 30275 34554 27769 34743 44118 54468 33254 45080 68830 97938 3940 4930 6260 7145 10592 13252 16828 19206 11603 14516 20035 26028 46024 57582 73120 90275 43326 58734 89678 127601 8293 10376 13176 16267 87413 118500 180930 257443
PGA=.60 48099 158275 6414 39937 33614 57507 26796 116620 255908 170605 52229 54935 97500 66800 179926 133291 43594 89568 15678 37415 67153 133821 7737 21698 32719 111298 174352 20056 351766 ....
0> ()J
Reproduced w
ith permission of the copyright ow
ner. Further reproduction prohibited without perm
ission.
~ tJ ~ ..... R)
0 ..... ::.:J ~ ::.:J 0 I
t:"' t:"'
:J: 0 go ..... R)
0 t:: .... 'g .... ..... .... ..... R)
Section 1N1EOB 1N1EOB 1N1E08 1N1E08 1N1EOB 1N1EOB 1N1EOB 1N1E08 1N1EOB 1N1EOB 1N1EOB 1N1E08 1N1EOB 1N1EOB 1N1EOB 1N1EOB 1N1EOB 1N1E08 1N1EOB 1N1EOB 1N1EOB 1N1E08 1N1E08 1N1E08 1N1EOB 1N1EOB 1N1EOB 1N1EOB 1N1E08 1N1EOB
Sequence Number
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
--
Structure Number Type People w 0-10 C1 0-10 w 11-30 w 11-30 w 0-10 w 11-30 w 11-30
URM 11-30 RM 11-30 RM 30-100 w 0-10 RM 11-30 RM 11-30
URM 11-30 C2 11-30 w 30-100 w 30-100 RM 0-10 w 0-10 w 30-100 w 11-30 C2 11-30 w 0-10 w 0-10 w 0-10 w 30-100 C2 30-100 w 11-30 C2 30-100
Occupancy PGA=.OB
Residential 3.8BE-05 Commercial 5.49E-05 Commercial 1.55E-04 Commercial 1.58E-04 Emer. Serv. 3.88E-05
Office 1.55E-04 Residential 1.55E-04 Commercial 1.99E-03
Office 1.70E-04 Office 5.54E-04
Commercial 3.8BE-05 Commercial 1.70E-04 Commercial 1.85E-04 Commercial 1.99E-03 Commercial 1.78E-04 Pub. Assem. 5.05E-04 Pub. Assem. 5.05E-04 Commercial 4.26E-05 Commercial 3.8BE-05 Pub. Assem. 5.05E-04 Commercial 1.55E-04
Industrial 1.78E-04 Commercial 3.88E-05
Office 3.88E-05 Commercial 3.88E-05 Commercial 5.05E-04 Commercial 5.78E-04 Commercial 1.55E-04 Commercial 5.78E-04
LOSS OF LIFE AFTER RETROFIT PGA=.20 PGA=.27 PGA-.39 PGA-.50 5.43E-05 6.56E-05 9.33E-05 1.42E-04 4.27E-04 2.27E-03 4.86E-02 5.68E-02 2.13E-04 2.47E-04 3.00E-04 3.73E-04 2.37E-04 2.97E-04 4.68E-04 8.55E-04 5.32E-05 6.17E-05 7.51E-05 8.56E-05 2.13E-04 2.47E-04 3.00E-04 3.42E-04 2.13E-04 2.47E-04 3.26E-04 4.40E-04 3.24E-02 3.92E-02 5.23E-02 0.065143 3.74E-04 6.89E-04 2.49E-03 0.010267 1.22E-03 2.24E-03 8.10E-03 3.34E-02 5.43E-05 6.56E-05 9.33E-05 1.42E-04 3.74E-04 6.89E-04 2.49E-03 0.010267 5.04E-04 1.33E-03 2.09E-02 0.382353 3.24E-02 3.92E-02 5.23E-02 0.065143 3.21E-04 4.58E-04 9.32E-04 2.23E-03 6.92E-04 8.02E-04 1.06E-03 1.43E-03 6.92E-04 8.02E-04 1.06E-03 1.43E-03 9.36E-05 1.72E-04 6.23E-04 2.57E-03 5.32E-05 6.17E-05 7.51E-05 9.33E-05 6.92E-04 8.02E-04 9.76E-04 1.11E-03 2.13E-04 2.47E-04 3.00E-04 3.73E-04 3.21E-04 4.58E-04 9.32E-04 2.23E-03 5.32E-05 6.17E-05 7.51E-05 8.56E-05 5.32E-05 6.17E-05 7.51E-05 8.56E-05 5.32E-05 6.17E-05 8.14E-05 1.10E-04 6.92E-04 8.02E-04 9.76E-04 1.21E-03 1.04E-03 1.49E-03 3.03E-03 7.24E-03 2.13E-04 2.47E-04 3.00E-04 3.73E-04 1.04E-03 1.49E-03 3.03E-03 7.24E-03
PGA=.60 2.44E-04 6.39E-02 4.87E-04 1.93E-03 9.82E-05 3.93E-04 6.16E-04 0.076573 0.043993 0.142976 2.44E-04 0.043993 0.429935 0.076573 6.52E-03, 2.00E-03 2.00E-03 1.10E-02 1.22E-04 1.21E-03 4.87E-04 6.52E-03 9.34E-05 9.82E-05 i
1.54E-04' 1.58E-03 2.12E-02 4.87E-04 2.12E-02
1
.... CD 4=
Section Sequence Structure Number Occupancy LOSS OF LIFE AFTER RETROFIT 1N1EOB Number Type People PGA=.OB PGA=.20 PGA=.27 PGA-.39 PGA-.50 PGA=.60 1N1E08 1 W 0-10 Residential 3.88E-05 5.43E-05 6.56E-05 9.33E-05 1.42E-04 2.44E-04 1N1EOB 2 C1 0-10 Commercial 5.49E-05 4.27E-04 2.27E-03 4.86E-02 5.6BE-02 6.39E-02 1N1E08 3 W 11-30 Commercial 1.55E-04 2.13E-04 2.47E-04 3.00E-04 3.73E-04 4.B7E-04 1N1E08 4 W 11-30 Commercial 1.5BE-04 2.37E-04 2.97E-04 4.68E-04 B.55E-04 1.93E-03 1N1EOB 5 W 0-10 Emer. Servo 3.BBE-05 5.32E-05 6. 17E-05 7.51E-05 B.56E-05 9.82E-05 1N1EOB 6 W 11-30 Office 1.55E-04 2.13E-04 2.47E-04 3.00E-04 3.42E-04 3.93E-04 1N1EOB 7 W 11-30 Residential 1.55E-04 2.13E-04 2.47E-04 3.26E-04 4.40E-04 6.16E-04 1N1E08 8 URM 11-30 Commercial 1.99E-03 3.24E-02 3.92E-02 5.23E-02 0.065143 0.076573 1N1EOB 9 RM 11-30 Office 1.70E-04 3.74E-04 6.89E-04 2.49E-03 0.010267 0.043993 1N1EOB 10 RM 30-100 Office 5.54E-04 1.22E-03 2.24E-03 B.10E-03 3.34E-02 0.142976 1N1EOB 11 W 0-10 Commercial 3.8BE-05 5.43E-OS 6.SSE-OS 9.33E-OS 1.42E-04 2.44E-04 1N1EOB 12 RM 11-30 Commercial 1.70E-04 3.74E-04 6.89E-04 2.49E-03 0.010267 0.043993 1N1EOB 13 RM 11-30 Commercial 1.BSE-04 S.04E-04 1.33E-03 2.09E-02 0.382353 0.429935 1N1E08 14 URM 11-30 Commercial 1.99E-03 3.24E-02 3. 92 E-02 5.23E-02 0.065143 0.076573 1N1EOB 15 C2 11-30 Commercial 1.7BE-04 3.21E-04 4.58E-04 9.32E-04 2.23E-03 6.52E-03 1N1EOB 16 W 30-100 Pub. Assem. 5.05E-04 6.92E-04 8.02E-04 1.06E-03 1.43E-03 2.00E-03 1N1E08 17 W 30-100 Pub. Assem. 5.0SE-04 S.92E-04 B.02E-04 1.0SE-03 1.43E-03 2.00E-03 1N1EOB 1B RM 0-10 Commercial 4.26E-05 9.36E-05 1.72E-04 6.23E-04 2.57E-03 1.10E-02 1N1EOB 19 W 0-10 Commercial 3.BBE-05 5.32E-05 6.17E-05 7.51E-05 9.33E-05 1.22E-04 1N1EOB 20 W 30-100 Pub. Assem. 5.0SE-04 6.92E-04 B.02E-04 9.76E-04 1.11E-03 1.21E-03 1N1EOB 21 W 11-30 Commercial 1.SSE-04 2.13E-04 2.47E-04 3.00E-04 3.73E-04 4.87E-04 1N1EOB 22 C2 11-30 Industrial 1.78E-04 3.21E-04 4.58E-04 9. 32E-04 2.23E-03 6.52E-03 1N1EOB 23 W 0-10 Commercial 3.8BE-05 5.32E-05 6. 17E-05 7.S1E-OS B.5SE-OS 9.34E-05 1N1EOB 24 W 0-10 Office 3.8BE-OS 5.32E-OS 6. 17E-OS 7.S1E-05 B.S6E-05 9.82E-05 1N1EOB 2S W 0-10 Commercial 3.BBE-05 S.32E-OS 6. 17E-05 8.14E-05 1.10E-04 1.54E-04 1N1E08 26 W 30-100 Commercial S.OSE-04 6.92E-04 B.02E-04 9.76E-04 1.21E-03 1.SBE-03 1N1EOB 27 C2 30-100 Commercial 5.78E-04 1.04E-03 1.49E-03 3.03E-03 7.24E-03 2.12E-02 1N1EOB 2B W 11-30 Commercial 1.55E-04 2. 13E-04 2.47E-04 3.00E-04 3.73E-04 4.87E-04 1N1EOB 29 C2 30-100 Commercial S.7BE-04 1.04E-03 1.49E-03 3.03E-03 7.24E-03 2.12E-02
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
185
RFBEN-MM Modules
Input: The ouput from the DAM-FACT, DAMAGE-$,
LIFELOSS, RETRO-$D, and RETRO-LL modules, by section.
Processing: The dollar value of building damage
avoided and lives saved are determined in accordance with
the Retrofit Direct Benefit Model presented in Chapter V,
for each building in the inventory. The retrofit cost is
determined for each building in accordance with the
Retrofit Cost Model presented in chapter V. There is a
separate RFBEN-•* module for each PGA value.
output: A data file, by section, of one line for
each building, consisting of sequence number, building
type, soil modification factor, structural score,
occupancy type, area, retrofit cost, value of damage
avoided, and lives saved. This output file contains all
the information needed to "stuff" the retrofit data of
each building into the regional Building Classification
System, the next step.
REGIONAL rtETROFIT CLASSIFICATION, PRIORITIZATION, AND OPTIMIZATION MODULES
BCS-RBEN Module
Input: The output from RFBEN-MM, by section.
Processing: The retrofit direct benefits and costs
for each building from the RFBEN-•* output are summed into
a regional Building Classification System (BCS), in
185
RFBEN-** Modules
Input: The ouput from the DAM-FACT, DAMAGE-$,
LIFELOSS, RETRO-$D, and RETRO-LL modules, by section.
Processing: The dollar value of building damage
avoided and lives saved are determined in accordance with
the Retrofit Direct Benefit Model presented in Chapter V,
for each building in the inventory. The retrofit cost is
determined for each building in accordance with the
Retrofit Cost Model presented in chapter V. There is a
separate RFBEN-** module for each PGA value.
output: A data file, by section, of one line for
each building, consisting of sequence number, building
type, soil modification factor, structural score,
occupancy type, area, retrofit cost, value of damage
avoided, and lives saved. This output file contains all
the information needed to "stuff" the retrofit data of
each building into the regional Building Classification
System, the next step.
REGIONAL rtETROFIT CLASSIFICATION, PRIORITIZATION, AND OPTIMIZATION MODULES
BCS-RBEN Module
Input: The output from RFBEN-**, by section.
Processing: The retrofit direct benefits and costs
for each building from the RFBEN-** output are summed into
a regional Building Classification System (BCS) , in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
accordance with the BCS model presented in Chapter VI.
The minimum number of building classes required to
classify all the buildings in the region are used.
186
Output: A data file cotaining one line for each
building class in the region, with the five BCS indices,
total retrofit cost, total value o£ building damage
avoided, and total number of lives saved for each class.
RETRO-BC Module
Input: The output file from the BCS-RBEH module.
Processing: Total dollar value of retrofit benefit
and net benefit values are determined for each building
class, in accordance with the scenario earthquaKe
cost-benefit model presented in Chapter VI.
Output: A data file with one line for each building
class, with the five BCS indices, total retrofit cost, net
benefit, and B/C ratio for each building class. There is a
separate run and output file for each scenario PGA event.
EXP-VAL Module
Input: The output files from RETRO-BC, for a range
of scenario PGA values; annual exceedance probabilities
for each PGA; discount rate; planning horizon in years.
Processing: The PV, HPV, and PV/C ratios are
determined for the range of PGA event values, according to
the expected value analysis procedure presented in Ch. VI.
accordance with the BCS model presented in Chapter VI.
The minimum number of building classes required to
classify all the buildings in the region are used.
186
Output: A data file cotaining one line for each
building class in the region, with the five BCS indices,
total retrofit cost, total value of building damage
avoided, and total number of lives saved for each class.
RETRO-BC Module
Input: The output file from the BCS-RBEH module.
Processing: Total dollar value of retrofit benefit
and net benefit values are determined for each building
class, in accordance with the scenario earthquake
cost-benefit model presented in Chapter VI.
Output: A data file with one line for each building
class, with the five BCS indices, total retrofit cost, net
benefit, and B/C ratio for each building class. There is a
separate run and output file for each scenario PGA event.
EXP-VAL Module
Input: The output files from RETRO-BC, for a range
of scenario PGA valuesj annual exceedance probabilities
for each PGAj discount ratej planning horizon in years.
Processing: The PV, HPV, and PV/C ratios are
determined for the range of PGA event values, according to
the expected value analysis procedure presented in Ch. VI.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
187
Output: A data file containing one line for each
building class, with the five BCS indices, total retrofit
cost, net present value (HPV) of benefit, and PV/C ratio
for each building class.
SORT-BC Module
Input: The output file from the RETRO-BC or EXP-VAL
modules.
Processing: The Building Classes are prioritized for
retrofit by sorting from high to low B/C (or PV/C) ratio,
using a selection-sort algorithm.
Output: A file containing one line for each building
class, sorted from high to low B/C (or PV/C) ratio,
consisting of a retrofit ranK number, the five BCS
indices, total retrofit cost, net benefit (or HPV), and
B/C (or PV/C) ratio for each prioritized building class.
There is a separate table for each PGA value in the
scenario earthquake event analysis, and one table for the
expected value analysis. The TABLE-BC sub-module writes
this output file in table form; the tables in Appendix C
are examples of the TABLE-BC output.
SYST-EFF Module
Input: The output file from the SORT-BC module.
Processing: This module calculates a running sum of
the retrofit cost and net benefit (or NPV) for each
187
Output: A data file containing one line for each
building class, with the five BCS indices, total retrofit
cost, net present value (HPV) of benefit, and PV/C ratio
for each building class.
SORT-BC Module
Input: The output file from the RETRO-BC or EXP-VAL
modules.
Processing: The Building Classes are prioritized for
retrofit by sorting from high to low B/C (or PV/C) ratio,
using a selection-sort algorithm.
Output: A file containing one line for each building
class, sorted from high to low B/C (or PV/C) ratio,
consisting of a retrofit rank number, the five BCS
indices, total retrofit cost, net benefit (or HPV) , and
B/C (or PV/C) ratio for each prioritized building class.
There is a separate table for each PGA value in the
scenario earthquake event analysis, and one table for the
expected value analysis. The TABLE-BC sub-module writes
this output file in table formj the tables in Appendix C
are examples of the TABLE-BC output.
SYST-EFF Module
Input: The output file from the SORT-BC module.
Processing: This module calculates a running sum of
the retrofit cost and net benefit (or HPV) for each
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
188
prioritized building class.
Output: A file containing one line for each
prioritized building class, consisting of a retrofit ranK
number, the five BCS indices, running sum of retrofit cost
and net benefit (or HPV), and aggregate B/C (or PV/C)
ratio. System Efficiency curves in Chapter IX were
plotted from this file. The TABLE-BC Module will write
this output file in Table form.
188
prioritized building class.
Output: A file containing one line for each
prioritized building class, consisting of a retrofit rank
number, the five BCS indices, running sum of retrofit cost
and net benefit (or HPV) , and aggregate B/C (or PV/C)
ratio. System Efficiency curves in Chapter IX were
plotted from this file. The TABLE-BC Module will write
this output file in Table form.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER VIII
EARTHQUAKE LOSS ESTIMATE OF THE PORTLAND, OREGON QUADRANGLE
The Regional Earthquake Loss and Retrofit Analysis
Program (REAL-RAP) Software, which was presented in
Chapter VII, was utilized to conduct an earthquake loss
estimate of 7,690 non-residential buildings in the
Portland, Oregon Quadrangle.
As previously discussed, REAL-RAP and its
underlying models are designed to analyze a regional
ATC-21 building survey. Therefore, a "test-run" of the
methodology and program on an actual survey database is an
important part of this research.
NON-RESIDENTIAL BUILDING STOCK DATA
The test-run of REAL-RAP was made on data taken from
the 1993 Portland Seismic Hazards Survey (PSHS-93),
conducted by the PSU Civil Engineering Department and the
City of Portland Bureau of Buildings. The survey database
was discussed in detail in Chapter II.
The number of buildings from the survey used for the
analysis are:
CHAPTER VIII
EARTHQUAKE LOSS ESTIMATE OF THE PORTLAND, OREGON QUADRANGLE
The Regional Earthquake Loss and Retrofit Analysis
Program (REAL-RAP) Software, which was presented in
Chapter VII, was utilized to conduct an earthquake loss
estimate of 7,690 non-residential buildings in the
Portland, Oregon Quadrangle.
As previously discussed, REAL-RAP and its
underlying models are designed to analyze a regional
ATC-21 building survey. Therefore, a "test-run" of the
methodology and program on an actual survey database is an
important part of this research.
NON-RESIDENTIAL BUILDING STOCK DATA
The test-run of REAL-RAP was made on data taken from
the 1993 Portland Seismic Hazards Survey (PSHS-93),
conducted by the PSU Civil Engineering Department and the
City of Portland Bureau of Buildings. The survey database
was discussed in detail in Chapter II.
The number of buildings from the survey used for the
analysiS are:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Portland State University Portland Bureau of Bldgs
Total
3, 561 4, 129
7, 690 buildings
Building Type, Area, and Structural Score
190
A summary of the building stock by structure type,
including the number of buildings, square footage, and
average Structural Score (S = BSH + E(PMF) + SMF, using
the ATC-21 Moderate Seismicity Scoring system) is shown in
Table XLVII, following.
TABLE XLVII
NUMBER OF BUILDINGS, AREA, AND AVERAGE SCORE BY BUILDING TYPE
BLDG NO. OF TOTAL AVERAGE TYPE BLDGS AREA SCORE
w 3019 2. 87215E+07 5. 55 S1 44 8838408 1. 55 S2 26 1127057 2. 16 S3 309 3768319 4.62 S4 82 1. 121426E+07 2. 59 C1 159 1. 266035E+07 1. 35 C2 1656 3. 239901E+07 2.46 C3/S5 63 1606882. . 61 PC1 515 1.484026E+07 2. 59 PC2 39 3462993 . 85 RM 916 8016094 2.48 URM 862 1. 544921E+07 . 66
------------TOTAL: 7690 1. 421 044E+08 3. 52
About 12Y. of the buildings are URM and C3/S5, which
are generally known to be high risk buildings in
Portland State University Portland Bureau of Bldgs
Total
3,561 4,129
7,690 buildings
Building Type, Area, and Structural Score
190
A summary of the building stock by structure type,
including the number of buildings, square footage, and
average Structural Score (S = BSH + E(PMF) + SMF, using
the ATC-21 Moderate Seismicity Scoring system) is shown in
Table XLVII, following.
TABLE XLVII
NUMBER OF BUILDINGS, AREA, AND AVERAGE SCORE BY BUILDING TYPE
BLDG NO. OF TOTAL AVERAGE TYPE BLDGS AREA SCORE
W 3019 2.87215E+07 5. 55 S1 44 8838408 1. 55 S2 26 1127057 2.16 S3 309 3768319 4.62 S4 82 1. 121426E+07 2. 59 C1 159 1.266035E+07 1. 35 C2 1656 3. 239901E+07 2.46 C3jS5 63 1606882 .61 PCl 515 1. 484026E+07 2. 59 PC2 39 3462993 .85 RM 916 8016094 2.48 URM 862 1. 544921E+07 .66
------------TOTAL: 7690 1. 421044E+08 3. 52
About 12Y. of the buildings are URM and C3jS5, which
are generally known to be high risK buildings in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
191
earthquakes. Since very few of these buildings have been
retrofit, clearly there is a large hazard in Portland.
The number of buildings in each of five ranges of
Structural Score are shown in Table XLVIII, following.
TABLE XLVIII
NUMBER OF BUILDINGS BY RANGE OF STRUCTURAL SCORE (S)
SCORE RANGE NO. BLDGS DAMAGE RISK
< 0 282 VERY HIGH 0 to 1 725 HIGH 1 to 2 1,054 MODERATE 2 to 3 1, 314 LOW
> 3 4, 315 VERY LOW
Soil Modification Factor
The number of buildings occurring in four ranges of
Soil Modification Factor are shown in Table XLIX.
TABLE XLIX
NUMBER OF BUILDINGS BY RANGE OF SOIL MODIFICATION FACTOR (SMF)
SMF RANGE NO. BLDGS SOIL HAZARD
0 to -. 1 9 LOW -. 2 to -. 5 5,623 MODERATE -. 6 to -.9 1, 070 HIGH
<= -1. 0 988 VERY HIGH
191
earthquakes. Since very few of these buildings have been
retrofit, clearly there is a large hazard in Portland.
The number of buildings in each of five ranges of
Structural Score are shown in Table XLVIII, following.
TABLE XLVIII
HUMBER OF BUILDINGS BY RANGE OF STRUCTURAL SCORE (S)
SCORE RANGE NO. BLDGS DAMAGE RISK
< 0 282 VERY HIGH 0 to 1 725 HIGH 1 to 2 1,054 MODERATE 2 to 3 1,314 LOW
> 3 4,315 VERY LOW
Soil Modification Factor
The number of buildings occurring in four ranges of
Soil Modification Factor are shown in Table XLIX.
TABLE XLIX
HUMBER OF BUILDINGS BY RANGE OF SOIL MODIFICATION FACTOR (SHF)
SHF RANGE NO. BLDGS SOIL HAZARD
0 to -. 1 9 LOW -. 2 to -. 5 5,623 MODERATE -.6 to -.9 1,070 HIGH
<= -1. 0 988 VERY HIGH
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
192
Table XLIX shows that 27Y. of the inventory buildings
are in hazardous soil areas.
Occupancy Type and Number of Occupants
The Lowpoint and Midpoint of the range of number of
people in the buildings have been summed by Occupancy
Type, and presented in Table L, following.
TABLE L
LOWPOINT AND MIDPOINT OF RANGE OF HUMBER OF PEOPLE AND NUMBER OF BUILDINGS BY OCCUPANCY TYPE
RANGE OF NO. PEOPLE OCCUPANCY NO. OF ---------------------
TYPE BLDGS LOWPOINT MIDPOINT
Residential 1,859 42,904 107, 575 Commercial 2,229 41,263 96,595 Office 1,049 56, 340 131,085 Industrial 1,899 26,384 60,965 Pub. Assem. 281 51,714 94,970 School 156 27,240 69,850 Govt. Bldg. 150 11,090 25, 125 Emer. Serv. 36 4,927 11,125 Historic Bldg 31 111 462 17,960
------ ------- -------TOTAL: 7,690 273,324 615,250
The number of people range Midpoint, from the field
survey data, represents the field inspectors judgement on
the number of people that are likely to be in a building
when it is in "full use." Since it is not probable that
all the buildings in the inventory will at full use at the
192
Table XLIX shows that 27Y. of the inventory buildings
are in hazardous soil areas.
Occupancy Type and Number of Occupants
The Lowpoint and Midpoint of the range of number of
people in the buildings have been summed by Occupancy
Type, and presented in Table L, following.
TABLE L
LOWPOINT AND MIDPOINT OF RANGE OF NUMBER OF PEOPLE AND NUMBER OF BUILDINGS BY OCCUPANCY TYPE
RANGE OF NO. PEOPLE OCCUPANCY NO. OF ---------------------
TYPE BLDGS LOWPOINT MIDPOINT
Residential 1,859 4-2,904- 107,575 Commercial 2,229 4-1,263 96,595 Office 1,04-9 56, 340 131,085 Industrial 1,899 26,384- 60,965 Pub. Assem. 281 51,714- 94-,970 School 156 27,240 69,850 Govt. Bldg. 150 11,090 25, 125 EDler. Servo 36 4-,927 11,125 Historic Bldg 31 11,462 17,960
------ ------- -------TOTAL: 7,690 273,324 615,250
The number of people range MidpOint, from the field
survey data, represents the field inspectors judgement on
the number of people that are likely to be in a building
when it is in "full use." Since it is not probable that
all the buildings in the inventory will at full use at the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
193
same time, the Lowpoint is useful in studying an aggregate
value of occupancy at a lower level.
Judgement points in the direction of looKing at a
number of people ranging, at "full use," somewhere between
the Lowpoint and Midpoint values.
LOSS ESTIMATES FOR FOUR SCENARIO EARTHQUAKE EVENTS
Losses for the 7,690 non-residential buildings in
the Portland Quadrangle were determined by REAL-RAP for
six values of PGA, four of which are presented in Tables
LI through LIV, following, The PGA values input and
corresponding return periods are:
PGA RETURN
. 08 100 year
. 20 500 year (DBE)
. 27 1,000 year
. 39 2, 500 year
"Scenario Earthquak.e Event" means that the losses
shown are the full losses from the model for that PGA.
The return times are not used in this estimate, and are
shown here for reference only.
The average damage factor, dollar value of building
damage, and loss of life is shown by Building Type for
each PGA event. The loss of life shown was determined from
the Midpoint of the range of Humber of People from the
inventory.
193
same time, the Lowpoint is useful in studying an aggregate
value of occupancy at a lower level.
Judgement pOints in the direction of looKing at a
number of people ranging, at "full use," somewhere between
the Lowpoint and Midpoint values.
LOSS ESTIMATES FOR FOUR SCENARIO EARTHQUAKE EVENTS
Losses for the 7,690 non-residential buildings in
the Portland Quadrangle were determined by REAL-RAP for
six values of PGA, four of which are presented in Tables
LI through LIV, following. The PGA values input and
corresponding return periods are:
pGA RETURN
.08 100 year
.20 500 year (DBE)
.27 1,000 year
.39 2, 500 year
"Scenario Earthquake Event" means that the losses
shown are the full losses from the model for that PGA.
The return times are not used in this estimate, and are
shown here for reference only.
The average damage factor, dollar value of building
damage, and loss of life is shown by Building Type for
each PGA event. The loss of life shown was determined from
the Midpoint of the range of Number of People from the
inventory.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LI
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA = .08 (100 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 3. 59 $ 8.612823E+07 1.0
S1 4.99 $ 3.687408E+07 . 6
S2 2.97 $ 2325002 . 1
S3 2.2 $ 5219876 . 1
S4 6. 11 $ 5.246866E+07 .9
C1 14. 6 $ 1.577566E+08 1480. 3
C2 5. 74 $ 1. 4477 49E+08 14. 3
C3/S5 23. 5 $ 4.725515E+07 748. 1
PC1 10.62 $ 1. 164073E+08 253.0
PC2 7.99 $ 3.070333E+07 12.8
RM 6.73 $ 6.021686E+07 324.4
URM 31 $ 4. 58262E+08 3009. 1
TOTAL: 8. 36 $ 1. 198392E+09 5844.7
194
TABLE LI
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA = .08 (100 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 3. 59 $ 8. 612823E+07 1.0
Sl 4.99 $ 3. 687408E+07 .6
S2 2.97 $ 2325002 . 1
S3 2.2 $ 5219876 . 1
S4 6. 11 $ 5. 246866E+07 .9
C1 14. 6 $ 1. 577566E+08 1480. 3
C2 5. 74 $ 1. 447749E+08 14. 3
C3/S5 23. 5 $ 4. 725515E+07 748. 1
PCl 10.62 $ 1. 164073E+08 253.0
PC2 7.99 $ 3.070333E+07 12.8
RM 6.73 $ 6.021686E+07 324.4
URM 31 $ 4. 58262E+08 3009. 1
TOTAL: 8. 36 $ 1. 198392E+09 5844.7
194
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA: .20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 7.84 $ 1.885347E+08 1.7
S1 13. 69 $ 1. 136741E+08 245. 3
S2 6.68 $ 5515891 0 1
S3 5. 16 $ 1.221718E+07 . 1
S4 14. 19 $ 1. 212542E+08 2. 1
C1 42.85 $ 4. 536644E+08 2956. 3
C2 13.78 $ 3. 715052E+08 274. 5
C3/S5 67.05 $ 9.602283E+07 1937. 5
PC1 25.29 $ 2.73.1!-362E+08 281. 7
PC2 22.47 $ 7. 21!4E+07 16.7
RM 16.78 $ 1. 169726E+08 282. 1
URM 69. 75 $ 8.686221E+08 6998.4
TOTAL: 19. 57 $ 2.693859E+09 12996. 5
195
TABLE LII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA = .20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 7.84 $ 1. 885347E+08 1.7
Sl 13.69 $ 1. 136741E+08 245. 3
S2 6.68 $ 5515891 . 1
S3 5.16 $ 1. 221718E+07 . 1
S4 14.19 $ 1. 212542E+08 2. 1
C1 42.85 $ 4. 536644E+08 2956. 3
C2 13.78 $ 3.715052E+08 274. 5
C3/S5 67.05 $ 9.602283E+07 1937.5
PCl 25.29 $ 2. 73l!-362E+08 281. 7
PC2 22.47 $ 7. 244E+07 16.7
RM 16.78 $ 1. 169726E+08 282.1
URM 69. 75 $ 8. 686221E+08 6998.4
TOTAL: 19.57 $ 2. 693859E+09 12996.5
195
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LIII
SUMMARY OF LOSSES FOR THE PORTLAHD QUAD AT PGA = . 27 (1,000 YEAR RETURH)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 9.859999 $ 2. 373614E+08 2.9
S1 24. 91 $ 2. 159162E+08 562. 3
S2 8. 399999 $ 7023393 . 1
S3 6.74 $ 1. 59657E+07 . 1
S4 17.05 $ 1.441428E+08 3.0
C1 56.23 $ 6. 180308E+08 4995. 8
C2 18.43 $ 4.986351E+08 312. 1
C3/S5 74.61 $ 1.005628E+08 1457. 3
PC1 33. 22 $ 3.620145E+08 630. 6
PC2 40. 14 $ 1. 392571E+08 137.6
RM 24.02 $ 1. 677315E+08 600. 8
URM 79.84 $ 9.71185E+08 9279. 2
TOTAL: 24.48 $ 3.477826E+09 17981. 8
196
TABLE LIII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA = .27 (1,000 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 9.859999 $ 2. 373614E+08 2.9
S1 24. 91 $ 2.159162E+08 562. 3
S2 8. 399999 $ 7023393 . 1
S3 6.74 $ 1. 59657E+07 . 1
S4 17.05 $ 1. 441428E+08 3. 0
C1 56.23 $ 6.180308E+08 4995. 8
C2 18.43 $ 4. 986351E+08 312. 1
C3/S5 74.61 $ 1.005628E+08 1457. 3
PC1 33. 22 $ 3. 620145E+08 630. 6
PC2 40. 14 $ 1. 392571E+08 137.6
RM 24.02 $ 1. 677315E+08 600. 8
URM 79.84 $ 9. 71185E+08 9279. 2
TOTAL: 24.48 $ 3. 477826E+09 17981. 8
196
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LIV
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA: . 39 (2,500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 12. 91 $ 3. 116605E+08 142.7
S1 31. 66 $ 2.827823E+08 808. 8
S2 10.66 $ 9112080 . 1
S3 9. 310001 $ 2. 207283E+07 . 1
S4 22.7 $ 1. 892474E+08 8.6
C1 71. 75 $ 7.405977E+08 6361. 9
C2 27. 58 $ 7. 669451E+08 918.6
C3/S5 84. 61 $ 1. 106024E+08 2124. 8
PC1 42.91 $ 4. 678059E+08 931. 0
PC2 61. 93 $ 1. 763162E+08 375. 6
RM 36.44 $ 2.435513E+08 1624.9
URM 87. 14 $ 1. 037381E+09 9460. 2
TOTAL: 31. 31 $ 4. 358074E+09 22757. 3
197
TABLE LIV
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA = .39 (2,500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 12. 91 $ 3. 116605E+08 142.7
S1 31. 66 $ 2. 827823E+08 808. 8
S2 10.66 $ 9112080 . 1
S3 9. 310001 $ 2. 207283E+07 . 1
S4 22.7 $ 1.892474E+08 8. 6
C1 71. 75 $ 7.405977E+08 6361. 9
C2 27. 58 $ 7.669451E+08 918. 6
C3/S5 84. 61 $ 1. 106024E+08 2124. 8
PC1 42.91 $ 4. 678059E+08 931. 0
PC2 61. 93 $ 1. 763162E+08 375. 6
RM 36.44 $ 2. 435513E+08 1624.9
URM 87. 14 $ 1. 037381E+09 9460. 2
TOTAL: 31. 31 $ 4. 358074E+09 22757. 3
197
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
198
LOSS ESTIMATES FOR THE DESIGN BASIS EARTHQUAKE
The Design Basis EarthquaKe (DBE) is defined as the
PGA with a 10Y. chance of exceedance in 50 years. This
definition has become the national standard for
establishing design criteria for new structures (Building
Seismic Safety Council 1988). Therefore, the losses found
for the DBE are the most important of the earthquake
losses determined in this study.
Damage and Loss of Life by Structure Type
The loss summary by structure type for the DBE, of
PGA = . 20 (Table LII), is repeated below for the
convenience of the reader.
As previously discussed, these losses were
determined using the Midpoint of the Number of People
range, considering all the buildings being in "full use."
For this inventory, "full use" would be during regular
business hours, Monday - Friday, 8:00 to 5:00.
198
LOSS ESTIMATES FOR THE DESIGN BASIS EARTHQUAKE
The Design Basis Earthquake (DBE) is defined as the
PGA with a 101. chance of exceedance in 50 years. This
definition has become the national standard for
establishing design criteria for new structures (Building
Seismic Safety Council 1988). Therefore, the losses found
for the DBE are the most important of the earthquake
losses determined in this study.
Damage and Loss of Life by Structure TyPe
The loss summary by structure type for the DBE, of
PGA = .20 (Table LII), is repeated below for the
convenience of the reader.
As previously discussed, these losses were
determined using the Midpoint of the Number of People
range, considering all the buildings being in "full use."
For this inventory, "full use" would be during regular
business hours, Monday - Friday, 8:00 to 5:00.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA: .20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 7. 84 $ 1.885347E+08 1.7 S1 13. 69 $ 1. 136741E+08 245. 3 S2 6. 68 $ 5515891 . 1 S3 5. 16 $ 1. 221718E+07 . 1 S4 14. 19 $ 1.212542E+08 2. 1 C1 42.85 $ 4. 536644E+08 2956. 3 C2 13.78 $ 3.715052E+08 274. 5 C3/S5 67.05 $ 9.602283E+07 1937. 5 PC1 25.29 $ 2. 734362E+08 281. 7 PC2 22.47 $ 7. 244E+07 16.7 RM 16. 78 $ 1. 169726E+08 282. 1 URM 69.75 $ 8.686221E+08 6998.4
TOTAL: 19. 57 $ 2.693859E+09 12996. 5
Table LII shows that most of the loss of life is
attributable to three building types, as follows:
LOSS OF Y. OF LIFE TOTAL
URM 6,998 53. 8Y. C1 2,956 22.7Y. C3/S5 1,937 14. 9Y.
------- ------11, 891 91. 4Y.
199
TABLE LII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD AT PGA : .20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 7. 84 $ 1. 885347E+08 1.7 S1 13.69 $ 1. 136741E+08 245. 3 S2 6. 68 $ 5515891 . 1 S3 5.16 $ 1. 221718E+07 . 1 S4 14.19 $ 1. 212542E+OB 2. 1 C1 42.B5 $ 4. 536644E+OB 2956. 3 C2 13.7B $ 3. 715052E+OB 274. 5 C3/S5 67.05 $ 9.602283E+07 1937. 5 PCl 25.29 $ 2. 734362E+OB 2B1. 7 PC2 22.47 $ 7. 244E+07 16.7 RH 16. 7B $ 1. 169726E+OB 282. 1 URH 69.75 $ B.686221E+OB 6998.4
TOTAL: 19. 57 $ 2. 693859E+09 12996. 5
Table LII shows that most of the loss of life is
attributable to three building types, as follows:
LOSS OF X OF LIFE TOTAL
URM 6,998 53. 8X Ci 2.956 22. 7X C3/S5 1,937 14. 9Y.
------- ------11.891 91. 4X
199
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
200
Total Dollar Value of Loss
The Value of Life (VOL) adopted for use in Chapter
VI, which also accounts for accompanying serious injuries,
was$ 2,250,000. Based on this, the total dollar value of
building damage, loss of life, and serious injury for the
DBE is:
Damage, from Table VIII-VI = $ 2.69 G ( 8.4:t.)
(91. 6:t.) Casualties: 12,996 w $ 2. 25 M = $ 29.24 G
$ 31.93 G (100:t.)
Since more than 90:t. of the value of the loss is from
the loss of life, loss of life can be expected to be the
critical factor in retrofit planning.
Loss of Life by Structural Score
The loss of life by Structural Score range is shown
in Table LV, following:
TABLE LV
LOSS OF LIFE BY RANGE OF STRUCTURAL SCORE (S)
STRUCTURAL LOSS OF LIFE SCORE RANGE (MIDPOINT)
< 0 7,973 0 to 1 5,006 1 to 2 10 2 to 3 3
> 3 3
DAMAGE RISK
VERY HIGH HIGH MODERATE LOW VERY LOW
200
Total Dollar Value of Loss
The Value of Life (VOL) adopted for use in Chapter
VI, which also accounts for accompanying serious injuries,
was $ 2,250,000. Based on this, the total dollar value of
building damage, loss of life, and serious injury for the
DBE is:
Damage, from Table VIII-VI = $ 2.69 G ( 8.4:1.)
(91. 6:1.) Casualties: 12,996 W $ 2.25 M = $ 29.24 G
$ 31.93 G (100:1.)
Since more than 90:1. of the value of the loss is from
the loss of life, loss of life can be expected to be the
critical factor in retrofit planning.
Loss of Life by Structural Score
The loss of life by Structural Score range is shown
in Table LV, following:
TABLE LV
LOSS OF LIFE BY RANGE OF STRUCTURAL SCORE (S)
STRUCTURAL LOSS OF LIFE SCORE RANGE (MIDPOINT)
< 0 7,973 0 to 1 5,006 1 to 2 10 2 to 3 3
> 3 3
DAMAGE RISK
VERY HIGH HIGH MODERATE LOW VERY LOW
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table LV shows clearly that virtually 1001. of
the loss of life occurs in buildings with a Structural
Score of 1 or less.
Loss of Life by Soil Modification Factor
The loss of life occurring in four ranges of Soil
Modification Factor are shown in Table LVI, following:
TABLE LVI
LOSS OF LIFE BY RANGE OF SOIL MODIFICATION FACTOR (SMF)
LOSS OF LIFE SOIL SMF RANGE (MIDPOINT) HAZARD
0 to -. 1 0 LOW -. 2 to -. 5 7, 542 LOW -. 6 to -. 9 3,651 HIGH
<= -1. 0 1,802 HIGH
Table LVI appears to show that soil conditions do
201
not have a statistically large impact on the distribution
of loss of life in the Portland Quad.
Loss of Life by Humber of People
As previously mentioned, the loss of life in Table
LII was determined using the Midpoint of the Humber of
People Range.
To study the effect on the loss of life of the
surveyed range of number of people, REAL-RAP was re-run
Table LV shows clearly that virtually 1001. of
the loss of life occurs in buildings with a Structural
Score of 1 or less,
Loss of Life by Soil Modification Factor
The loss of life occurring in four ranges of Soil
Modification Factor are shown in Table LVI, following:
TABLE LVI
LOSS OF LIFE BY RANGE OF SOIL MODIFICATION FACTOR (SMF)
LOSS OF LIFE SOIL SMF RANGE (MIDPOINT) HAZARD
0 to -, 1 0 LOW -, 2 to -, 5 7,542 LOW -, 6 to -, 9 3,651 HIGH
<= -1. 0 1,802 HIGH
Table LVI appears to show that soil conditions do
201
not have a statistically large impact on the distribution
of loss of life in the Portland Quad,
Loss of Life by Humber of People
As previously mentioned, the loss of life in Table
LII was determined using the Midpoint of the Humber of
Peopl e Range,
To study the effect on the loss of life of the
surveyed range of number of people, REAL-RAP was re-run
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
202
for the inventory using the Lowpoint Humber of the range.
The loss of life for both the Lowpoint and Midpoint is
shown by Occupancy Type in Table LVII, following.
TABLE LVII
LOSS OF LIFE BY LOWPOINT AND MIDPOINT OF NUMBER OF PEOPLE RANGE, BY OCCUPANCY TYPE
LOSS OF LIFE FOR OCCUPANCY ---------------------
TYPE LOWPOINT MIDPOINT
Residential 661 1,734 Commercial 1,306 2,902 Office 1,264 3,229 Industrial 384 898 Pub. Ass em. 412 1,042 School 760 1,914 Govt. Bldg. 178 467 Emer. Serv. 104 257 Historic Bldg 225 552
------- -------TOTAL: 5,294 12,995
Comparison of Tables LII and LVII shows that
the loss of life is proportionate to the number of people
occupying the buildings.
Based on the preceding discussion on the range of the
total number of people, it is concluded that the
methodology estimates a range of loss of life: from 5, 294
to 12, 995.
202
for the inventory using the Lowpoint Number of the range.
The loss of life for both the Lowpoint and Midpoint is
shown by Occupancy Type in Table LVII, following.
TABLE LVII
LOSS OF LIFE BY LOWPOINT AND MIDPOINT OF NUMBER OF PEOPLE RANGE, BY OCCUPANCY TYPE
LOSS OF LIFE FOR OCCUPANCY ---------------------
TYPE LOWPOINT MIDPOINT
Residential 661 1,734 Commercial 1,306 2,902 Office 1,264 3, 229 Industrial 384 898 Pub. Assem. 412 1,042 School 760 1,914 Govt. Bldg. 178 467 EDler. Servo 104 257 Historic Bldg 225 552
------- -------TOTAL: 5,294 12,995
Comparison of Tables LII and LVII shows that
the loss of life is proportionate to the number of people
occupying the buildings.
Based on the preceding discussion on the range of the
total number of people, it is concluded that the
methodology estimates a range of loss of life: from 5,294
to 12,995.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Damage State by Building Type
The number of buildings in two ranges of Damage
State are presented by Building Type in Table LVIII.
The
these two
DF = DF = DF =
TABLE LVIII
HUMBER OF BUILDINGS IN 2 DAMAGE STATES BY BUILDING TYPE
DAMAGE FACTOR (DF) BLDG ----------------------TYPE 100Y. 50Y. TO 99Y.
w 0 0 S1 1 0 S2 0 0 S3 0 0 S4 0 0 C1 25 21 C2 3 4 C3/S5 31 15 PC1 20 40 PC2 1 4 RM 17 26 URM 410 182
TOTAL: 508 292
percentage of buildings in the inventory
damage states is:
100Y. 6. 6Y. of inventory
50Y. to 99Y. 3. 8 ------
50Y. or more 10. 2Y. of inventory
in
203
This percentage calculation helps clarify the fact
that virtually 100Y. of the loss of life (which occurs in
Damage State by Building Type
The number of buildings in two ranges of Damage
State are presented by BUilding Type in Table LVIII.
The
these two
DF =
DF =
DF =
TABLE LVIII
HUMBER OF BUILDINGS IN 2 DAMAGE STATES BY BUILDING TYPE
DAMAGE FACTOR (DF) BLDG ----------------------TYPE 100Y. 50Y. TO 991.
W 0 0 Sl 1 0 S2 0 0 S3 0 0 S4 0 0 Cl 25 21 C2 3 4 C3/S5 31 15 PCl 20 40 PC2 1 4 RM 17 26 URM 410 182
TOTAL: 508 292
percentage of buildings in the inventory
damage states 1s:
100Y. 6. 6Y. of inventory
50Y. to 99Y. 3. 8 ------
50Y. or more 10. 2Y. of inventory
in
203
This percentage calculation helps clarify the fact
that virtually 100Y. of the loss of life (which occurs in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
204
major damage states, near DF = 100) in the Portland Quad
inventory is taking place in about 10X of the buildings.
Comparison to Kobe City Earthquake of 1995
The 1995 Kobe City earthquaKe has potential for
comparison to a DBE earthquaKe in Portland, owing to the
following similarities:
1. Kobe and Portland are cities of comparable size.
2. The 1995 event was considered a "moderate"
earthquaKe, Richter Magnitude 6. 8
3. Both Kobe and Portland are "modern" cities, but
historically lacKing in stringent seismic design
requirements for buildings such as would be found
in California.
A comparison of buildings in major damage states from
the present study, with that reported by Comartin et al
(1995) for Kobe, is summarized in Table LIX, following.
TABLE LIX
DAMAGE STATE COMPARISON TO KOBE CITY 1995 EARTHQUAKE
DAMAGE STATE
100X 50X to 99X
50X or more
X OF BLDGS IH DAMAGE STATE
Table VIII-XII
6.61. 3. 8X
10. 21.
Kobe City
4. 9Y. 7.7X
12.6X
204
major damage states, near DF = 100) in the Portland Quad
inventory is taking place in about 10X of the buildings.
Comparison to Kobe City Earthquake of 1995
The 1995 Kobe City earthquake has potential for
comparison to a DBE earthquake in Portland, owing to the
following similarities:
1. Kobe and Portland are cities of comparable size.
2. The 1995 event was considered a "moderate"
earthquake, Richter Magnitude 6.8
3. Both Kobe and Portland are "modern" cities, but
historically lacKing in stringent seismic design
requirements for buildings such as would be found
in California.
A comparison of buildings in major damage states from
the present study, with that reported by Comartin et al
(1995) for Kobe, is summarized in Table LIX, following.
TABLE LIX
DAMAGE STATE COMPARISON TO KOBE CITY 1995 EARTHQUAKE
DAMAGE STATE
100X 50X to 99X
50X or more
X OF BLDGS IN DAMAGE STATE
Table VIII-XII
6.61. 3. 8X
10. 21.
Kobe City
4.91. 7.71.
12.61.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
205
Table LIX indicates that the percentage of buildings
shown by REAL-RAP to be in major damage states is
comparable to the Kobe City earthquaKe.
Losses Attributable to the Performance Modifiers
In order to study the e£fect of the Per£ormance
Modifiers, a version of REAL-RAP was run which used the
SMF as usual, but set E(PMF) = 0 £or all buildings in the
inventory. The ef£ect of the Performance Modifiers is
shown in Table LX, £ollowing.
TABLE LX BUILDING DAMAGE AND LOSS OF LIFE ATTRIBUTABLE
TO PERFORMANCE MODIFIERS
AVERAGE TOT BLDG LOSS OF DF DAMAGE LIFE
From Table VIII-VI 19. 57 $ 2. 69 G 12,996 With E(PMF) = 0 11.85 $ 1. 39 G ll-77
------ --------- -------Attributed to E(PMF) 7.72. $ 1. 30 G 12, 519
Table LX shows that loss estimates which are
based on "average" structures only, without Performance
Modifiers taKen into account, seriously underestimate the
losses, especially the loss o£ li£e.
205
Table LIX indicates that the percentage of buildings
shown by REAL-RAP to be in major damage states is
comparable to the Kobe City earthquaKe.
Losses Attributable to the Performance Modifiers
In order to study the effect of the Performance
Modifiers, a version of REAL-RAP was run which used the
SMF as usual, but set E(PMF) = 0 for all buildings in the
inventory. The effect of the Performance Modifiers is
shown in Table LX, following.
TABLE LX BUILDING DAMAGE AND LOSS OF LIFE ATTRIBUTABLE
TO PERFORMANCE MODIFIERS
AVERAGE TOT BLDG LOSS OF DF DAMAGE LIFE
From Table VIII-VI 19. 57 $ 2. 69 G 12,996 With E (PMF) = 0 11.85 $ 1.39 G 477
------ --------- -------Attributed to E(PMF) 7.72. $ 1. 30 G 12,519
Table LX shows that loss estimates which are
based on "average" structures only, without Performance
Modifiers taKen into account, seriously underestimate the
losses, especially the loss of life.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER IX
RETROFIT ANALYSIS OF THE PORTLAND, OREGON QUADRANGLE
The Regional Earthquake Loss and Retrofit Analysis
Program (REAL-RAP) was utilized to conduct a retrofit
analysis of the 7,690 buildings in the Portland, Oregon
Quadrangle. The building inventory used and losses to be
mitigated are the same as those presented previously in
Chapter VIII, using the midpoint of the number of people
range (except as noted in the following).
Five retrofit analyses were made in this study:
1. For four Scenario Earthquake Events, the
reduced losses resulting from a Full Retrofit
(where all buildings in the inventory are
retrofit) are analyzed and presented.
2. For a Design Basis Earthquake, a cost-benefit
and system efficiency analysis was made, to
determine an Optimal Retrofit Program. This is
then compared to the Full Retrofit.
3. The sensitivity of the optimal retrofit program
to the number of people and to the dollar
value of life was analyzed. This was done by
determining two alternate optimal retrofit
programs: 1) using the lowpoint of the number
CHAPTER IX
RETROFIT ANALYSIS OF THE PORTLAND, OREGON QUADRANGLE
The Regional EarthquaKe Loss and Retrofit Analysis
Program (REAL-RAP) was utilized to conduct a retrofit
analysis of the 7,690 buildings in the Portland, Oregon
Quadrangle. The building inventory used and losses to be
mitigated are the same as those presented previously in
Chapter VIII, using the midpoint of the number of people
range (except as noted in the following).
Five retrofit analyses were made in this study:
1. For four Scenario EarthquaKe Events, the
reduced losses resulting from a Full Retrofit
(where all buildings in the inventory are
retrofit) are analyzed and presented.
2. For a Design Basis EarthquaKe, a cost-benefit
and system efficiency analysis was made, to
determine an Optimal Retrofit Program. This is
then compared to the Full Retrofit.
3. The sensitivity of the optimal retrofit program
to the number of people and to the dollar
value of life was analyzed. This was done by
determining two alternate optimal retrofit
programs: 1) using the lowpoint of the number
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
207
of people range, and 2) using about 25/. of the
dollar value of life.
4. An Expected Value cost-benefit and system
efficiency analysis was made, as an alternate
determination of an Optimal Retrofit Program.
This program is compared to the DBE-based
Optimal Retrofit Program.
POST-RETROFIT EARTHQUAKE LOSS ESTIMATES FOR FOUR SCENARIO EARTHQUAKE EVEHTS
FOLLOWING FULL RETROFIT
Post-retrofit losses were determined for the
Portland Quad inventory, based on 100/. of the buildings
having been retrofit. This is not an "efficient" retrofit
program, but is an excercise to determine the maximum
mitigation attainable, according to the retrofit model.
The same four PGA scenarios run in Chapter VIII are
presented for the post-retrofit buildings, in Tables LXI
through LXIV, following. The average damage factor,
dollar value of building damage, and loss of life is shown
by building type for each PGA event. The loss of life
shown was determined from the Midpoint of the range of
Humber of People from the inventory.
207
of people range, and 2) using about 251. of the
dollar value of life.
4. An Expected Value cost-benefit and system
efficiency analysis was made, as an alternate
determination of an Optimal Retrofit Program.
This program is compared to the DBE-based
Optimal Retrofit Program.
POST-RETROFIT EARTHQUAKE LOSS ESTIMATES FOR FOUR SCENARIO EARTHQUAKE EVENTS
FOLLOWING FULL RETROFIT
Post-retrofit losses were determined for the
Portland Quad inventory, based on 1001. of the buildings
having been retrofit. This is not an "efficient" retrofit
program, but is an excercise to determine the maximum
mitigation attainable, according to the retrofit model.
The same four PGA scenariOS run in Chapter VIII are
presented for the post-retrofit buildings, in Tables LXI
through LXIV, following. The average damage factor,
dollar value of building damage, and loss of life is shown
by building type for each PGA event. The loss of life
shown was determined from the Midpoint of the range of
Number of People from the inventory.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXI
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = .08 (100 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 2.59 $ 6.096544E+07 .9
S1 3. 3 $ 2.430852E+07 .5
S2 1. 96 $ 1581728 .o S3 1. 66 $ 3910518 . 1
S4 4. 38 $ 3. 848339E+07 .8
C1 9.7 $ 1.052362E+08 40.9
C2 3.67 $ 9. 203788E+07 1.5
C3/S5 14. 19 $ 2. 852689E+07 12.4
PC1 6.4 $ 6. 690674E+07 1.7
PC2 5. 13 $ 1. 924749E+07 .3
RM 4.86 $ 4. 307256E+07 15. 2
URM 15.7 $ 2. 316E+08 18. 5
TOTAL: 5.06 $ 7. 158775E+08 92. 8
208
TABLE LXI
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = .08 (100 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 2.59 $ 6.096544E+07 .9
S1 3. 3 $ 2. 430852E+07 .5
S2 1. 96 $ 1581728 .0
S3 1. 66 $ 3910518 . 1
S4 4. 38 $ 3. 848339E+07 .8
C1 9.7 $ 1.052362E+08 40.9
C2 3.67 $ 9. 203788E+07 1.5
C3/S5 14.19 $ 2. 852689E+07 12.4
PC1 6.4 $ 6. 690674E+07 1.7
PC2 5.13 $ 1. 924749E+07 .3
RM 4.86 $ 4. 307256E+07 15.2
URM 15.7 $ 2. 316E+08 18. 5
TOTAL: 5.06 $ 7. 158775E+08 92. 8
208
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = . 20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 5. 68 $ 1. 337343E+08 1.2
S1 9. 399999 $ 7. 760756E+07 9.4
S2 4. 62 $ 4027448 . 1
S3 3. 74 $ 8810041 . 1
S4 10. 21 $ 8. 830522E+07 1.4
C1 29. 29 $ 3. 1122E+08 111. 4
C2 9 $ 2.408885E+08 8. 1
C3/S5 42. 16 $ 6.037205E+07 40. 9
PC1 15, 21 $ 1. 591369E+08 3. 3
PC2 14.62 $ 4. 671833E+07 . 6
RM 12. 36 $ 8. 614131E+07 18. 6
URM 37.61 $ 4. 684373E+08 61. 4
TOTAL: 12. 23 $ 1. 685399E+09 256. 5
209
TABLE LXII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA : .20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 5. 68 $ 1.33734-3E+08 1.2
S1 9. 399999 $ 7.760756E+07 9.4-
S2 4.62 $ 4027448 . 1
S3 3. 74 $ 8810041 . 1
S4 10. 21 $ 8. 830522E+07 1.4-
C1 29. 29 $ 3. 1122E+08 111.4
C2 9 $ 2.4-08885E+08 8. 1
C3/S5 42. 16 $ 6.037205E+07 40. 9
PC1 15,21 $ 1.591369E+08 3. 3
PC2 14.62 $ 4.671833E+07 .6
RM 12. 36 $ 8.614131E+07 18. 6
URM 37.61 $ 4. 684373E+08 61. 4-
TOTAL: 12. 23 $ 1.685399E+09 256. 5
209
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXIII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = . 27 (1,000 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 7. 16 $ 1. 686144E+08 1.5
S1 17. 37 $ 1.497794E+08 25.9
S2 5. 93 $ 5263660 . 1
S3 4.77 $ 1. 123457E+07 . 1
S4 12.49 $ 1. 077262E+08 1.8
C1 39.08 $ 4. 304779E+08 206.0
C2 12. 13 $ 3.265227E+08 11. 3
C3/S5 47. 88 $ 6.45393E+07 38. 2
PC1 20. 13 $ 2. 136936E+08 7. 1
PC2 26.09 $ 9.021271E+07 3. 7
RM 17. 96 $ 1. 25342E+08 43. 2
URM 44. 53 $ 5.416366E+08 94.4
TOTAL: 15. 68 $ 2. 235043E+09 433. 3
210
TABLE LXIII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = .27 (1,000 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 7.16 $ 1. 686144E+08 1.5
Sl 17.37 $ 1. 497794E+08 25.9
S2 5. 93 $ 5263660 . 1
S3 4. 77 $ 1. 123457E+07 . 1
S4 12.49 $ 1. 077262E+OB 1.8
Cl 39.08 $ 4. 304779E+08 206.0
C2 12. 13 $ 3. 265227E+08 11. 3
C3/S5 47. 88 $ 6. 45393E+07 38. 2
PCl 20. 13 $ 2.136936E+08 7. 1
PC2 26.09 $ 9.021271E+07 3. 7
RM 17.96 $ 1. 25342E+08 43. 2
URM 44. 53 $ 5. 416366E+08 94.4
TOTAL: 15.68 $ 2. 235043E+09 433. 3
210
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXIV
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = . 39 (2, 500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 9. 359999 $ 2. 205176E+08 5. 4
S1 22. 74 $ 2.019095E+08 54.7
S2 7. 82 $ 7105022 . 1
S3 6.4 $ 1. 512223E+07 . 1
S4 16.86 $ 1. 428168E+08 3. 2
C1 51.24 $ 5.299333E+08 323. 7
C2 18.49 $ 5. 129665E+08 38.2
C3/S5 55. 97 $ 7. 316269E+07 63.0
PC1 26.65 $ 2. 84768E+08 13. 7
PC2 41. 19 $ 1. 172277E+08 12. 3
RM 27. 89 $ 1.863266E+08 139.7
URM 51. 01 $ 6.072673E+08 131. 6
TOTAL: 20. 8 $ 2. 899124E+09 785. 7
211
TABLE LXIV
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = . 39 (2, 500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 9. 359999 $ 2.205176E+08 5.4
S1 22. 74 $ 2.019095E+08 54.7
S2 7. 82 $ 7105022 . 1
S3 6.4 $ 1. 512223E+07 . 1
S4 16.86 $ 1. 428168E+08 3. 2
C1 51.24 $ 5. 299333E+08 323. 7
C2 18.49 $ 5. 129665E+08 38.2
C3/S5 55. 97 $ 7.316269E+07 63.0
PC1 26.65 $ 2. 84768E+08 13.7
PC2 41.19 $ 1.172277E+08 12.3
RM 27. 89 $ 1. 863266E+08 139.7
URM 51. 01 $ 6.072673E+08 131. 6
TOTAL: 20. 8 $ 2. 899124E+09 785. 7
211
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEVELOPMENT OF AN OPTIMAL RETROFIT PROGRAM FOR THE DESIGN BASIS EARTHQUAKE
212
The Design Basis EarthquaKe (DBE) is defined as the
PGA with a 10Y. chance of exceedance in 50 years. This is
the same as a return of 500 years, which is PGA = . 20.
The summary of losses at PGA = .20 if all the buildings
are retrofit (Table LXII) is repeated below.
TABLE LXII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = . 20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
w 5. 68 $ 1. 337343E+08 1.2 S1 9. 399999 $ 7. 760756E+07 9.4 S2 4. 62 $ 4027448 . 1 S3 3. 74 $ 8810041 . 1 S4 10. 21 $ 8. 830522E+07 1.4 C1 29.29 $ 3. 1122E+08 111. 4 C2 9 $ 2.408885E+08 8. 1 C3/S5 42. 16 $ 6.037205E+07 40.9 PC1 15.21 $ 1. 591369E+08 3. 3 PC2 14.62 $ 4.671833E+07 . 6 RM 12. 36 $ 8.614131E+07 18.6 URM 37.61 $ 4.684373E+08 61.4
TOTAL: 12. 23 $ 1. 685399E+09 256.4
Total Value of Retrofit Benefits
The effectiveness of retrofitting all buildings is
summarized in Table LXV, following.
DEVELOPMENT OF AN OPTIMAL RETROFIT PROGRAM FOR THE DESIGN BASIS EARTHQUAKE
212
The Design Basis EarthquaKe (DBE) is defined as the
PGA with a 10Y. chance of exceedance in 50 years. This is
the same as a return of 500 years, which is PGA = . 20.
The summary of losses at PGA = .20 if all the buildings
are retrofit (Table LXII) is repeated below.
TABLE LXII
SUMMARY OF LOSSES FOR THE PORTLAND QUAD IF ALL BLDGS ARE RETROFIT AT PGA = .20 (500 YEAR RETURN)
BLDG AVERAGE TOTAL TOT LOSS TYPE DF DAMAGE OF LIFE
W 5. 68 $ 1. 337343E+08 1.2 S1 9. 399999 $ 7. 760756E+07 9.4 S2 4. 62 $ 4027448 . 1 S3 3. 74 $ 8810041 . 1 S4 10. 21 $ 8. 830522E+07 1.4 C1 29.29 $ 3. 1122E+08 111.4 C2 9 $ 2.408885E+08 8. 1 C3/S5 42. 16 $ 6.037205E+07 40.9 PC1 15.21 $ 1.591369E+08 3. 3 PC2 14.62 $ 4. 671833E+07 .6 RM 12. 36 $ 8. 614131E+07 18.6 URM 37.61 $ 4. 684373E+08 61. 4
TOTAL: 12. 23 $ 1. 685399E+09 256.4
Total Value of Retrofit Benefits
The effectiveness of retrofitting all buildings is
summarized in Table LXV, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXV
EFFECT OF RETROFITTING ALL BUILDINGS
Average DF Tot Bldg Damage Tot Loss of Life
BEFORE RETROFIT
19. 57 $ 2. 69 G
12,996
ALL BLDGS RETROFIT
12. 23 $ 1. 69 G
256
213
Table LXV shows that 98Y. of the loss of life and 37Y.
of the building damage is mitigated by retrofitting all
the buildings.
The total benefit of retrofitting all the buildings,
based on the Value of Life (VOL) used previously, is:
B: $ 2. 69 G- $ 1.69 G + ((12,996- 256) * ($ 2. 25 M)l
= $ 29. 67 G
Retrofit Prioritization by Building Class
The Retrofit Group of program modules in REAL-RAP
was used to place all the buildings in the inventory in a
Building Class, determine the direct benefits and costs of
retrofit for each Building Class, and prioritize the
Building Classes by B/C ratio.
The prioritized list of Building Classes is shown in
Appendix c. In Appendix c, the first column is the
ranKing of the Building Class by B/C ratio. The next five
columns are the indices which identify the Building
TABLE LXV
EFFECT OF RETROFITTING ALL BUILDINGS
Average DF Tot Bldg Damage Tot Loss of Life
BEFORE RETROFIT
19.57 $ 2. 69 G
12,996
ALL BLDGS RETROFIT
12.23 $ 1.69 G
256
213
Table LXV shows that 98Y. of the loss of life and 37Y.
of the building damage is mitigated by retrofitting all
the buildings.
The total benefit of retrofitting all the buildings,
based on the Value of Life (VOL) used previously, is:
B = $ 2.69 G - $ 1.69 G + (12,996 - 256) • ($ 2.25 H)l
= $ 29. 67 G
Retrofit Prioritization by Building Class
The Retrofit Group of program modules in REAL-RAP
was used to place all the buildings in the inventory in a
Building Class. determine the direct benefits and costs of
retrofit for each Building Class, and prioritize the
Building Classes by BIC ratio.
The prioritized list of Building Classes is shown in
Appendix C. In Appendix C, the first column is the
ranking of the Building Class by B/C ratio. The next five
columns are the indices which identify the Building
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
214
Class (STG, SHZ, SSR, OTG, and BSG). The last three
columns are the total retrofit fit cost and net benefit of
all the buildings in that class, and the B/C ratio for
that class.
A Rey to the symbols used in the Building
Classification System columns in Appendix C follows (an
explanation of each index was presented in Chapter VI).
COLUMN
SHF
s
OCCUP TYPE
SIZE
H
L
SYMBOL
1-
1+
S,PA
G,ES
O,C,H
R
I
s M
L
MEANING
Higher soil hazard zone
Lower soil hazard zone
Struc Score less than 1.0
Struc Score greater than 1.0
School, Public Assembly
Govt. Bldg., Emer. Serv.
Office, Commercial, Historic
Residential
Industrial
Small
Medium
Large or Very Large
214
Class (STG, SHZ, SSR, OTG, and BSG). The last three
columns are the total retrofit fit cost and net benefit of
all the buildings in that class, and the BIC ratio for
that class.
A Rey to the symbols used in the Building
Classification System columns in Appendix C follows (an
explanation of each index was presented in Chapter VI).
COLUMN
SHF
S
OCCUP TYPE
SIZE
H
L
SYMBOL
1-
1+
S,PA
G,ES
O,C,H
R
I
S
M
L
MEANING
Higher soil hazard zone
Lower soil hazard zone
Struc Score less than 1.0
Struc Score greater than 1.0
School, Public Assembly
Govt. Bldg., Emer. Servo
Office, Commercial, Historic
Residential
Industrial
Small
Medium
Large or Very Large
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
215
From Appendix c, it can be seen that 284 Building
Classes (of a possible 540) are used to classify all the
buildings in the inventory.
System Efficiency
The maximum B/C ratio is 982. The first B/C less
than 1.0 occurs at R/F RanK number 86. This is also where
the Net Benefit becomes negative.
Therefore, the point of maximum efficiency, where
system net benefit is maximized, occurs at retrofit RanK
number 85. This marks the point of Optimum Retrofit
Investment, and is identified in the retrofit
prioritization of Appendix C.
A running sum of the costs and benefits was made
from Appendix c. From the running sum, a System Efficiency
Curve was made (Fig. 44). The System Efficiency Curve
shows Benefit (B) and Net Benefit (NB = B - C) vs.
Retrofit Investment (C). From Fig. 44, the Optimal
Retrofit Investment can be seen as the point of maximum
Net Benefit on the NB curve.
Parameters of the Optimal Retrofit Program
System parameters of the Optimal Retrofit were
derived from the running sum of the costs and benefits,
and are presented in Table LXVI, following.
215
From Appendix C, it can be seen that 284 Building
Classes (of a possible 540) are used to classify all the
buildings in the inventory.
System Efficiency
The maximum B/C ratio is 982. The first B/C less
than 1.0 occurs at RIF RanK number 86. This is also where
the Net Benefit becomes negative.
Therefore, the point of maximum efficiency, where
system net benefit is maximized, occurs at retrofit RanK
number 85. This marks the point of Optimum Retrofit
Investment, and is identified in the retrofit
prioritization of Appendix C.
A running sum of the costs and benefits was made
from Appendix C. From the running sum, a System Efficiency
Curve was made (Fig. 44). The System Efficiency Curve
shows Benefit (B) and Net Benefit (NB = B - C) vs.
Retrofit Investment (C). From Fig. 44, the Optimal
Retrofit Investment can be seen as the point of maximum
Net Benefit on the NB curve.
Parameters of the Optimal Retrofit Program
System parameters of the Optimal Retrofit were
derived from the running sum of the costs and benefits,
and are presented in Table LXVI, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
25
20
~ I z
0 ::J 15 ...J m
10
5
0
0
SYSTEM EFFICIENCY CURVES Retrofit for DBE (PGA=.20)
Portland Quadrangle - 7,690 Buildings
I
~! I I 8
I' ~ I N~
I 1\ I
I i\ I
I I \ J ~ I rvF "'IVt:l>L - ~~~~v.u VIIIIIUII
Net Benefit~J28.31 Billion .. at . = ~ 0.7
I I
I
1 2 3 RETROFIT INVESTMENT, BILLION-$
216
I
-
I
4
Figure 44. System Efficiency Curves for Retrofit for DBE (PGA = . 20) for the Portland Quad.
30
25
20 ~ , z o 15 .....J .....J m
10
5
o o
SYSTEM EFFICIENCY CURVES Retrofit for DBE (PGA=.20)
Portland Quadrangle - 7,690 Buildings
I
~! I I B I
VI {' \. I -~ h
I \ I
I I I I
1\ !
! \ I '~p~irm:lIIVr-IIIIVt:;:;L - l~l:I~~·U Vlllllurl Net BenefitrlJ2s.31 Billion /1, at J[ rimar = ~ 0.7
I I
123 RETROFIT INVESTMENT, BILLlON-$
I
4
216
Figure 44. System Efficiency Curves for Retrofit for DBE (PGA : . 20) for the Portland Quad.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
217
TABLE LXVI
SYSTEM PARAMETERS OF OPTIMAL RETROFIT INVESTMENT FOR A SCENARIO EARTHQUAKE EVENT OF PGA = . 20
Total Retrofit Cost $ 953. 6 Million
Net System Benefit $ 28. 31 Billion
Aggregate BIC 30.7
No. Bldgs. Retrofit 903 bldgs
Bldg Damage Avoided $ 646. Million
Lives Saved 12,717 lives
Running sums were also made of Direct Benefits by
Building Class (see Appendix). Figs. 45 and 46 present
curves of lives saved and building damage avoided,
respectively, vs. retrofit investment. From these curves,
the Direct Benefit 1 Retrofit Cost ratios for an optimal
retrofit investment ($ 953.6 Million) are:
a) 13.3 lives saved 1 $ 1-Million invested, and
b) $. 68 damage avoided 1 $ 1.00 invested.
Comparison of Optimal Retrofit to Full Retrofit
Table LXVII, following, compares the amount of
loss mitigation provided by the optimal retrofit versus a
"complete" retrofit (i.e., all buildings retrofit).
217
TABLE LXVI
SYSTEM PARAMETERS OF OPTIMAL RETROFIT INVESTMENT FOR A SCENARIO EARTHQUAKE EVENT OF PGA = . 20
Total Retrofit Cost $ 953. 6 Million
Net System Benefit $ 28.31 Billion
Aggregate B/C 30.7
No. Bldgs. Retrofit 903 bldgs
Bldg Damage Avoided $ 646. Million
Lives Saved 12,717 lives
Running sums were also made of Direct Benefits by
Building Class (see Appendix). Figs. 45 and 46 present
curves of lives saved and building damage avoided,
respectively, vs. retrofit investment. From these curves,
the Direct Benefit I Retrofit Cost ratios for an optimal
retrofit investment ($ 953.6 Million) are:
a) 13.3 lives saved I $ 1-Million invested, and
b) $ .68 damage avoided I $ 1.00 invested.
Comparison of Optimal Retrofit to Full Retrofit
Table LXVII, following, compares the amount of
loss mitigation provided by the optimal retrofit versus a
"complete" retrofit (i.e., all buildings retrofit).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
12
11 en 10 Cl z 9 <{ en :::> 8 0 I 7 1-6 6 w ~ 5 en en 4 w > 3 :::i
2
1
0
I I
I
I
LIVES SAVED vs. RETROFIT INVESTMENT Retrofit for DBE (PGA=.20)
Portland Quadrangle - 7,690 Buildings
' I ;, ~ I l I
1/ I
I _\ I \ I I
\ A,t Optim~l Retrofi lnvestn ent: Lives Sa ed = 12 i717
I !
A~gregat~ 8/C = 3.3 Live s/$Millio~ ' ' ' I
I I I
I
I
!
I
I
!
I
0 1 2 3 4
RETROFIT INVESTMENT, BILLION-$
218
Figure 45. Lives Saved vs. Retro£it Investment £or Retro£it £or DBE (PGA = .20).
13
12
11 en 10 0 Z 9 « en :::l 8 0 I 7 I-0 6 W
~ 5 en en 4 w > 3 -I
2
1
0
I I I
I
I
LIVES SAVED vs. RETROFIT INVESTMENT Retrofit for DBE (PGA=.20)
Portland Quadrangle - 7,690 Buildings
, 1
L1 ~ 1 I I
iL I I \ I \ /1
\.Ai OPtim~1 Retrofi Investn ent: Uves Sa ed = 12 717
I ! A~gregat~ BIC = 3.3 Live
, , ~/$Millio~
I
1 I I I
1
I , j
I
I
I
I
o 1 234 RETROFIT INVESTMENT, BILLlON-$
218
Figure 45. Lives Saved vs. Retrofit Investment for Retrofit for DBE (PGA = .20).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
El7 I z
0 :::i .....J
~ I
0 0 ~
0 w Q
~ w C> <( ~
C§
BLDG DAMAGE AVOIDED vs. RETROFIT INVESTMENT Retrofit for DBE (PGA=.20)
Portland Quadrangle- 7,690 Buildings
10 I I I I i I ~
s~l+--+--+-~1--~'--~~~r~--+-~ 8 ---4--1 ---+--! --+-1 -~11 ~--+i -------+--! -+--------i-1 --H
I I l/ 7~~~~~y+-~~--~~--~~--r--rl ~II
/"'1~ i I I I 6 -H---1--r~~-+---+---r---r--~---H I 'I\ At oltimal Ritrofit lm estmeJ I 5 ---H-----+V-I--+---orrr.ljla1go'Hu'""';a""'m""'a~;~""'~~-m I-\1V""o~~~~ ~~~*la.4-==~ '11~1!:Hl> Mtlltmm.lo...-n t---+1
1 Aggrj:!gate B~ = 0.6 .. ,. I
4 -H--~---+--~---+---r--~---r---r
3~~--~~~~~~~--~---~-~ 2 -H-1--+--/1 ----+----t--+---------+--------+----+----t-
/ 1 I 0 1 2 3 4
RETROFIT INVESTMENT, BILLION-$
Figure 46. Bldg Damage Avoided vs. Retrofit Investment for Retrofit for DBE (PGA = .20).
219
10
9 EI7
I
Z 8 0 --l --l 7 ~
I 0
6 0 ~
0 5 w 0 0 4 > « w 3 C> « ~ 2 « 0
1
0
I I I
I
I I I
I
BLDG DAMAGE AVOIDED vs. RETROFIT INVESTMENT Retrofit for DBE (PGA=.20)
Portland Quadrangle - 7,690 Buildings
I I
I i I ~ I ~ J
I I I~! I I
! 1/ 1 A I I I i II
i I I I
/ I \ At Oltimal Ritrofit 1m I I estment:
V I?lug uamag.!!. I'\VOI~~ IC =:\1041 Million
I I Aggr j:!gate B~ = O.6e
r I
II II I
II I
o 1 2 3 4 RETROFIT INVESTMENT, BILLlON-$
Figure 46. Bldg Damage Avoided vs. Retrofit Investment for Retrofit for DBE (PGA = .20).
219
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
220
TABLE LXVII
COMPARISON BETWEEN COMPLETE RETROFIT AND OPTIMAL RETROFIT
Tot Bldg Damage
Tot Loss of Life
Retrofit Investment
BEFORE RETROFIT
$ 2. 69 G
12,996
0
ALL BLDGS RETROFIT
$ 1. 69 G
256
$ 4.42 G
OPTIMAL RETROFIT
$ 2.05 G
279
$ .95 G
From Table LXVII, it is clear that the optimal
retrofit provides essentially the same life-safety benefit
as retrofitting all the buildings, but at one-fifth of the
cost. Retrofitting past optimal does provide additional
avoided building dmage, but at one-sixth the B/C ratio as
optimal.
Building Class Indices in the Optimal Retrofit
Table LXVIII, following, presents a summary of the
Building Types and Occupancy Groups of the 903 buildings
included within the optimal retrofit program.
220
TABLE LXVII
COMPARISON BETWEEN COMPLETE RETROFIT AND OPTIMAL RETROFIT
Tot Bldg Damage
Tot Loss of Life
Retrofit Investment
BEFORE RETROFIT
$ 2. 69 G
12,996
o
ALL BLDGS RETROFIT
$ 1.69 G
256
$ 4.42 G
OPTIMAL RETROFIT
$ 2.05 G
279
$ .95 G
From Table LXVII, it is clear that the optimal
retrofit provides essentially the same life-safety benefit
as retrofitting all the buildings, but at one-fifth of the
cost. Retrofitting past optimal does provide additional
avoided building dmage, but at one-sixth the BIC ratio as
optimal.
Building Class Indices in the Optimal Retrofit
Table LXVIII, following, presents a summary of the
Building Types and Occupancy Groups of the 903 buildings
included within the optimal retrofit program.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXVIII
SUMMARY OF BUILDING TYPES AND OCCUPANCY GROUPS INCLUDED IN THE OPTIMAL RETROFIT PROGRAM
OCCUPANCY TYPE GROUP BLDG -----------------------------------
221
TYPE S,PA G,ES O,C,H R I TOTAL
S1,2,4 0 0 22 0 0 22 C1 5 6 .q.4 3 4 62 C2 0 1 9 0 3 13 C3/S5 5 0 29 3 8 .q.5 PC1 1 3 11 0 76 91 PC2 0 0 0 0 1 1 RM 0 1.q. 30 3 31 78 URM 38 6 328 116 103 591 W,S3 0 0 0 0 0 0
TOTAL 49 30 .q.73 125 226 903
It can be concluded that the number Unreinforced
Masonry (URM) buildings included in the optimal retrofit
(591) is an order of magnitude more than any other
building type. Of the other building types, Concrete
Frame (C1), URM Frame Infill (C3/S5), Tilt-up (PC1), and
Reinforced Masonry (RM) all have similar numbers of
buildings (.q.5 to 91) included in the optimal retrofit.
More than 75-percent of the total number of buildings
included in the optimal retrofit are included in the
Office-Commercial-Historical and Industrial Occupancies.
Table LXIX, following, presents a summary of the
number of Building Types in the optimal retrofit program,
TABLE LXVIII
SUMMARY OF BUILDING TYPES AND OCCUPANCY GROUPS INCLUDED IN THE OPTIMAL RETROFIT PROGRAM
OCCUPANCY TYPE GROUP BLDG -----------------------------------
221
TYPE S,PA G,ES O,C,H R I TOTAL
S1,2,4 0 0 22 0 0 22 Cl 5 6 J!.4 3 4 62 C2 0 1 9 0 3 13 C3/S5 5 0 29 3 B 45 PCl 1 3 11 0 76 91 PC2 0 0 0 0 1 1 RM 0 lJ!. 30 3 31 78 URM 38 6 328 116 103 591 W,S3 0 0 0 0 0 0
TOTAL 49 30 J!.73 125 226 903
It can be concluded that the number Unreinforced
Masonry (URM) buildings included in the optimal retrofit
(591) is an order of magnitude more than any other
building type. Of the other building types, Concrete
Frame (Cl), URM Frame Infill (C3/S5), Tilt-up (PC1), and
Reinforced Masonry (RM) all have similar numbers of
buildings (J!.5 to 91) included in the optimal retrofit.
Hore than 75-percent of the total number of buildings
included in the optimal retrofit are included in the
Office-Commercial-Historical and Industrial occupancies.
Table LXIX, following, presents a summary of the
number of Building Types in the optimal retrofit program,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
as a percentage of the total number of its type in the
inventory.
TABLE LXIX
PERCENTAGE OF BUILDINGS TYPES INCLUDED IN OPTIMAL RETROFIT PROGRAM
S1,2,4 14. 5Y. C1 39.01. C2 0.8Y. C3/S5 71. 4Y. PC1 17.71. PC2 2.6Y. RM 8. 5Y. URM 68.61. W,S3 OY.
TOTAL 11. 7Y.
Unreinforced Masonry (URM) and URM Frame Infill
(C3/S5) show a very high percentage of the inventory
222
included in the optimal retrofit. Additionally, Concrete
Frame (C1) shows a large percentage of the inventory
included.
Table LXX, following, presents a summary of the
number of Occupancy Type Groups in the optimal retrofit
program, as a percentage of the total number of its type
group in the inventory.
as a percentage of the total number of its type in the
inventory.
TABLE LXIX
PERCENTAGE OF BUILDINGS TYPES INCLUDED IN OPTIMAL RETROFIT PROGRAM
S1,2,4 14. 5'l. C1 39.0'l. C2 0.8'l. C3/S5 71. 4'l. PC1 17.7'l. PC2 2.6'l. RM 8.5'l. URM 68. 6'l. W,S3 O'l.
TOTAL 11. 7'l.
Unreinforced Masonry (URM) and URM Frame Infill
(C3/S5) show a very high percentage of the inventory
222
included in the optimal retrofit. Additionally, Concrete
Frame (C1) shows a large percentage of the inventory
included.
Table LXX, following. presents a summary of the
number of Occupancy Type Groups in the optimal retrofit
program, as a percentage of the total number of its type
group in the inventory.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXX
PERCENTAGE OF OCCUPANCY TYPE GROUPS INCLUDED IN OPTIMAL RETROFIT PROGRAM
School, Pub. Assem. Govt. Bldg. , Emer. Serv. Office, Commercial, Historic Residential Industrial
TOTAL
11. 2Y. 16. 1 Y. 14. 3Y.
6. 7Y. 11. 9Y.
11. n
223
The percentage of each occupancy type group included
in the optimal retrofit is of the same order of magnitude
as the percentage of the entire inventory, which is 11. 7Y..
The number of buildings by Soil Hazard Zone (SHZ)
range included in the optimal retrofit are as follows:
H (high) L (low)
TOTAL
179 bldgs 724
903 bldgs
The fraction of buildings in each Soil Hazard Zone
included in the optimal retrofit roughly matches the
fraction of the entire building inventory in each zone.
The number of buildings by Building Size Group
(SIZE) included in the optimal retrofit are as follows:
S (small) M (medium) L (large)
TOTAL
325 bldgs 433 145
903 bldgs
Again, it appears that the fraction of buildings in each
TABLE LXX
PERCENTAGE OF OCCUPANCY TYPE GROUPS INCLUDED IN OPTIMAL RETROFIT PROGRAM
School, Pub. Assem. Govt. Bldg., Emer. Servo Office, Commercial, Historic Residential Industrial
TOTAL
11. 2Y. 16. 1 Y. 14. 3Y.
6. 7Y. 11. 9Y.
11. 7Y.
223
The percentage of each occupancy type group included
in the optimal retrofit is of the same order of magnitude
as the percentage of the entire inventory, which is 11.7Y..
The number of buildings by Soil Hazard Zone (SHZ)
range included in the optimal retrofit are as follows:
H (high) L (low)
TOTAL
179 bldgs 724
903 bldgs
The fraction of buildings in each Soil Hazard Zone
included in the optimal retrofit roughly matches the
fraction of the entire building inventory in each zone.
The number of buildings by Building Size Group
(SIZE) included in the optimal retrofit are as follows:
S (small) M (medium) L (large)
TOTAL
325 bldgs 433 145
903 bldgs
Again, it appears that the fraction of buildings in each
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
224
size group roughly matches the fraction of the building
inventory in each group.
The number of buildings by Structural Score Range
(SSR) included in the optimal retrofit are as follows:
1- (S < +1) 1+ (S > +1)
TOTAL
881 bldgs 22
903 bldgs
It is clear that 98-percent of the buildings in the
optimal retrofit have a Structural Score of less than 1.0.
Sensitivity Analysis
Clearly, the optimal retrofit program is being
driven by the benefit of the dollar value of lives saved
by retrofit. Two parameters which affect the dollar value
of lives saved and could possibly impact the scope
of the optimal retrofit program are the number of people
(HP) in the buildings, and the dollar value used for human
life (VOL). To study the sensitivity of the optimal
retrofit program to these two parameters, the following
two alternate optimal retrofit programs were determined:
1. An optimal retrofit program using for HP the
Lowpoint (in lieu of Midpoint) of the range of
the number of people in each building. A system
efficiency curve based on the Lowpoint HP is
shown in Figure 47, following.
22~
size group roughly matches the fraction of the building
inventory in each group.
The number of buildings by Structural Score Range
(SSR) included in the optimal retrofit are as follows:
1- (S < +1) 1+ (S > +1)
TOTAL
881 bldgs 22
903 bldgs
It is clear that 98-percent of the buildings in the
optimal retrofit have a Structural Score of less than 1.0.
Sensitivity Analysis
Clearly, the optimal retrofit program is being
driven by the benefit of the dollar value of lives saved
by retrofit. Two parameters which affect the dollar value
of lives saved and could possibly impact the scope
of the optimal retrofit program are the number of people
(HP) in the buildings, and the dollar value used for human
life (VOL). To study the sensitivity of the optimal
retrofit program to these two parameters, the following
two alternate optimal retrofit programs were determined:
1. An optimal retrofit program using for HP the
Lowpoint (in lieu of Midpoint) of the range of
the number of people in each building. A system
efficiency curve based on the Lowpoint HP is
shown in Figure ~7, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
225
2. An optimal retrofit program using for VOL
$ 500, 000. (in 1 ieu of $ 2. 25 M). A system
efficiency curve based on VOL=$ 500,000. is
shown in Figure 48, following.
A comparison between the optimal retrofit program
determined previously (based on NP = Midpoint, and
VOL = $ 2. 25 M) and the two alternate programs is made in
Table LXXI, following.
TABLE LXXI
SENSITIVITY ANALYSIS RESULTS
RETROFIT NO. OF RETROFIT PROGRAM BLDGS INVESTMENT
DBE Optimal 903 $ 954. M Retrofit
DBE Optimal 892 $ 947. M using LOWPOINT
DBE Optimal 860 $ 743. M using LOW VOL
From Table LXXI, it is clear that the scope of the
retrofit program changes only slightly when the Value of
Life (VOL) or the Humber of People (HP) used in the
analysis are varied.
Variations in the VOL or range of number of people
in the buildings will not change the decision to retrofit.
225
2. An optimal retrofit program using for VOL
$ 500,000. (in I ieu of $ 2. 25 M). A system
efficiency curve based on VOL = $ 500,000. is
shown in Figure 48, following.
A comparison between the optimal retrofit program
determined previously (based on NP = Midpoint, and
VOL = $ 2.25 M) and the two alternate programs is made in
Table LXXI, following.
TABLE LXXI
SENSITIVITY ANALYSIS RESULTS
RETROFIT NO. OF RETROFIT PROGRAM BLDGS INVESTMENT
DBE Optimal 903 $ 954. M Retrofit
DBE Optimal 892 $ 947. M using LOWPOINT
DBE Optimal 860 $ 743. M using LOW VOL
From Table LXXI, it is clear that the scope of the
retrofit program changes only slightly when the Value of
Life (VOL) or the Number of People (NP) used in the
anal ysis are varied.
Variations in the VOL or range of number of people
in the buildings will not change the decision to retrofit.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(:17 I z
226
SYSTEM EFFICIENCY CURVES Retrofit for DBE (PGA=.20) at Lowpoint of NP Range
Portland Quadrangle - 7,690 Buildings
30 I I i
! I I
+ 41 I l i II i I
I I !
!
I I I I II I I I
I I I
I I
25
I I 20
! !
~ 15 ; I i I I I ! ! I ...J
ill 10
5
0
I v ,~ I 8 I -+-- ...... L \ - " .. -- -
I IV !llllr'-"L-~~~~0./ !"lihiun Net Benefit=? $11.35 Billion
'~lit'! ~t . =1 .99
I I
I I I
I I 0 1 2 3 4
RETROFIT INVESTMENT, BILLION-$
Figure 47. System efficiency curves for retrofit for DBE (PGE = • 20) for the Portland Quad, using the Lowpoint of HP range.
226
SYSTEM EFFICIENCY CURVES Retrofit for DBE (PGA=.20) at Lowpoint of NP Range
Portland Quadrangle - 7,690 Buildings
30
25
20
~ I
Z o 15 .-I .-I ill
10
5
o
I I J !
I ! + 41 I ! J II i I
I I ! I
I I I i i I I I I
1 I ! I I
I
I I
! !
; I I I I I ! ! i
I
V l~ , B I
-"It
L ~ - ~ --/ 'N~t
,.nell rut IIIIVC:Jl-I~~~U.' !vlil.iu" Benefit ~ $11.35 Billion
II. 1=1/r. ~t '1r rjm~I=1 .99
/ I
I I I I I I
o 1 2 3 4
RETROFIT INVESTMENT, BILLlON-$
Figure 47. System efficiency curves for retrofit for DBE (PGE : . 20) for the Portland Quad, using the Lowpoint of HP range.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
25
20
~ I z
0 ::i 15 _J
Ill
10
5
0
I
I
I
I
i
I I
SYSTEM EFFICIENCY CURVES Retrofit for DBE(PGA=.20) with Value
of Life = $500,000 Portland Quadrangle- 7,690 Buildings
I I I I
I I I
i I
I I I
I I I ' i
I
I I I
I
!
I I
! I
I I I : I ! I I !
I
I I i I i ! I I i I
I I Op{imal R/ Invest~ $743 ~illion
Net = !1:61 n Rill inn 1 / Aggregate 8/C = 9.21 ! I I 1
i
1/ I
I
I I
I
i I
/I I I
/ 8 k::f: I
I
I ~ v I
_NI~
i
i I
I I . I ! i i:
II
II I: I
I
II I
/ 1-- -v I
0 1 2 3 4
RETROFIT INVESTMENT, BILLION-$
227
Figure 48. System efficiency curves for retrofit for DBE (PGA = .20) for the Portland Quad, using VOL : $ 500, 000.
30
25
20 ER-
I Z o 15 ....J ....J OJ
10
5
o
I
I I
I
I I I
SYSTEM EFFICIENCY CURVES Retrofit for DBE(PGA=.20) with Value
of Life = $500,000 Portland Quadrangle - 7,690 Buildings
I I , I I
I I I
i I
I I I
I ! I I
i i !
I I I
I I I
i I I i I
I ! I I ! I I I i I i ! I I i I
I / optimal RI Invest ~ $743 ~iIIiOn Net Benefit = !l:S1 0 Rillinn
I I Aggregate SIC = 9.21 ! i I ! i
1/ I I
I I
I i I
II I I
I. B II I
I ~ I £,
I Nt:s
! I I j I I I! I! ' I ! i i,
Ii
Ii I: I
I
II i
/ I -./
I I o 1 2 3 4
RETROFIT INVESTMENT, BILLlON-$
227
Figure 48. System efficiency curves for retrofit for DBE (PGA = .20) for the Portland Quad, using VOL = $ 500, 000.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEVELOPMENT OF AN OPTIMAL RETROFIT PROGRAM BASED ON AN EXPECTED VALUE ANALYSIS
An expected value analysis of retrofit was made,
using the methodology described in Chapter VI. The
resulting optimal retrofit program resulting from this
228
analysis is compared to the DBE optimal retrofit program.
Present Value of Full Retrofit
The annual expected value of a "full" retrofit,
wherein all the buildings in the inventory are retrofit,
is presented in Table LXXII, following. In Table LXXII,
the sum of the annual benefits for a range of four PGA
scenario events is taKen as the total annual expected
value of retrofit. The values of Damage Avoided and Lives
Saved are the difference between the bottom-line values in
Tables LI to LIV, and Tables LXI to LXIV. The Scenario
Benefit is taKen as the dollar value of avoided damage
plus the lives saved times the value of life (Eq. 6. 1).
EAE is the expected annual number of earthquaKes, as
presented in Chapter VI. The annual expected value of
benefit is taKen as EAE times the scenario benefit.
DEVELOPMENT OF AN OPTIMAL RETROFIT PROGRAM BASED ON AN EXPECTED VALUE ANALYSIS
An expected value analysis of retrofit was made,
using the methodology described in Chapter VI. The
resulting optimal retrofit program resulting from this
228
analysis is compared to the DBE optimal retrofit program.
Present Value of Full Retrofit
The annual expected value of a "full" retrofit,
wherein all the buildings in the inventory are retrofit,
is presented in Table LXXII, following. In Table LXXII,
the sum of the annual benefits for a range of four PGA
scenario events is taken as the total annual expected
value of retrofit. The values of Damage Avoided and Lives
Saved are the difference between the bottom-line values in
Tables LI to LIV, and Tables LXI to LXIV. The Scenario
Benefit is taken as the dollar value of avoided damage
plus the lives saved times the value of life (Eq. 6.1).
EAE is the expected annual number of earthquakes, as
presented in Chapter VI. The annual expected value of
benefit is taken as EAE times the scenariO benefit.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
229
TABLE LXXII
ANNUAL EXPECTED VALUE OF FULL RETROFIT
PGA DAMAGE LIVES SCENARIO EAE ANNUAL EXP AVOID (M$) SAVED BEN (M$) BEN (M$)
. 08 $ 482. 5 5, 752. $ 13,424. . 01 $ 134. 24
. 20 $ 1, 008. 5 12, 740. $ 29, 674. . 002 $ 59. 35
. 27 $ 1,242. 8 17, 549. $ 40, 727. . 001 $ 40. 73
. 39 $ 1, 458. 9 21, 972. $ 50, 895. . 0004 $ 20. 36
TOTAL ANNUAL EXPECTED VALUE OF BENEFIT: $ 254. 67
The Expected Present Value (PV) of the full retrofit
is the sum of the annual expected value over a planning
horizon, discounted to present value. As described in
Chapter VI, a planning horizon of 50 years and discount
rate of 7-percent results in a Uniform Series; Present
Worth Factor (PWF) of 13.801.
PV: $ 254.67 M * 13.801
: $ 3, 514. 7 M
This PV is roughly one-tenth of the Benefit (B) from
a full retrofit based on the Design Basis Earthquake, from
page 213.
The Net Present Value of full retrofit is the
difference between the Present Value and the Retrofit Cost:
229
TABLE LXXII
ANNUAL EXPECTED VALUE OF FULL RETROFIT
PGA DAMAGE LIVES SCENARIO EAE ANNUAL EXP AVOID (M$) SAVED BEN (M$) BEN (M$)
.08 $ 482. 5 5,752. $ 13,424. .01 $ 134.24
.20 $ 1,008.5 12,740. $ 29,674. .002 $ 59. 35
.27 $ 1,242.8 17,549. $ 40,727. .001 $ 40. 73
.39 $ 1,458.9 21,972. $ 50,895. .0004 $ 20. 36
TOTAL ANNUAL EXPECTED VALUE OF BENEFIT: $ 254. 67
The Expected Present Value (PV) of the full retrofit
is the sum of the annual expected value over a planning
horizon, discounted to present value. As described in
Chapter VI, a planning horizon of 50 years and discount
rate of 7-percent results in a Uniform Series/ Present
Worth Factor (PWF) of 13.801.
PV : $ 254.67 M * 13.801
: $ 3, 514. 7 M
This PV is roughly one-tenth of the Benefit (B) from
a full retrofit based on the Design Basis Earthquake, from
page 213.
The Net Present Value of full retrofit is the
difference between the Present Value and the Retrofit Cost:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
NPV : PV - C
= $ 3, 514.7 M- $ 4,418. 3M
= - $ 903. 60 M
230
The negative result for NPV clearly indicates that,
from an expected value approach, there will be a loss of
value to retrofit all the buildings in the inventory (this
was not the case with the DBE, where the NB at full
retrofit was still a large positive number).
Therefore, from an expected value viewpoint, an
optimum retrofit program taKes on added significance, in
that the optimal retrofit investment is necessary not only
to maximum benefit, but also to avoid actual loss.
Retrofit Prioritization by Building Class
The Retrofit Group of program modules in REAL-RAP
was used to place all the buildings in the inventory in a
Building Class, determine the PV and NPV of retrofit (as
presented above) for each Building Class, and prioritize
the Building Classes by PV/C ratio.
The prioritized list of Building Classes is shown in
Appendix D. In Appendix D, the first column is the ranKing
of the Building Class by PV/C ratio. The next five
columns are the indices which identify the Building Class
(STG, SHZ, SSR, OTG, and BSG). The last three columns are
the total retrofit cost, NPV of retrofit, and PV/C ratio
for all the buildings in that class.
HPV = PV - C
= $ 3,514.7 M - $ 4,418.3 M
= - $ 903. 60 M
230
The negative result for HPV clearly indicates that,
from an expected value approach, there will be a loss of
value to retrofit all the buildings in the inventory (thiS
was not the case with the DBE, where the HB at full
retrofit was still a large positive number).
Therefore, from an expected value viewpoint, an
optimum retrofit program takes on added significance, in
that the optimal retrofit investment is necessary not only
to maximum benefit, but also to avoid actual loss.
Retrofit Prioritization by Building Class
The Retrofit Group of program modules in REAL-RAP
was used to place all the buildings in the inventory in a
Building Class, determine the PV and HPV of retrofit (as
presented above) for each Building Class, and prioritize
the Building Classes by PV/C ratio.
The prioritized list of Building Classes is shown in
Appendix D. In Appendix D, the first column is the ranking
of the Building Class by PV/C ratio. The next five
columns are the indices which identify the Building Class
(STG, SHZ, SSR, OTG, and BSG). The last three columns are
the total retrofit cost, HPV of retrofit, and PV/C ratio
for all the buildings in that class.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
231
A Key to the symbols used in the Building
Classification System columns in Appendix D follows (an
explanation of each index was presented in Chapter VI).
COLUMN
SMF
s
OCCUP TYPE
SIZE
H
L
SYMBOL
1-
1+
S,PA
G,ES
O,C,H
R
I
s M
L
MEANING
Higher soil hazard zone
Lower soil hazard zone
Struc Score less than 1.0
Struc Score greater than 1.0
School, Public Assembly
Govt. Bldg. , Emer. Serv.
Office, Commercial, Historic
Residential
Industrial
Small
Medium
Large or Very Large
In Appendix D, the same 284 Building Classes used
in the DBE Optimal Retrofit are used to classify the
buildings for the Expected Value Optimal Retrofit (because
it is the same building inventory).
231
A Key to the symbols used in the Building
Classification System columns in Appendix D follows (an
explanation of each index was presented in Chapter VI).
COLUMN
SMF
S
OCCUP TYPE
SIZE
H
L
SYMBOL
1-
1+
S,PA
G,ES
O,C,H
R
I
S
M
L
MEANING
Higher soil hazard zone
Lower soil hazard zone
Struc Score less than 1.0
Struc Score greater than 1. 0
School, Public Assembly
Govt. Bldg., Emer. Servo
Office, Commercial, Historic
Residential
Industrial
Small
Medium
Large or Very Large
In Appendix D, the same 284 Building Classes used
in the DBE Optimal Retrofit are used to classify the
buildings for the Expected Value Optimal Retrofit (because
it is the same building inventory).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
232
System Efficiency
The maximum PV/C ratio is 182. The first B/C less
than 1.0 occurs at R/F RanK number 79. This is also where
the Net Present Value becomes negative.
Therefore, the point of maximum efficiency, where
system net benefit is maximized, occurs at retrofit RanK
number 78. This marks the point of Optimum Retrofit
Investment, and is identified in the retrofit
prioritization of Appendix D.
A running sum of the costs and benefits was made
from Appendix D. From the running sum, a System
Efficiency Curve was made (Fig. 49). The System Efficiency
Curve shows Present Value of retrofit benefit (PV) and Net
Present Value of retrofit benefit (NPV = PV - C) vs.
Retrofit Investment (C). From Fig. 49, the Optimal
Retrofit Investment can be seen as the point of maximum
Net Present Value on the NPV curve.
Parameters of the Optimal Retrofit Program
System parameters of the Optimal Retrofit were
derived from the running sum of the costs and benefits,
and are presented in Table LXXIII, following.
232
System Efficiency
The maximum PV/C ratio is 182. The first B/C less
than 1.0 occurs at RIF RanK number 79. This is also where
the Net Present Value becomes negative.
Therefore, the point of maximum efficiency, where
system net benefit is maximized, occurs at retrofit Rank
number 78. This marks the point of Optimum Retrofit
Investment, and is identified in the retrofit
prioritization of Appendix D.
A running sum of the costs and benefits was made
from Appendix D. From the running sum, a System
Efficiency Curve was made (Fig. 49). The System Efficiency
Curve shows Present Value of retrofit benefit (PV) and Net
Present Value of retrofit benefit (NPV = PV - C) vs.
Retrofit Investment (C). From Fig. 49, the Optimal
Retrofit Investment can be seen as the point of maximum
Net Present Value on the NPV curve.
Parameters of the Optimal Retrofit Program
System parameters of the Optimal Retrofit were
derived from the running sum of the costs and benefits,
and are presented in Table LXXIII, following.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
3
EA-1 z ~ 2 ....I CD
1
0
I
I
0
SYSTEM EFFICIENCY CURVES Retrofit for Expected Value Analysis Portland Quadrangle- 7,690 Buildings
I I
I l I
I PV l I I
I
111 /O~mal R~rofit lnv~stment} $565.~ Million ] I Nel Prese Value = $2.65 illion
I I )(I Agbregate I
~/C =4.97
I I
~I ~~ i I I
1/
~ I I
~ NPV
I~ ~
1 2 3 4 RETROFIT INVESTMENT, BILLION-$
233
Figure 49. System Efficiency Curves for Retrofit for Expected Value Analysis for the Portland Quad.
4
3
2
1
o
I
o
SYSTEM EFFICIENCY CURVES Retrofit for Expected Value Analysis Portland Quadrangle - 7,690 Buildings
I I
I I I
I PV 1 I I
I
1f1/o~mal Rr lnv~stment} $56s.J Million ! I Net Prese Value = $2.65 illion
/ I~ Agbregate I
~/C =4.97
I I
VI ~I i I I
f{
I ~ ~ NPV
~ ~
123 RETROFIT INVESTMENT, BILLlON-$
I
4
233
I
Figure 49. System Efficiency Curves for Retrofit for Expected Value Analysis for the Portland Quad.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE LXXIII
SYSTEM PARAMETERS OF OPTIMAL RETROFIT INVESTMENT BASED ON AH EXPECTED VALUE ANALYSIS
Total Retrofit Cost $ 665.6 Million
Net System Benefit $ 2. 646 Billion
Aggregate B/C 4.97
No. Bldgs. Retrofit 846 bldgs
Building Class Indices in the Optimal Retrofit
234
Table LXXIV summarizes Building Types and Occupancy
Groups of the 846 buildings in the optimal retrofit.
TABLE LXXIV
SUMMARY OF BUILDING TYPES AND OCCUPANCY GROUPS INCLUDED IN THE OPTIMAL RETROFIT PROGRAM
BASED ON AH EXPECTED VALUE ANALYSIS
OCCUPANCY TYPE GROUP BLDG -----------------------------------TYPE S,PA G,ES O,C,H R I
S1,2,4 0 0 0 1 0 C1 5 4 37 4 1 C2 1 0 0 0 3 C3/S5 5 0 28 3 4 PC1 1 3 5 0 76 PC2 1 0 0 0 1 RM 2 1.1!- 30 4 31 URM 38 6 32.1!- 114 100 W,S3 0 0 0 0 0
TOTAL 53 27 42.1!- 126 216
TOTAL
1 51
4 40 85
2 81
582 0
846
TABLE LXXIII
SYSTEM PARAMETERS OF OPTIMAL RETROFIT INVESTMENT BASED ON AN EXPECTED VALUE ANALYSIS
Total Retrofit Cost $ 665. 6 Hi 11 ion
Net System Benefit $ 2.646 Billion
Aggregate BIC 4.97
No. Bldgs. Retrofit 846 bldgs
Building Class Indices in the Optimal Retrofit
234
Table LXXIV summarizes Building Types and Occupancy
Groups of the 846 buildings in the optimal retrofit.
TABLE LXXIV
SUMMARY OF BUILDING TYPES AND OCCUPANCY GROUPS INCLUDED IN THE OPTIMAL RETROFIT PROGRAM
BASED ON AN EXPECTED VALUE ANALYSIS
OCCUPANCY TYPE GROUP BLDG -----------------------------------TYPE S,PA G,ES O,C,H R I
S1,2,4 0 0 0 1 0 C1 5 4 37 4 1 C2 1 0 0 0 3 C3/S5 5 0 28 3 4 PC1 1 3 5 0 76 PC2 1 0 0 0 1 RM 2 1J!. 30 J!. 31 URH 38 6 324 114 100 W,S3 0 0 0 0 0
TOTAL 53 27 J!.24 126 216
TOTAL
1 51
4 40 85
2 81
582 0
8J!.6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
235
It can be concluded that the number Unreinforced
Masonry (URM) buildings included in the optimal retrofit
(582) is an order of magnitude more than any other
building type. Of the other building types, Concrete
Frame (C1), URM Frame Infill (C3/S5), Tilt-up (PC1), and
Reinforced Masonry (RM) all have similar numbers of
buildings (~0 to 85) included in the optimal retrofit.
More than 75-percent of the total buildings included
in the optimal retrofit are included in the
Office-Commercial-Historical and Industrial Occupancies.
Table LXXV, following, presents a summary of the
number of Building Types in the optimal retrofit program,
as a percentage of the total number of its type in the
inventory.
TABLE LXXV
PERCENTAGE OF BUILDIHGS TYPES IHCLUDED IH OPTIMAL RETROFIT PROGRAM BASED OH
AH EXPECTED VALUE AHALYSIS
S1,2,~ 0.7'/. C1 32. 1 t: C2 o. 2'/. C3/S5 63. 5'/. PC1 16.5t: PC2 5. 1 t: RM 8. 8'/. URM 67. 5'/. W,S3 0'/.
TOTAL 11. 0'/.
235
It can be concluded that the number Unreinforced
Masonry (URM) buildings included in the optimal retrofit
(582) is an order of magnitude more than any other
building type. Of the other building types, Concrete
Frame (Cl), URM Frame Infill (C3/S5), Tilt-up (PC1), and
Reinforced Masonry (RM) all have similar numbers of
buildings (~O to 85) included in the optimal retrofit.
More than 75-percent of the total buildings included
in the optimal retrofit are included in the
Office-Commercial-Historical and Industrial Occupancies.
Table LXXV, following, presents a summary of the
number of Building Types in the optimal retrofit program,
as a percentage of the total number of its type in the
inventory.
TABLE LXXV
PERCENTAGE OF BUILDINGS TYPES INCLUDED IN OPTIMAL RETROFIT PROGRAM BASED ON
AN EXPECTED VALUE ANALYSIS
Sl,2,~ 0.7Y. Cl 32. 1 I: C2 O. 21: C3/S5 63. 51: PCl 16. 51: PC2 5. lY. RM 8. 8Y. URH 67. 5Y. W,S3 01:
TOTAL 11. OX
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
236
Unreinforced Masonry (URM) and URM Frame Infill
(C3/S5) show a very high percentage of the inventory
included in the optimal retrofit. Additionally, Concrete
Frame (C1) shows a large percentage of the inventory
included.
Table LXXVI, following, presents a summary of the
number of Occupancy Type Groups in the optimal retrofit
program, as a percentage of the total number of its type
group in the inventory.
TABLE LXXVI
PERCENTAGE OF OCCUPANCY TYPE GROUPS INCLUDED IN OPTIMAL RETROFIT PROGRAM BASED ON
AH EXPECTED VALUE ANALYSIS
Schoo 1, Pub. Ass em. Govt. Bldg. , Emer. Serv. Office, Commercial, Historic Residential Industrial
TOTAL
12. 11: 14. 5Y. 12.8Y.
6. 8Y. 11. 4Y.
11.0Y.
The percentage of each occupancy type group included
in the optimal retrofit is of the same order of magnitude
as the percentage of the entire inventory, which is 11.0Y..
The number of buildings by Soil Hazard Zone (SHZ)
range included in the optimal retrofit are as follows:
236
Unreinforced Masonry (URM) and URM Frame Infill
(C3jS5) show a very high percentage of the inventory
included in the optimal retrofit. Additionally, Concrete
Frame (C1) shows a large percentage of the inventory
included.
Table LXXVI, following, presents a summary of the
number of Occupancy Type Groups in the optimal retrofit
program, as a percentage of the total number of its type
group in the inventory.
TABLE LXXVI
PERCENTAGE OF OCCUPANCY TYPE GROUPS INCLUDED IN OPTIMAL RETROFIT PROGRAM BASED ON
AN EXPECTED VALUE ANALYSIS
School, Pub. Assem. Govt. Bldg., Emer. Servo Office, CommerCial, Historic Residential Industrial
TOTAL
12. 11: 14. 5X 12.8X
6. 8X 11. 4X
11. OX
The percentage of each occupancy type group included
in the optimal retrofit is of the same order of magnitude
as the percentage of the entire inventory, which is 11.0X.
The number of buildings by Soil Hazard Zone (SHZ)
range included in the optimal retrofit are as follows:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
H (high) L (low)
TOTAL
161 bldgs 685
846 bldgs
2.37
The fraction of buildings in each Soil Hazard Zone included
in the optimal retrofit roughly matches the fraction of the
entire building inventory in each zone.
The number of buildings by Building Size Group
(SIZE) included in the optimal retrofit are as follows:
S (small) M (medium) L (large)
TOTAL
321 bldgs 416 109
846 bldgs
Again, it appears that the fraction of buildings in each
size group roughly matches the fraction of the building
inventory in each group.
The number of buildings by Structural Score Range
(SSR) included in the optimal retrofit are as follows:
1- (S < +1) 1+ (S > +1)
TOTAL
82.9 bldgs 17
846 bldgs
It is clear that 98-percent of the buildings in the
optimal retrofit have a Structural Score of less than 1.0.
Comparison of Expected Value and Design Basis Retrofits
A comparison of the Building Indices in the expected
value optimal retrofit with those presented previously for
the design basis optimal retrofit show that there is
little difference betweeen the buildings captured for
H (high) L (low)
TOTAL
161 bldgs 685
846 bldgs
237
The fraction of buildings in each Soil Hazard Zone included
in the optimal retrofit roughly matches the fraction of the
entire building inventory in each zone.
The number of buildings by Building Size Group
(SIZE) included in the optimal retrofit are as follows:
S (small) M (medium) L (large)
TOTAL
321 bldgs 416 109
846 bldgs
Again, it appears that the fraction of buildings in each
size group roughly matches the fraction of the building
inventory in each group.
The number of buildings by Structural Score Range
(SSR) included in the optimal retrofit are as follows:
1- (S < +1) 1+ (S > +1)
TOTAL
829 bldgs 17
846 bldgs
It is clear that 98-percent of the buildings in the
optimal retrofit have a Structural Score of less than 1.0.
Comparison of Expected Value and Design Basis Retrofits
A comparison of the Building Indices in the expected
value optimal retrofit with those presented previously for
the design basis optimal retrofit show that there is
little difference betweeen the buildings captured for
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
238
retrofit in both programs.
In order to compare the Direct Benefits of the
Expected Value Optimal Retrofit with the Design Basis
Optimal Retrofit, the Direct Benefits corresponding to an
earthquake scenario of PGA = . 20 were compiled from the
Expected Value Optimal Retrofit Building Classes, and
running sums were made of Direct Benefits by Building
Class (see Appendix). A comparison of system parameters is
presented in Table LXXVII, following.
TABLE LXXVII
COMPARISON OF SYSTEM PARAMETERS OF EXPECTED VALUE VS. DESIGN BASIS OPTIMAL RETROFIT PROGRAMS
Total Retrofit Cost
NPV and HB
Aggregate B/C
No. Bldgs. Retrofit
Dam. Avoid. (PGA:, 2)
Lives Saved (PGA=. 2)
EXPECTED VALUE RETROFIT
$ 665.6 M
$ 2,646. M
4.97
846 bldgs
$ 559.6 M
11,943 lives
DESIGN BASIS RETROFIT
$ 953.6 M
$ 28, 306. M
30.68
903 bldgs
$ 646. M
12,717 lives
As can be seen from Table LXXVII, there is not a
large difference between these two retrofit programs.
238
retrofit in both programs.
In order to compare the Direct Benefits of the
Expected Value Optimal Retrofit with the Design Basis
Optimal Retrofit, the Direct Benefits corresponding to an
earthquake scenario of PGA = . 20 were compiled from the
Expected Value Optimal Retrofit Building Classes, and
running sums were made of Direct Benefits by Building
Class (see Appendix). A comparison of system parameters is
presented in Table LXXVII, following.
TABLE LXXVII
COMPARISON OF SYSTEM PARAMETERS OF EXPECTED VALUE VS. DESIGN BASIS OPTIMAL RETROFIT PROGRAMS
Total Retrofit Cost
NPV and NB
Aggregate BIC
No. Bldgs. Retrofit
Dam. Avoid. (PGA=.2)
Lives Saved (PGA=.2)
EXPECTED VALUE RETROFIT
$ 665.6 H
$ 2,646. H
4.97
846 bldgs
$ 559.6 H
11,943 lives
DESIGN BASIS RETROFIT
$ 953.6 H
$ 28, 306. H
30.68
903 bldgs
$ 646. H
12,717 lives
As can be seen from Table LXXVII, there is not a
large difference between these two retrofit programs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The high value of life "penetrates" the benefit
reduction caused by the earthquake probabilities in the
expected value approach. The great majority of the
live-safety hazardous buildings are captured in both
retrofit programs.
239
The high value of life "penetrates" the benefit
reduction caused by the earthquake probabilities in the
expected value approach. The great majority of the
live-safety hazardous buildings are captured in both
retrofit programs.
239
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER X
SUMMARY AND CONCLUSIONS
EARTHQUAKE LOSS AND RETROFIT MODELING
The "average" building damage estimates provided by
ATC-13 (1985) were extended in the present research to
include earthquake-sensitive effects of variations in
design, construction, and site-specific soil conditions
which are commonly found in buildings. The structural
scoring system and performance modifiers provided in
ATC-21 (1988) were used as the basis of these variations.
Probability theory was used to derive a relationship
between Structural Score (S) and Damage Factor (DF). This
relationship was transformed to fragility curve families,
providing a relationship between Peak Ground Acceleration
(PGA) and DF, based on the performance modifiers in
individual buildings.
Site-specific soil effects were included in the
damage model performance modifiers. Local earthquake
hazard maps were used as the basis of assigning
performance modifiers to account for the effect on damage
of ground motion amplification, liquefaction/lateral
spread risk, and dynamic slope instability risk.
The Damage Factor was used to calculate the
CHAPTER X
SUMMARY AND CONCLUSIONS
EARTHQUAKE LOSS AND RETROFIT MODELING
The "average" building damage estimates provided by
ATC-13 (1985) were extended in the present research to
include earthquake-sensitive effects of variations in
deSign, construction, and site-specific soil conditions
which are commonly found in buildings. The structural
scoring system and performance modifiers provided in
ATC-21 (1988) were used as the basis of these variations.
Probability theory was used to derive a relationship
between Structural Score (S) and Damage Factor (DF). This
relationship was transformed to fragility curve families,
providing a relationship between Peak Ground Acceleration
(PGA) and DF, based on the performance modifiers in
individual buildings.
Site-specific soil effects were included in the
damage model performance modifiers. Local earthquake
hazard maps were used as the basis of assigning
performance modifiers to account for the effect on damage
of ground motion amplification, liquefaction/lateral
spread risk, and dynamic slope instability risk.
The Damage Factor was used to calculate the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
241
probabilistic dollar value of building damage, loss of
life, and serious injuries in a building, based on the
occupancy type and an estimated range of number of persons
in the building while in use.
The retrofit effectiveness values provided in
FEMA-227 (1992) were extended to estimate the beneficial
avoided building damage and lives saved following retrofit
of buildings. Retrofit costs were estimated from
published values in FEMA-156 (1994) and Goettel and Horner
(1995).
REGIONAL RETROFIT PLAHHIHG MODEL
A system was developed to classify buildings in a
region, based on similarities in expected earthquaKe loss
and retrofit cost. Individual buildings from a Rapid
Visual Screening survey are aggregated into one of 540
possible building classes based on structural type, soil
hazards, structural score, occupancy type, and size.
A cost-benefit analysis model was developed for the
building classes, including the value of avoided damage,
value of lives saved and serious injuries avoided, and
cost of structural and non-structural retrofit. The cost-
benefit model was developed for a scenario earthquaKe
event, and for a range of earthquaKe events using an
expected value approach.
241
probabilistic dollar value of building damage, loss of
life, and serious injuries in a building, based on the
occupancy type and an estimated range of number of persons
in the building while in use.
The retrofit effectiveness values provided in
FEMA-227 (1992) were extended to estimate the beneficial
avoided building damage and lives saved following retrofit
of buildings. Retrofit costs were estimated from
published values in FEMA-156 (1994) and Goettel and Horner
(1995).
REGIONAL RETROFIT PLANNING MODEL
A system was developed to classify buildings in a
region, based on similarities in expected earthquaKe loss
and retrofit cost. Individual buildings from a Rapid
Visual Screening survey are aggregated into one of 540
possible building classes based on structural type, soil
hazards, structural score, occupancy type, and size.
A cost-benefit analysis model was developed for the
building classes, including the value of avoided damage,
value of lives saved and serious injuries avoided, and
cost of structural and non-structural retrofit. The cost-
benefit model was developed for a scenario earthquaKe
event, and for a range of earthquaKe events using an
expected value approach.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
242
A technique was developed to prioritize the building
classes, so that the most cost-effective buildings are
retrofit first in a regional plan.
A system efficiency analysis was developed to
determine the optimum retrofit investment for a region.
COMPUTER SOFTWARE DEVELOPMENT
Regional EArthquake Loss and Retrofit Analysis
Program (REAL-RAP) was developed to conduct the loss
estimates and retrofit analysis for a regional ATC-21
(1988) survey inventory.
The REAL-RAP program is composed of modules, which
operate sequentially. Data files of intermediate results
are made at each module step, to enable thorough checking
of results, and processing of large amounts of data on
micro-computers. The modular design philosophy also
permits additions or adjustments to the program to be more
easily integrate~
EARTHQUAKE LOSS ESTIMATE OF THE PORTLAND QUADRANGLE
ATC-21 survey inventory data collected from more
than 7,500 buildings in the Portland Quadrangle during the
1993 Portland Seismic Hazards survey was analyzed with
REAL-RAP.
242
A technique was developed to prioritize the building
classes, so that the most cost-effective buildings are
retrofit first in a regional plan.
A system efficiency analysis was developed to
determine the optimum retrofit investment for a region.
COMPUTER SOFTWARE DEVELOPMENT
Regional EArthquake Loss and Retrofit Analysis
Program (REAL-RAP) was developed to conduct the loss
estimates and retrofit analysis for a regional ATC-21
(1988) survey inventory.
The REAL-RAP program is composed of modules, which
operate sequentially. Data files of intermediate results
are made at each module step, to enable thorough checking
of results, and processing of large amounts of data on
micro-computers. The modular design philosophy also
permits additions or adjustments to the program to be more
easily integrated.
EARTHQUAKE LOSS ESTIMATE OF THE PORTLAND QUADRANGLE
ATC-21 survey inventory data collected from more
than 7,500 buildings in the Portland Quadrangle during the
1993 Portland Seismic Hazards Survey was analyzed with
REAL-RAP.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The loss estimate focused on the Design Basis
Earthquake Scenario, PGA = 0.2 g. The results of this
estimate were:
Building Damage
Loss of Life
$ 2. 7 Billion
5,294 to 12,995
243
The loss of life varies directly with the building
occupant load. The range of loss of life shown above
derives from the lowpoint to the midpoint of the range of
the estimated number of occupants, from the field survey.
Over two-thirds of the loss of life was in
unreinforced masonry or unreinforced masonry infill
buildings (URM and C3/S5). Nearly one quarter of the loss
of life was in concrete frame buildings (C1).
The total dollar value of loss, including the value
of life, was $ 31.9 Billion. Over ninety-percent of the
total dollar value of loss is attributable to the value of
lost human life.
Nearly all of the loss of life occurred in buildings
with a Structural Score of less than 1.0. Over half of
the loss of life occurred in the Commercial, Office, and
Industrial occupancies. Nearly one quarter of the loss of
life occurred in the School and Public Assembly
occupancies. There was not a significant correlation
between loss of life and Soil Modification Factor.
Ten-percent of the buildings in the inventory have a
The loss estimate focused on the Design Basis
Earthquake Scenario, PGA = 0.2 g. The results of this
estimate were:
Building Damage
Loss of Life
$ 2.7 Billion
5,294 to 12,995
243
The loss of life varies directly with the building
occupant load. The range of loss of life shown above
derives from the lowpoint to the midpoint of the range of
the estimated number of occupants, from the field survey.
Over two-thirds of the loss of life was in
unreinforced masonry or unreinforced masonry infill
buildings (URM and C3/S5). Hearly one quarter of the loss
of life was in concrete frame buildings (C1).
The total dollar value of loss, including the value
of life, was $ 31.9 Billion. Over ninety-percent of the
total dollar value of loss is attributable to the value of
lost human life.
Hearly all of the loss of life occurred in buildings
with a Structural Score of less than 1.0. Over half of
the loss of life occurred in the Commercial, Office, and
Industrial occupancies. Hearly one quarter of the loss of
life occurred in the School and Public Assembly
occupancies. There was not a significant correlation
between loss of life and Soil Modification Factor.
Ten-percent of the buildings in the inventory have a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
244
damage factor above 50 (major damage or collapse). For
comparison to a similar earthquake in a modern city, the
1994 Kobe earthquake had 12. 6-percent of the buildings at
damage factor 50 or greater.
When REAL-RAP was run without any performance
modifiers in the model (corresponding to the ~(PMF) = 0
line on the fragility curves), the midpoint loss of life
dropped from 12,995 to 477. This indicates that more than
95-percent of the loss of life derives from the variations
in design and construction as modeled in the performance
modifiers.
RETROFIT ANALYSIS OF THE PORTLAND QUADRANGLE
The REAL-RAP software was first used to determine the
result of a "full" retrofit retrofitting every building
in the inventory. For the design basis earthquake (DBE,
PGA = .20), the value of damage was reduced to$ 1. 69
Billion and loss of life was reduced to 256 lives lost, at
a retrofit cost of$ 4.42 Billion. (a 37'l. reduction in
damage and a 98Y. reduction in loss of life).
Hext, the retrofit benefits and costs were
classified and prioritized from high to low B/C ratio for
the Design BAsis Earthquake (DBE). A system efficiency
analysis was made, resulting in an optimal retrofit
program as follows.
244
damage factor above 50 (major damage or collapse). For
comparison to a similar earthquake in a modern city, the
1994 Kobe earthquake had 12.6-percent of the buildings at
damage factor 50 or greater.
When REAL-RAP was run without any performance
modifiers in the model (corresponding to the E(PMF) = 0
line on the fragility curves), the midpoint loss of life
dropped from 12,995 to 477. This indicates that more than
95-percent of the loss of life derives from the variations
in design and construction as modeled in the performance
modifiers.
RETROFIT ANALYSIS OF THE PORTLAND QUADRANGLE
The REAL-RAP software was first used to determine the
result of a "full" retrofit retrofitting every building
in the inventory. For the design basis earthquake (DBE,
PGA = .20), the value of damage was reduced to $ 1.69
Billion and loss of life was reduced to 256 lives lost, at
a retrofit cost of $ 4.42 Billion. (a 371. reduction in
damage and a 98Y. reduction in loss of life).
Uext, the retrofit benefits and costs were
classified and prioritized from high to low B/C ratio for
the Design BAsis Earthquake (DBE). A system efficiency
analysis was made, resulting in an optimal retrofit
program as follows.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
245
$ 953. 6 Million retrofit investment lives saved
damage avoided number of buildings
12,717 (based on midpoint) $ 646. Hi 11 ion 903
The retrofit prioritization by building class also
allows a maximally efficient "partial" retrofit program to
be determined. The utility of this can be seen by loo~ing
at a retrofit program that provides 50Y. of the savings in
life of the optimal program, as follows.
retrofit investment lives saved
damage avoided number of buildings
$ 171. Million 6,393 (based on midpoint) $ 152. Million 272
For every dollar spent on retrofit, $ o. 68 is saved
in reduced building damage. The savings in human life is
about the same as the full retrofit, but at one-fifth the
cost. Of the building types contained in the optimal
retrofit, two-thirds are unreinforced masonry (URM).
Significant numbers of concrete frame (Ci), unreinforced
masonry infill (C3/S5), tilt-up (PC1), and reinforced
masonry (RM) buildings are also included. Over 75Y. of the
buildings in the optimal retrofit program are in the
office-commercial-historical-industrial occupancy types,
and 98-percent have a structural score of less than +1.0.
It is concluded that the building classification
system and cost-benefit analysis was successful in
selecting for retrofit those buildings (about 10-percent of
the inventory) that are life-safety critical, while greatly
245
$ 953. 6 Million retrofit investment lives saved
damage avoided number of buildings
12,717 (based on midpoint) $ 646. Million 903
The retrofit prioritization by building class also
allows a maximally efficient "partial" retrofit program to
be determined. The utility of this can be seen by loo~ing
at a retrofit program that provides 50Y. of the savings in
life of the optimal program, as follows.
retrofit investment lives saved
damage avoided number of buildings
$ 171. Million 6,393 (based on midpoint) $ 152. Million 272
For every dollar spent on retrofit, $ 0.68 is saved
in reduced building damage. The savings in human life is
about the same as the full retrofit, but at one-fifth the
cost. Of the building types contained in the optimal
retrofit, two-thirds are unreinforced masonry (URM).
Significant numbers of concrete frame (Ci), unreinforced
masonry infill (C3/S5), tilt-up (PC1), and reinforced
masonry (RM) buildings are also included. Over 75Y. of the
buildings in the optimal retrofit program are in the
office-commercial-historical-industrial occupancy types,
and 98-percent have a structural score of less than +1.0.
It is concluded that the building classification
system and cost-benefit analysis was successful in
selecting for retrofit those buildings (about iO-percent of
the inventory) that are life-safety critical, while greatly
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
246
reducing the retrofit investment from the full retrofit
program. The choice of an ATC-21 structural score of +1.0
as a classification index was successful in identifying the
important buildings for retrofit.
The REAL-RAP software was also used to make an
expected value analysis of retrofit. Although the present
values of retrofit benefit are only one-tenth those of the
Design Basis EarthquaKe, the PV/C (present value to cost)
ratios are still high enough to capture most of the
life-safety critical buildings in the optimal retrofit
program. The optimal retrofit program based on the
expected value analysis is as follows:
retrofit investment lives saved
damage avoided number of buildings
$ 665.6 Million 11,943 (based on midpoint) $ 559.6 Million 846
The categories of buildings in the expected value
optimal retrofit are roughly the same as the DBE program.
It is concluded that seismic retrofit for Portland
has a very high B/C ratio, and it makes sense to retrofit.
Variations in the number of building occupants and dollar
value of human life do not change this conclusion. The
same conclusion is reached whether a scenario earthquake
event or an expected value anaysis of retrofit benefit are
used
The initial B/C ratios, at the "top" of the
prioritized building classes, approach 1,000 (for the DBE
246
reducing the retrofit investment from the full retrofit
program. The choice of an ATC-21 structural score of +1.0
as a classification index was successful in identifying the
important buildings for retro~it.
The REAL-RAP software was also used to make an
expected value analysis of retrofit. Although the present
values of retrofit benefit are only one-tenth those of the
Design Basis Earthquake, the PV/C (present value to cost)
ratios are still high enough to capture most of the
life-safety critical buildings in the optimal retrofit
program. The optimal retrofit program based on the
expected value analysis is as follows:
retrofit investment lives saved
damage avoided number of buildings
$ 665.6 Million 11.943 (based on midpoint) $ 559.6 Million 846
The categories of buildings in the expected value
optimal retro~it are roughly the same as the DBE program.
It is concluded that seismic retrofit for Portland
has a very high B/C ratio, and it makes sense to retrofit.
Variations in the number of building occupants and dollar
value of human life do not change this conclusion. The
same conclusion is reached whether a scenario earthquake
event or an expected value anaysis of retrofit benefit are
used
The initial B/C ratios, at the "top" of the
prioritized building classes, approach 1,000 (for the DBE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
247
retrofit). The aggregate B/C ratio for the optimal DBE is
about 30. With such high initial B/C ratios, the question
is not whether to retrofit, but where to "cut-off" a
retrofit program. The optimal retrofit program is one
answer to that question.
LIMITATIONS IN APPLYING THIS RESEARCH
In applying the results of this research, it is
important to understand the probabilistic roots of the
loss estimation techniques developed.
This methodology is intended to evaluate hundreds or
thousands of buildings in a region, and is applicable only
to categories or groups of buildings. This methodology is
definitely not intended to fully determine the seismic
safety of individual buildings.
The methodology developed herein is intended to be
the first step of a several step process in identifying
those buildings that require additional, more detailed
investigation by qualified engineers. The goal is to
broadly identify most of the potentially seismically
hazardous buildings, and to eliminate most of the
relatively adequate buildings from further review.
The utility of this methodology in taKing this first
step in an actual retrofit program is dependent upon the
qualifications of the responsible building officials, and
247
retrofit). The aggregate BIC ratio for the optimal DBE is
about 30. With such high initial BIC ratios, the question
is not whether to retrofit, but where to "cut-off" a
retrofit program. The optimal retrofit program is one
answer to that question.
LIMITATIONS IN APPLYING THIS RESEARCH
In applying the results of this research, it is
important to understand the probabilistic roots of the
loss estimation techniques developed.
This methodology is intended to evaluate hundreds or
thousands of buildings in a region, and is applicable only
to categories or groups of buildings. This methodology is
definitely not intended to fully determine the seismic
safety of individual buildings.
The methodology developed herein is intended to be
the first step of a several step process in identifying
those buildings that require additional, more detailed
investigation by qualified engineers. The goal is to
broadly identify most of the potentially seismically
hazardous buildings, and to eliminate most of the
relatively adequate buildings from further review.
The utility of this methodology in taking this first
step in an actual retrofit program is dependent upon the
qualifications of the responsible building officials, and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24-8
the quality o£ the input information. Field survey crews
must be qualified, well trained, and adequately supervised
and checked by professional engineering personnel. The
soil hazard mapping information is approximate in nature;
its detail and quality can be expected to vary greatly
£rom community to community, and should be scrutinized by
professional engineering personnel qualified to judge its
value. The building o££icials in responsible charge o£
implementing this information should have the technical
background to £ully understand these limitations.
The loss estimates and retrofit analyses made in
this research consider the buildings in "£ull use" during
regular business hours. Additionally, the e££ect o£
foundation type (deep vs. shallow) are not considered in
the ATC-21 inventory nor the loss/retrofit models.
RECOMMENDATIONS FOR FURTHER RESEARCH
This methodology bases its loss estimates and
retrofit recommendations on a 11 £ar-£ield" seismic source,
wherein every building in the region is considered to
subject to the same PGA.
For a local earthquake source, near-£ield
attenuation e££ects would result in di££erent values o£
PGA £or di££erent buildings, depending upon their
individual proximity to the £ault rupture. In Portland,
248
the quality of the input information. Field survey crews
must be qualified, well trained, and adequately supervised
and checked by professional engineering personnel. The
soil hazard mapping information is approximate in naturej
its detail and quality can be expected to vary greatly
from community to community, and should be scrutinized by
professional engineering personnel qualified to judge its
value. The building officials in responsible charge of
implementing this information should have the technical
background to fully understand these limitations.
The loss estimates and retrofit analyses made in
this research consider the buildings in "full use" during
regular business hours. Additionally, the effect of
foundation type (deep vs. shallow) are not considered in
the ATC-21 inventory nor the loss/retrofit models.
RECOMMENDATIONS FOR FURTHER RESEARCH
This methodology bases its loss estimates and
retrofit recommendations on a "far-field" seismic source,
wherein every building in the region is considered to
subject to the same PGA.
For a local earthquake source, near-field
attenuation effects would result in different values of
PGA for different buildings, depending upon their
individual proximity to the fault rupture. In Portland,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
249
for example, a west-hills fault rupture would likely cause
buildings downtown to have PGA values higher than the DBE,
while buildings in east Portland may be at or below the
DBE level of PGA. The resulting loss estimates and
retrofit programs may be different, depending upon where
the hazardous buildings are relative to damaging levels of
PGA.
A recommended next step in this model will be to
"map" near-field PGA values to the individual buildings.
An attenuation model could be linked to the building
inventory via parcel numbers and GIS technology.
The present model would be adjusted so that a PGA
value individual to each building will be used to
determine that building's damage factor. This would
require the addition of modules preceding the DAM-FACT
module, plus adjustments in the DAM-FACT module. The
remaining REAL-RAP modules would require little if any
additional work.
Further research also needs to be done on the
estimated damage from long-duration subduction zone
earthquakes. The damage estimates used (ATC-13 1985) are
based on experince with short-duration crustal events.
Very little information is available on the effect of the
long-duration shaking, which can be up to 3-minutes long.
Research also needs to be done on the synergistic
249
for example, a west-hills fault rupture would likely cause
buildings downtown to have PGA values higher than the DBE,
while buildings in east Portland may be at or below the
DBE level of PGA. The resulting loss estimates and
retrofit programs may be different, depending upon where
the hazardous buildings are relative to damaging levels of
PGA.
A recommended next step in this model will be to
"map" near-field PGA values to the individual buildings.
An attenuation model could be linked to the building
inventory via parcel numbers and GIS technology.
The present model would be adjusted so that a PGA
value individual to each building will be used to
determine that building's damage factor. This would
require the addition of modules preceding the DAM-FACT
module, plus adjustments in the DAM-FACT module. The
remaining REAL-RAP modules would require little if any
addi tional work.
Further research also needs to be done on the
estimated damage from long-duration subduction zone
earthquakes. The damage estimates used (ATC-13 1985) are
based on experince with short-duration crustal events.
Very little information is available on the effect of the
long-duration shaking, which can be up to 3-minutes long.
Research also needs to be done on the synergistic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
250
effects of building losses and lifeline-utility losses, as
discussed in Chapter I, realizing that building losses are
one part of the larger puzzle.
The loss estimation methodology developed in the
present research has potential as a "pre-processor" to
other regional loss estimation methodologies which still
look only at "average" buildings, such as the recently
developed HAZUS program (HIBS 1995).
The methodology developed herein is not
experimentally verifiable without monumental cost.
However, the loss estimating and retrofit effectiveness
factors could be adjusted and refined based on detailed
reconnaissance information from actual earthquakes.
Therefore, reconnaissance data gathering from future
earthquakes should be more detailed to aid this type of
refinement.
250
effects of building losses and lifeline-utility losses, as
discussed in Chapter I, realizing that building losses are
one part of the larger puzzle.
The loss estimation methodology developed in the
present research has potential as a "pre-processor" to
other regional loss estimation methodologies which still
look only at "average" buildings, such as the recently
developed HAZUS program (HIBS 1995).
The methodology developed herein is not
experimentally verifiable without monumental cost.
However, the loss estimating and retrofit effectiveness
factors could be adjusted and refined based on detailed
reconnaissance information from actual earthquakes.
Therefore, reconnaissance data gathering from future
earthquakes should be more detailed to aid this type of
refinement.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES CITED
Aki, K., 1988, "Local Site Effects on Strong Ground Motion," Earthquake Engineering and Soil Dynamics II: Recent Advances in Ground Motion Evaluation, ASCE Geotechnical Special Publication 20, pp. 103-155.
Applied Technology Council (ATC-13), 1985, Earthquake Damage Evaluation Data for California, FEMA Contract No. EMW-C-0912, Redwood City, California.
Applied Technology Council (ATC-21), 1988, Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook, FEMA-154, Redwood City, California.
Applied Technology Council (ATC-21-1), 1988, Rapid Visual Screening of Buildings for Potential Seismic Hazards: Supporting Documentation, FEMA-155, Redwood City, California.
Atwater, B. F., 1987, "Evidence for Great Holocene Earthquakes Along the Outer Coast of Washington State, " Science, V. 236, pp. 942-944.
Bartlett, S.F., and Youd, T.L., 1992, "Empirical prediction of lateral spread displacement", US-Japan Workshop on Earthquake Resistant Design Lifeline Facilities and Countermeasures for Soil Liquefaction, 4th, Honolulu, Hawaii, Proceedings, National Center for Earthquake Engineering Research Technical Report NCEER-92-0019, v. 2, p. 351-366.
Benuska, L., 1990, Earthquake Spectra Supplement to Volume 6: Lorna Prieta Earthquake Reconnaissance Report, Earthquake Engineering Research Institute, El Cerrito, California.
Building Seismic Safety Council, 1988, NEHRP Recommended Provisions for the Development of Seismic Regulations for Hew Buildings, Washington, D.C.
Comartin, C. D., Greene, M., and Tubbesing, S.K., 1995, The Hyogo-Ken Hanbu Earthquake: Preliminary Reconnaissance Report, Earthquake Engineering Research Institue, Oakland, California.
REFERENCES CITED
Aki, K., 1988, "Local Site Effects on Strong Ground Motion," Earthquake Engineering and Soil Dynamics II: Recent Advances in Ground Motion Evaluation, ASCE Geotechnical Special Publication 20, pp. 103-155.
Applied Technology Council (ATC-13), 1985, Earthquake Damage Evaluation Data for California, FEHA Contract No. EMW-C-0912, Redwood City, California.
Applied Technology Council (ATC-l1), 1988, Rapid Visual Screening of Buildings for Potential Seismic Hazards: A Handbook, FEHA-154, Redwood City, California.
Applied Technology Council (ATC-21-1), 1988, Rapid Visual Screening of Buildings for Potential Seismic Hazards: Supporting Documentation, FEMA-155, Redwood City, Cal ifornia.
Atwater, B.F., 1987, "Evidence for Great Holocene Earthquakes Along the Outer Coast of Washington State," SCience, V. 236, pp. 942-944.
Bartlett, S.F., and Youd, T.L., 1992, "Empirical prediction of lateral spread displacement", US-Japan Workshop on Earthquake Resistant Design Lifeline Facilities and Countermeasures for Soil Liquefaction, 4th, Honolulu, Hawaii, Proceedings, National Center for Earthquake Engineering Research Technical Report NCEER-92-0019, v. 2, p. 351-366.
Benuska, L., 1990, Earthquake Spectra Supplement to Volume 6: Loma Prieta Earthquake Reconnaissance Report, Earthquake Engineering Research Institute, El Cerrito, California.
Building Seismic Safety CounCil, 1988, NEHRP Recommended Provisions for the Development of Seismic Regulations for Hew Buildings, Washington, D.C.
Comartin, C.D., Greene, M., and Tubbesing, S.K., 1995, The Hrogo-Ken Nanbu Earthquake: Preliminary Reconnaissance Report, Earthquake Engineering Research Institue, Oakland, California.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
252
Federal Emergency Management Agency (FEMA-172), 1989, Techniques for SeismicallY Rehabilitating Existing Buildings <Preliminary>, URS/John A. Blume & Associates, Washington, D.C.
Federal Emergency Management Agency (FEMA-178), 1992, NEHRP HandbooK for the Seismic Evaluation of Existing Buildings, Building Seismic Safety Council, Washington, D.C.
Federal Emergency Management Agency (FEMA-227 & 228), 1992, A Benefit-Cost Model for the Seismic Rehabilitation of Buildings, VSP Associates, Vol. I & II, Washington, D.C.
Federal Emergency Management Agency (FEMA-156), 1994, Typical Costs for Seismic Rehabilitation of Existing Buildings, Second Edition, Vols. 1 and 2, Hart Consultant Group, Washington, D.C.
Goettel & Horner Inc., 1995, EarthquaKe RisK Analysis, Final Report, Vol. 1 & 2, Submitted to the City of Portland, Sacramento, California.
Geomatrix Consultants, 1995 Seismic Design Mapping, State of Oregon, Final Report, ODOT Contract 11688, San Francisco, California.
International Conference of Building Officials, 1991, Uniform Building Code, Whittier, California.
Oregon Department of Geology & Mineral Industries (GMS-79), 1993, EarthquaKe Hazard Maps of the Portland Quadrangle, Portland, Oregon.
National Bureau of Standards (NBS 55), 1964, HandbooK of Mathematical Functions, Applied Mathematics Series 55, Washington, D.C.
National Institue of Building Sciences, 1995, Development of a Standardized EarthquaKe Loss Estimation Methodology, Volume I, Draft Technical Manual 100/. Submittal, RisK Management Solutions, Inc., Menlo ParK, California.
Rad, F.N., & McCormacK, T.C., 1994, Rapid Visual Screening of 4500 Buildings in Portland, submitted to Metro, Portland, Oregon.
252
Federal Emergency Management Agency (FEMA-172), 1989, Techniques for Seismically Rehabilitating EXisting Buildings (Preliminary), URS/John A. Blume & Associates, Washington, D.C.
Federal Emergency Management Agency (FEMA-178), 1992, NEHRP Handbook for the Seismic Evaluation of EXisting Buildings, Building Seismic Safety Council, Washington, D.C.
Federal Emergency Management Agency (FEMA-227 & 228), 1992, A Benefit-Cost Model for the Seismic Rehabilitation of Buildings, VSP Associates, Vol. I & II, Washington, D.C.
Federal Emergency Management Agency (FEMA-156), 1994, Typical Costs for Seismic Rehabilitation of Existing Buildings, Second Edition, Vols. 1 and 2, Hart Consultant Group, Washington, D.C.
Goettel & Horner Inc., 1995, Earthquake Risk Analysis, Final Report, Vol. 1 & 2, Submitted to the City of Portland, Sacramento, California.
Geomatrix Consultants, 1995 Seismic Design Mapping, State of Oregon, Final Report, ODOT Contract 11688, San FranCiSCO, California.
International Conference of Building Officials, 1991, Uniform Building Code, Whittier, California.
Oregon Department of Geology & Mineral Industries (GMS-79), 1993, Earthquake Hazard Maps of the Portland Quadrangle, Portland, Oregon.
National Bureau of Standards (NBS 55), 1964, Handbook of Mathematical Functions, Applied Mathematics Series 55, Washington, D.C.
National Institue of Building Sciences, 1995, Development of a Standardized Earthquake Loss Estimation Methodology, Volume I, Draft Technical Manual 1001. Submittal, Risk Management Solutions, Inc., Menlo Park, California.
Rad, F.N., & McCormack, T.C., 1994, Rapid Visual Screening of 4500 Buildings in Portland, submitted to Metro, Portland, Oregon.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
253
Richter, c. F. , and Co.,
1958, Elementary Seismology, W.H. Freeman San Francisco, California.
Robertshaw, J.E., Mecca, S.J., and RericK, H.N., 1978, Problem Solving: A systems Approach, Petrocelli BooKs Inc., New YorK, H.Y.
Schnable, P.B., Lysmer, J., and Seed, H. B., 1972, SHAKE- A computer program for earthquaKe analysis of horizontally layered sites, EarthquaKe Engineering Research Center Report EERC 72-12, BerKeley, California, University of California.
Tri-Het, 1993, Line Section 5A Geotechnical Interpretive Report, Westside Corridor Proiect, Tri-County Metropolitan Transportation District of Oregon, Portland, Oregon.
Wood, H.O., and Newmann, F., 1931, "Modifed Mercalli Intensity Scale of 1931," Seismological Society of America Bulletin, Vol. 21, Ho. 4, pp. 277-283.
Wyllie, L.A., and Filson, J.R., 1989, EarthquaKe Spectra Special Supplement: Armenia EarthquaKe Reconnaissance Report, EarthquaKe Engineering Research Institute, El Cerrito, California.
253
Richter, C. F. , and Co.,
1958, Elementary Seismology, W.H. Freeman San Francisco, California.
Robertshaw, J.E., Mecca, S.J., and RericK, M.N., 1976, Problem Solving: A Systems Approach, Petrocelli BooKs Inc., New YorK, H.Y.
Schnable, P.B., Lysmer, J., and Seed, H.B., 1972, SHAKE - A computer program for earthquaKe analysis of horizontally layered Sites, EarthquaKe Engineering Research Center Report EERC 72-12, BerKeley, California, University of California.
Tri-Met, 1993, Line Section 5A Geotechnical Interpretive Report, Westside Corridor Project, Tri-County Metropolitan Transportation District of Oregon, Portland, Oregon.
Wood, H.O., and Newmann, F., 1931, "Modifed Mercalli IntenSity Scale of 1931," Seismological SOCiety of America Bulletin, Vol. 21, No.4, pp. 277-283.
Wyllie, L.A., and Filson, J.R., 1969, EarthquaKe Spectra Special Supplement: Armenia EarthquaKe Reconnaissance Report, EarthquaKe Engineering Research Institute, El Cerrito, California.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX A
POWER CURVE COEFFICIENTS FROM MEAN DAMAGE FACTOR VS. STRUCTURAL SCORE REGRESSIONS
APPENDIX A
POWER CURVE COEFFICIENTS FROM MEAN DAMAGE FACTOR VS. STRUCTURAL SCORE REGRESSIONS
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
255
W - WOOD FRAME
MMI PGA BSH b(O) b ( 1)
VI . 05 7.08 18. 20052 -1. 10604
VII . 10 8.46 24. 51270 -0.82999
VIII . 22 4. 76 25. 86136 -0.78016
IX . 47 3. 75 30. 20206 -0.63640
X 1. 02 1. 48 29.77830 -0.64946
S1 - STEEL MRF
MMI PGA BSH b(O) b ( 1)
VI . 05 2. 24 4. 80857 -2. 33550
VII . 10 5.42 19.47400 -1.04302
VIII . 22 6. 12 27. 30582 -0.72981
IX . 47 5.42 31. 92183 -0.58500
X 1. 02 3.43 34. 13045 -0.52297
255
W - WOOD FRAME
HMI PGA BSH b(O) b (1 )
VI .05 7.06 16.20052 -1. 10604
VII . 10 6.46 24.51270 -0. 82999
VIII .22 4.76 25.86136 -0.78016
IX .47 3. 75 30. 20206 -0. 63640
X 1. 02 1. 48 29.77830 -0.64946
S1 - STEEL MRF
HMI PGA BSH b(O) b (1)
VI .05 2. 24 4. 80857 -2. 33550
VII .10 5.42 19.47400 -1. 04302
VIII .22 6.12 27.30582 -0.72981
IX .47 5.42 31. 92183 -0.58500
X 1. 02 3.43 34.13045 -0.52297
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MMI
VI
VII
VIII
IX
X
MMI
VIII
IX
X
PGA
. 05
. 10
. 22
. 47
1. 02
PGA
. 22
. 47
1. 02
S2 - BRACED STL FRAME
BSH b(O)
2. 33 4.80857
3. 76 14. 11023
5. 30 24.92447
5.04 29. 81674
3. 36 32.00781
S3 - LIGHT METAL
BSH
8.08
5.40
3. 24
b(O)
25.43586
28. 38696
31. 80362
256
b ( 1)
-2. 33550
-1. 34190
-0. 81451
-0. 64834
-0. 58252
b ( 1)
-0. 79565
-0.69388
-0.58847
MMI
VI
VII
VIII
IX
X
MMI
VIII
IX
X
PGA
.05
.10
.22
. 47
1. 02
PGA
.22
. 47
1. 02
S2 - BRACED STL FRAME
BSH b(O)
2. 33 4.80857
3. 76 14. 11023
5. 30 24.92447
5.04 29.81674
3. 36 32.00781
S3 - LIGHT METAL
BSH
8.08
5.40
3. 24
b(O)
25.43586
28. 38696
31. 80362
256
b ( 1 )
-2. 33550
-1. 34190
-0. 81451
-0. 64834
-0. 58252
b (1)
-0. 79565
-0.69388
-0.58847
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
257
S4 - STL FRAME W/ CONC SHEARWALLS
MMI PGA BSH b(O) b ( 1)
VI . 05 9. 61 24. 69128 -0. 82320
VII . 10 6. 17 25.13170 -0.80662
VIII . 22 4.73 30.47236 -0.62809
IX . 47 3.49 34. 70487 -0. 50750
X 1. 02 1. 80 34.47071 -0. 51382
C1 - REINF CONC MRF
MMI PGA BSH b(O) b(1)
VI . 05 5. 12 18.43037 -1.09442
VII . 10 4.86 26.75175 -0. 74878
VIII . 22 2. 56 30.61683 -0. 62376
IX . 47 1. 33 35. 18993 -0.49472.
X 1. 02. o. 65 37.23838 -0.442.21
257
S4 - STL FRAME WI CONC SHEARWALLS
MMI PGA BSH b(O) b (1)
VI .05 9. 61 24.69128 -0. 82320
VII .10 6. 17 25.13170 -0.80662
VIII .22 4.73 30.47236 -0.62809
IX .47 3.49 34. 70487 -0. 50750
X 1. 02 1. 80 34.47071 -0.51382
C1 - REINF CONC MRF
MMI PGA BSH b(O) b(1 )
VI .05 5.12 16.43037 -1.09442
VII .10 4.86 26.75175 -0. 74878
VIII .22 2. 56 30.61683 -0. 62376
IX .47 1. 33 35.18993 -0.49472
X 1. 02 O. 65 37.23838 -0.44221
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
258
C2 - REINF CONC SHEARWALL
MMI PGA BSH b(O) b( 1)
VI . 05 7. 18 18. 89713 -1. 07089
VII . 10 3.85 20.43563 -0. 99845
VIII . 22 3. 80 29.02565 -0. 67320
IX . 47 2.25 32.27684 -0. 57477
X 1. 02 1. 21 37.07823 -0.44649
C3/S5 - URM INFILL
MMI PGA BSH b(O) b( 1)
VI . 05 3.06 14.40824 -1. 32260
VII . 10 2.27 19. 10022 -1. 06120
VIII . 22 1. 34 24.07250 -0. 84667
IX . 47 0.76 29.89975 -0.64564
X 1. 02 o. 34 34.03400 -0. 52571
258
C2 - REINF CONC SHEARWALL
MMI PGA BSH b(O) b (1)
VI .05 7. 18 18.89713 -1. 07089
VII . 10 3.85 20.43563 -0. 99845
VIII .22 3.80 29.02565 -0. 67320
IX .47 2.25 32.27684 -0. 57477
X 1. 02 1. 21 37.07823 -0.44649
C3/S5 - URM INFILL
MMI PGA BSH b(O) b(1)
VI .05 3.06 14.40824 -1. 32260
VII .10 2.27 19. 10022 -1. 06120
VIII .22 1. 34 24.07250 -0. 84667
IX .47 0.76 29.89975 -0.64564
X 1. 02 0.34 34.03400 -0. 52571
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
259
PC1 - TILT UP
MMI PGA BSH b(O) b ( 1)
VI . 05 5. 98 20. 10624 -1. 01358
VII . 10 5. 70 27. 55186 -0. 72151
VIII . 22 3.02 29. 66477 -0.65295
IX . 47 1. 97 33.63610 -0. 53654
X 1. 02 o. 95 34. 75106 -0. 50629
PC2 - PRECAST FRAME
MMI PGA BSH b(O) b ( 1)
VI . 05 3.02 10. 11581 -1. 65050
VII . 10 4. 15 21. 27473 -0. 96109
VIII . 22 2. 75 25.78235 -0. 78297
IX . 47 1. 05 34. 34272 -0.51727
X 1. 02 0. 33 37. 34354 -0.43962
259
PC1 - TILT UP
liM I PGA BSH b(O) b (1)
VI .05 5. 98 20. 10624 -1. 01358
VII .10 5. 70 27. 55186 -0.72151
VIII .22 3.02 29.66477 -0.65295
IX .47 1. 97 33. 63610 -0. 53654
X 1. 02 0.95 34. 75106 -0. 50629
PC2 - PRECAST FRAME
liM I PGA BSH b(O) b (1)
VI .05 3.02 10.11581 -1. 65050
VII .10 4.15 21. 27473 -0. 96109
VIII .22 2.75 25. 78235 -0. 78297
IX .47 1. 05 34. 34272 -0.51727
X 1. 02 O. 33 37.34354 -0.43962
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
260
RM - REINF MASONRY
MMI PGA BSH b(O) b ( 1)
VI . 05 6. 23 17.58770 -1. 13773
VII . 10 6.03 24. 83763 -0.81770
VIII . 22 3. 16 24. 92447 -0.81451
IX . 47 1. 93 27.49434 -0.72342
X 1. 02 1. 35 34. 36127 -0. 51679
URM - UNREINF MASONRY
MMI PGA BSH b (0) b ( 1)
VI . 05 4.69 22. 35209 -0. 91548
VII . 10 2.03 20. 92972 -0. 97639
VIII . 22 1. 24 28. 94623 -0. 67583
IX . 47 0.49 36. 89790 -0.47127
X 1. 02 0.07 42.07165 -0. 32905
260
RM - REINF MASONRY
MMI PGA BSH b(O) b (1)
VI .05 6. 23 17.58770 -1. 13773
VII .10 6.03 24. 83763 -0.81770
VIII .22 3. 16 24. 92447 -0.81451
IX .47 1. 93 27.49434 -0.72342
X 1. 02 1. 35 34. 36127 -0. 51679
URM - UNREINF MASONRY
MMI PGA BSH b (0) b (1)
VI .05 4.69 22. 35209 -0.91548
VII . 10 2.03 20. 92972 -0. 97639
VIII .22 1. 24 28. 94623 -0. 67583
IX .47 0.49 36. 89790 -0.47127
X 1. 02 0.07 42.07165 -0. 32905
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX B
HALF-STEP FRAGILITY CURVES
APPENDIX B
HALF-STEP FRAGILITY CURVES
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0::: 0 1-u <( 60 u. w (!) <( :! <(
40 0 z <( w :!
20
0.0
WOOD - WOOD FRAME NEHRP AREAS 3,4
--~--.P -3.L_
0.2 0.4 0.6 PGAON ROCK
0.8
262
~li
I
i i-0.5
I
i
0::: o l-t>
80
WOOD - WOOD FRAME NEHRP AREAS 3,4
262
~ 60 1-+----~--11 '-1.5 UJ (9 4: :2
; , I
(3 40 -H--i--I-+---i--/--L--
Z « UJ :2
20 -+1---1
I i I i-0.5 ! 1
~-+---/~~~--i-----b-~------!--+-c--4i+0.5
I
II--I--k~~~~~~~=f-==~=+=+=~I 1+1
.5 + I I
i ! I i
o ~==~~~==~~==~~~==~~' I -1 I~
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
80 0:: 0 1-(.) <(
60 LL w (!) <( :E <(
40 0 z <( w :E
20
0
1 i
S1 - STEEL MRF NEHRP AREAS 3,4
-55
t-+-1 I I
I
-4 5
I I /! I I
1/ I I / I It; I I I
- I I . v I il
I v I / I I
j I ! 1/ I / I
I v v / v I / /
1/ VI ~ v /J-------- ~ --I ~~ ~~~ ~~~
--r-- +-~~
~ ~ ~ ~I I ~ iP' I
I I I I '
0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
263
-3.5
-2.5
-1.5 -0.5 +0.5 +1.5
0.0
S1 - STEEL MRF NEHRP AREAS 3,4
-55 -45
0.2 0.4 0.6 0.8
PGAON ROCK
263
1.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82 - BRACED STEEL FRAME NEHRP AREAS 3,4
264
-2.5 UJ +•-~~~~~~~ I
20 -H---t-f---+---+---::~--+--+-=-F----1---±~+-i! -1.5
~~~~~~r~,§~t~~ -0.5 I. m 0 1 \ i
I
0.0 0.2 0.4 0.6 0.8
PGAON ROCK 1.0
S2 - BRACED STEEL FRAME NEHRP AREAS 3,4
80~-~--+-~--+-~--+-~~-+~~
~o I - - .-l-----+----+----+---+-----~'__+_-L
~ 60-H-~--+-~._L_J--~~-+,~--~ W I I (9 I I « :2 C§ 40 --+I--t------+-
Z « w :2
I 0.0 0.2 0.4 0.6 0.8
PGAON ROCK 1.0
254
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I
80 -t· 0:::
f 0 I-(.) I <( I
u. 60 -+ I w I
C) I <(
40 l ~ <( 0
I z <( t w ~
20 -t
0.0
83 - LIGHT METAL NEHRP AREAS 3,4
0.2 0.4 0.6 0.8 PGAON ROCK
265
I i i I
I I I I
I I I I I I
I I
-2.5
-1.5
1.0
I
80 -t· c:: .~ 0 l-() I ~
I
60 -+ u. I
W I C> I ~
40 J ~ ~ 0
I Z ~ t w ~
20 -t
a
0.0
83 - LIGHT METAL NEHRP AREAS 3,4
0.2 0.4 0.6 PGAON ROCK
0.8
265
I i i I I I I I I
I I I I I I
I I
-2.5
-1.5 -0.5 +0.5 +1.5
1.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
84 - STEEL FRAME W/ CONC SHEARWALLS NEHRP AREAS 3,4
0::: 0 I-()
~ 60 -t•--+--1---i---'-1-
w I
-2.5
(!) I ~ T~+-~-+~--~~~-~-F--H
~ 40~1 ~-~-~--~+--h~--+-~~-z <( w ~
20
0 L'
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
266
S4 - STEEL FRAME WI CONC SHEARWALLS NEHRP AREAS 3,4
100 I 1
I
!
80 r:t: 0 l-t) « 60 -t u.. w I (!) I « T :2 « 40 -1 0 z « w :2
20 f
0.2 0.4 0.6 0.8 PGAON ROCK
-2.5 1 =H
I I !
I I I:
I I I
1.0
266
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.0
C1 - REINF CONC MRF NEHRP AREAS 3,4
-2.5 -1.5
0.2 0.4 0.6 0.8 PGAON ROCK
267
1.0
C1 - REINF CONC MRF NEHRP AREAS 3,4
-2.5 -1.5
267
I .J I i
i i
I-i-~+--l+---+--I----+---+--l--.I~ ~5 1-1
J---!-----r +0.5
L-+--+-r--11+1.5
~-~~-4--~~-+~.~
I
0.0 0.2 0.4 0.6 0.8 1.0 PGA ON ROCK
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
268
C2 - REINF CONC SHEARWALL NEHRP AREAS 3,4
-3,5 -2.5 100 -+ -~
I I
I i -r-I I
80 i I
! -I
I
0::: I
i I 0 --- I
~ T I (.) I <(
u.. 60 --+ w
I (9 t <( :2 <(
40 Cl z <( w :2
20 I +-
0 ~\ I I
-j---J I i
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
C2 - REINF CONC SHEARWALL NEHRP AREAS 3,4
r-:==--=+====::;===-3~,5===+=-=:2.~5 ====:;====;===::,;-1 ~.5_==-. ___ ~, 100 -+1-- : II, ) .--,-.1 I I ' ! I I I I· i I I .. .
T
l: - : -I--I---I-~· -t--i-r'
~ I I I I [,
c::: 80 -! --I ; i 111 01- I : I ! r ------'----l-/---,---I---i---r---I Iii () I I t « I i u.. 60 --+1------+----11 : ; W I I[ (9 I I
268
« t'--t--f-t---+-f--+------t-/ r-~ i I 1-0.5 CI 40 ---r+' --I---I--i----'l---,----/-"---I--t-----+----+-~--rl
Z I[ « I . W 1+0.5 ~ 1+1.5
20 I I
+ --I-A-~'¥'"~-=. O-\~~==~+=~~~~~=P 1- i i T
0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
I
80 0::: 0 1-(.) <(
60 1.1.. w C) <(
I
~ <(
40 Cl z <( w ~
20
0
0.0
C3/S5 - URM INFILL NEHRP AREAS 3,4
-1 5 -05
I I
I I I
I
I i
I I I
/I I I I 1/ I
I I I I
! li i I I I i v
! I I I
I
I I
I
I I
I I
111 I
~-I v /i/r __ L : I /
IV/ ~ v 1.----:..--
I ~
~ I I
0.2 0.4 0.6 0.8
PGAON ROCK
269
~
l I I I
I u I ! I I I I
I i i I
1 i
tl I I I I
+0.5
+1.5
I I
I -, 1.0
100
I
80 c::: 0 I-0 « 60 L1.. W C9 « I
~ « 40 Cl Z « w ~
20
o I
0.0
C3/S5 - URM INFILL NEHRP AREAS 3,4
-15 -05
I I I / I
I
I i
/ I I
/1 I I I !/ I I I I I
! Ii i I I I i
V
I I I I
I
I
I
I I I I I
111 I 4i=l / /~r __ L: I /
IV/ ~ v V-:--
I ~
~ I I I I
0.2 0.4 0.6 0.8
PGAON ROCK
269
~
I I I
I I
U I ! I I I I
I i i I
-I i
tl I I I I
+0.5
+1.5
I I
I
I 1.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100 I
80 --
0:: 0 I-(.) <(
60 lJ..
I I
I I
w i
(.!) <( ~ <(
40 0 ~
z <( w ~
20
I
I
0
0.0
I
i
PC1 - TILT UP NEHRP AREAS 3,4
-2 5 I
I I I I
I i I I I I I / I v
I /! I
'/1 I I
i I
I I I
I I I
I
-1 5
v I II I I I i
I
II
I I ~ I
I I
I I I
I I
I I I I I I I
/ vn I / v ~~I
I .........
I VL,/ / I-"'
~~ -~- I --
~ ~ ~ ~I I I
~r, I I I I
I I I I I I I I
0.2 0.4 0.6 0.8 1.0 PGAON ROCK
270
-0.5
+0.5
+1.5
100 I
80 - -
0:: 0 I-0 « 60 lJ..
I I
I I
W i (!) « :2 « 40 0 ~
Z « w :2
20
I I
o 0.0
I
i
I I I
PC1 - TILT UP NEHRP AREAS 3,4
-25 I
I I I I
I i , I I I
I / I I V
I /! I '/1 I I i I
il V I
/ I / I
-1 5
V I ! I
i I I I i
I
II
I I ~ I
I I
I I I
I i I I I I I I I
./" /11
I I I I V ~~I _I / /
/ V~ /'" H:1 ---~ ~ ~ ~I I I
~ \ I I I I ,
I I I I I I I I
0.2 0.4 0.6 0.8 1.0 PGAON ROCK
2.70
-0.5
+0.5
+1.5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0::: 0 I-()
Lt w (!) <( ~ <( 0 z <( w ~
PC2 - PRECAST CONCRETE FRAME NEHRP AREAS 3,4
-4.5 100 1
I I I tl--1---+---1 -+---+---+--1'-t------l-+1 80 --+l---!----+-,J--+---J-.j--J---J------/t----l-------+----~
I I
271
I +0.5
I f----t-/17-li71-t---t:::::1:::::::::::=FI, +1.5
I I I I
I
I I I
0~~~==~~~~~~~~==-t~J I I
0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
0::: o l-t)
tt w (!) « :2 « o z « w :2
PC2 - PRECAST CONCRETE FRAME NEHRP AREAS 3,4
-4.5 -1.5 100 I
I I I tl-+---H -f----+---t---I'-+---l-+I 80 -+l--+----+-,--+-H'----+--+----/t---+----+--~
I I
I .
271
I +0.5
I f-r-/17-ii7i-r---t:::::1:::::::::::R, +1.5
I ! I I
I I I I
o -.~=t:==t-::::=::1r==:l==~=1===l===l:===1==±=1 \ Lj i -i~
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
80 0:: 0 1-0 <( 60 u... w (!) <( ~ <( 40 0 z <( w ~
20
0
RM - REINFORCED MASONRY NEHRP AREAS 3,4
-25 -1 5
I
I I v Ill u I
I I ! i
1/ v - ---l
I I I I i J
v - -t-h 1/ I ,
i-
i 1/ 7 I L..--1,...
.... L..--' v __....-~ I I ..........- L..--1,...
........ --1// k v L---~ ~ ~ ~I
L---: ~ -- i
~ ~ ~
I I
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
272
-0.5
+0.5
+1.5
100
80 0:: 0 .... 0 « 60 u... w (!) « ~ « 40 0 Z « w ~
20
o
RM - REINFORCED MASONRY NEHRP AREAS 3,4
-25 -1 5
1 I I V Ffl U I
I / ! i
II V _ ---l
I / I I i J
V - -t-h 1/ I '
l-
i II 7 I L..-- I...-
.... L..--, v ----~ / / ..........- L..--I...-
........ .-
1// k V L---~ ~ ~ ~I
L---: ~ -- i
~ ~ ~
\
I 0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
272
-0.5
+0.5
+1.5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
URM - UNREINFORCED MASONRY NEHRP AREAS 3,4
-1 5 -0 5
1 oo --r I I I ----r II II l i i
I W! -1-+--11--~~~~~--~--1 1-4
I I I l i I ! l I I 11 I I!
0.0 0.2 0.4 0.6 0.8 1.0 PGAON ROCK
273
URM - UNREINFORCED MASONRY NEHRP AREAS 3,4
-1 5 -05
100 --rl--_t-i!l_ I / II II II iii 1_ ~!
1 1--1 i I I il i I I!
i 1 II I II
0.0 0.2 0.4 0.6 0.8 1.0
PGAON ROCK
273
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT FOR A DESIGN BASIS EARTHQUAKE
APPENDIX C
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT FOR A DESIGN BASIS EARTHQUAKE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
275
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT FOR A DESIGN BASIS EARTHQUAKE (PGA : . 20)
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NET B/C RANK TYPE SMF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------1 PC1 H 1- S,PA s 133350 1. 30814E+08 981. 9825 2 C3/S5 L 1- S,PA s 232180 1.289201E+08 556.2591 3 RM H 1- G,ES s 404670 1. 626516E+08 402.9364 4 C3/S5 L 1- S,PA M 1791062 5. 369052E+08 300.7692 5 PC1 H 1- O,C,H s 148050 2.M7198E+07 193. 3133 6 URM H 1- R M 689800. 1 1. 307045E+08 190.4816 7 URM L 1- S,PA M 9809525 1. 807826E+09 185.2929 8 C3/S5 H 1- S,PA M 725340 1. 287739E+08 178.5358 9 C3/S5 L 1- S,PA L 6170303 1. 073498E+09 174.9782 10 URM L 1- S,PA s 1968781 3.350555E+08 171. 1843 11 PC1 H 1- G,ES s 171080 2.908538E+07 171.0104 12 C1 L 1- G,ES L 7149980 1. 057456E+09 148.8963 13 URM H 1- R s 17330 2178718 126.7194 14 C3/S5 H 1- O,C,H L 6515080 6.654997E+08 103. 1476 15 URM H 1- S,PA M 1683886 1. 591293E+08 95.50121 16 C1 H 1- R M 1364880 1. 26571E+08 93.73418 17 RM L 1- O,C,H L 3048508 2.558243E+08 84.91789 18 C1 H 1- S,PA L 1. 31824E+07 1.098994E+09 84.36829 19 C1 L 1- R L 8528352 6. 54137E+08 77.70146 20 C3/S5 L 1- O,C,H L 1. 404658E+07 1. 072172E+09 77. 329'11) 21 URM H 1- O,C,H L 8230580 6.02283E+08 74.17626 22 URM H 1- O,C,H s 2004596 1. 437458E+08 72.7081 23 URM L 1- S,PA L 1. 682362E+07 1. 104954E+09 66.67871 24 URM H 1- G,ES M 517350. 1 2.831E+07 55.72117 25 RM H 1- I s 232815. 1 1. 221851E+07 53.48159 26 URM L 1- G,ES L 2760188 1. 307346E+08 48.36438 27 C3/S5 L 1- R M 3973621 1. 837491E+08 47.24224 28 PC1 H 1- I s 245641. 2 1. 11835E+07 46.5278 29 URM H 1- O,C,H M 2. 801784E+07 1.224096E+09 44.68988 30 URM L 1- R L 3656768 1. 593194E+08 44. 56837 31 URM L 1- O,C,H s 2.422262E+07 1. 054948E+09 44. 5522 32 C1 H 1- G,ES M 2991776 1. 262329E+08 43. 1933 33 URM L 1- O,C,H L 5.300257E+07 2.226276E+09 43.00318 34 C1 H 1- O,C,H M 703120 2.724219E+07 39.74472 35 RM H 1- I M 2356257 8.543161E+07 37.25734 36 URM L 1- O,C,H M 7. 811875E+07 2.810444E+09 36.97656 37 C3/S5 L 1- O,C,H M 1.046626E+07 3. 711114E+08 36.45789 38 PC1 L 1- O,C,H M 966159.8 3. 166373E+07 33.77277 39 URM L 1- R s 4262868 1. 326453E+08 32. 11644 40 URM L 1- R M 7.8742E+07 2.42959E+09 31. 85508 41 URM L 1- I s 4581639 1.376947E+08 31. 05358 42 URM L 1- I M 1. 768281E+07 5. 157863E+08 30. 1688 43 C3/S5 L 1- O,C,H s 2085890 4.663695E+07 23. 3583 44 URM H 1- I M 4232282 9. 447211E+07 23.32179 45 RM L 1- O,C,H s 1335543 2.833194E+07 22. 2138 46 RM L 1- O,C,H M 1509190 2. 865204E+07 19.98505 47 URM L 1- I L 1. 703497E+07 3.079488E+08 19.07745 48 RM H 1- O,C,H s 82250 1417628 18.2356 49 C1 H 1- O,C,H L 3.213579E+07 5.490672E+08 18.08585 50 C1 L 1- O,C,H L 1. 449005E+08 2.451909E+09 17.92133 51 URM H 1- I s 665887.3 1. 101356E+07 17.53968 52 RM H 1- I L 535089 8732269 17.31928 53 PC1 H 1- O,C,H L 665175. 1 1. 035636E+07 16.56938 54 C1 L 1- O,C,H M 1. 128398E+07 1. 716497E+08 16.2118 55 PC1 H 1- I M 7027794 1. 059153E+08 16.07091 56 PC1 L 1- I M 7040823 9.812078E+07 14.93598
275
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT FOR A DESIGN BASIS EARTHQUAKE (PGA : .20)
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NET B/C RANK TYPE SHF S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------1 PCl H 1- S,PA S 133350 1. 30814E+08 981. 9825 2 C3/S5 L 1- S,PA S 232180 1. 289201E+08 556.2591 3 RH H 1- G,ES S 404670 1. 626516E+08 402.9364 4 C3/S5 L 1- S,PA H 1791062 5. 369052E+08 300.7692 5 PCl H 1- O,C,H S 148050 2.M7198E+07 193.3133 6 URH H 1- R H 689800. 1 1. 307045E+08 190.4816 7 URH L 1- S,PA H 9809525 1. 807826E+09 185.2929 8 C3/S5 H 1- S,PA H 725340 1. 287739E+08 178.5358 9 C3/S5 L 1- S,PA L 6170303 1. 073498E+09 174.9782 10 URH L 1- S,PA S 1968781 3. 350555E+08 171. 1M3 11 PCl H 1- G,ES S 171080 2. 908538E+07 171.0104 12 Cl L 1- G,ES L 7149980 1. 057456E+09 148.8963 13 URH H 1- R S 17330 2178718 126.7194 14 C3/S5 H 1- O,C,H L 6515080 6. 654997E+08 103. 1476 15 URH H 1- S,PA H 1683886 1. 591293E+08 95.50121 16 Cl H 1- R H 1364880 1. 26571E+08 93.73418 17 RH L 1- O,C,H L 3048508 2. 558243E+08 M.91789 18 Cl H 1- S,PA L 1. 31824E+07 1.098994E+09 M.36829 19 C1 L 1- R L 8528352 6.54137E+08 77.70146 20 C3/S5 L 1- O,C,H L 1. 404658E+07 1. 072172E+09 77.329"1) 21 URH H 1- O,C,H L 8230580 6.02283E+08 74.17626 22 URH H 1- O,C,H S 2004596 1. 437458E+08 72.7081 23 URH L 1- S,PA L 1. 682362E+07 1. 104954E+09 66.67871 24 URH H 1- G,ES H 517350. 1 2. 831E+07 55.72117 25 RH H 1- I S 232815.1 1. 221851E+07 53.48159 26 URH L 1- G,ES L 2760188 1. 307346E+08 48.36438 27 C3/S5 L 1- R H 3973621 1. 837491E+08 47.24224 28 PCl H 1- I S 245641. 2 1. 11835E+07 46.5278 29 URH H 1- O,C,H H 2.801784E+07 1. 224096E+09 44.68988 30 URH L 1- R L 3656768 1. 593194E+08 44. 56837 31 URH L 1- O,C,H S 2. 422262E+07 1. 054948E+09 44. 5522 32 Cl H 1- G,ES H 2991776 1. 262329E+08 43. 1933 33 URH L 1- O,C,H L 5.300257E+07 2. 226276E+09 43.00318 34 Cl H 1- O,C,H H 703120 2. 724219E+07 39.74472 35 RH H 1- I H 2356257 8. 543161E+07 37.25734 36 URH L 1- O,C,H H 7. 811875E+07 2. 810444E+09 36.97656 37 C3/S5 L 1- O,C,H H 1.046626E+07 3. 711114E+08 36.45789 38 PCl L 1- O,C,H H 966159.8 3. 166373E+07 33.77277 39 URH L 1- R S 4262868 1. 326453E+08 32. 11644 40 URH L 1- R H 7. 8742E+07 2. 42959E+09 31. 85508 41 URH L 1- I S 4581639 1. 376947E+08 31. 05358 42 URH L 1- I H 1. 768281E+07 5. 157863E+08 30. 1688 43 C3/S5 L 1- O,C,H S 2085890 4. 663695E+07 23. 3583 44 URH H 1- I H 4232282 9.447211E+07 23.32179 45 RH L 1- O,C,H S 1335543 2. 833194E+07 22.2138 46 RH L 1- O,C,H H 1509190 2. 865204E+07 19.98505 47 URH L 1- I L 1. 703497E+07 3.079488E+08 19.07745 48 RH H 1- O,C,H S 82250 1417628 18.2356 49 Cl H 1- O,C,H L 3. 213579E+07 5. 490672E+08 18.08585 50 Cl L 1- O,C,H L 1. 449005E+08 2. 451909E+09 17.92133 51 URH H 1- I S 665887.3 1. 101356E+07 17.53968 52 RH H 1- I L 535089 8732269 17.31928 53 PCl H 1- O,C,H L 665175. 1 1. 035636E+07 16.56938 54 Cl L 1- O,C,H H 1. 128398E+07 1. 716497E+08 16.2118 55 PCl H 1- I H 7027794 1. 059153E+08 16.07091 56 PCl L 1- I H 7040823 9. 812078E+07 14.93598
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
276
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT RET B/C RANK TYPE SHF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------57 C3/S5 L 1- I s 916618.9 1.255845E+07 14. 70084 58 C2 H 1- G,ES L 2143008 2. 804779E+07 14.08805 59 PC1 H 1- I L 6305523 8.044111E+07 13.75724 60 PC2 H 1- I M 445950 4888704 11.96245 61 c~ L 1- O,C,H L 4. 817625E+07 5.055421E+08 11.4936 62 PC1 L 1- I L 1. 191433E+07 1. 2236E+08 11. 26998 63 URM H 1- I L 3220450 3. 127442E+07 10.71119 64 C3/S5 L 1- I M 3193137 2.8928E+07 10. 05943 65 C1 H 1- I s 211791. 5 1806800 9. 531032 66 PC1 L 1- G,ES s 96347.66 698739. 1 8. 252269 67 C2 H 1- I M 3549821 2.573751E+07 8.250367 68 RM L 1- I M 945936.9 6291938 7.651541 69 URM L 1- G,ES s 655163. 6 4191031 7. 396923 70 C3/S5 H 1- O,C,H M 413600 2014498 5. 870642 71 RH L 1- R s 295590. 1 1256503 5. 250831 72 C1 L 1- S,PA M 1913755 7832591 5.092787 73 PC1 H 1- O,C,H M 2962663 1.042215E+07 4. 51783 74 PC1 L 1- I s 271845.6 907450.6 4. 338111 75 RH L 1- I s 790650 2104964 3.66232 76 URM L 1+ S,PA s 1345049 2565293 2. 907211 77 S1,2,4 L 1- O,C,H L 1. 959992E+08 3.671397E+08 2. 873169 78 C1 L 1- O,C,H s 1166888 2183128 2.870898 79 URM H 1+ O,C,H s 968920. 5 1693279 2.747594 80 URM H 1+ I s 58150 96603.09 2.661274 81 URM H 1+ I M 692700 560151. 1 1. 808649 82 URM H 1+ O,C,H M 5380440 4328500 t. 804488 83 Ct H 1- I L 1474445 1183235 1. 802495 84 C1 L 1- I M 2152566 535646 1.248841 85 C1 L 1- G,ES M 3047144 180924. 3 1. 059375
--------------------------------------------------------------------------RANK NO. 85 = POINT OF MAXIMUM EFFICIENCY (OPTIMAL RETROFIT)
--------------------------------------------------------------------------86 RM H 1+ S,PA M 348600 -29761. 75 . 9146249 87 PC2 L 1- O,C,H L 4.745484E+07 -4116480 .9132548 88 C2 H 1- R M 877220.7 -254000. 3 . 710449 89 PC2 L 1- I L 1.28628E+07 -4343112 . 662351 90 C2 H 1- I L 1850720 -770793.5 . 583517 91 PC1 L 1+ S,PA L 2865108 -1219087 . 5745057 92 RM L 1- I L 756338.6 -336659.7 . 5548825 93 PC1 H 1+ I L 1. 948893E+07 -8719790 . 5525773 94 RM L 1- R L 994562.4 -454456. 1 . 5430593 95 RM H 1+ O,C,H L 815260.6 -389862. 3 . 5217943 96 RM L 1- S,PA M 662462 -350496.7 . 4709181 97 C3/S5 H 1+ G,ES s 102181. 5 -56041 . 4515544 98 URM L 1+ R M 2.648288E+07 -1.485004E+07 . 4392588 99 PC1 L 1+ I L 1. 713839E+07 -9838918 . 4259136 100 RM L 1- S,PA s 165849.3 -98345.31 . 40702 101 RM L 1- G,ES s 65800 -39379.31 . 4015302 102 RM H 1+ I L 1114111 -668141. 6 . 4002918 103 URM L 1+ I s 3025033 -1874261 . 3804163 104 PCt L 1+ S,PA M 175694.4 -109340.7 . 3776656 105 URM L 1+ S,PA L 2646517 -1663822 • 3713163 106 PC1 H 1+ O,C,H M 3726468 -2353558 . 3684214 107 URM L 1+ I M 1. 606371E+07 -1. 018992E+07 . 3656557 108 PC1 H 1+ I s 771674.5 -491066.6 . 3636352 109 PC1 L 1+ I M 2.834568E+07 -1.80585E+07 . 362919 110 URM L 1+ O,C,H M 2.425889E+07 -1. 550113E+07 . 3610123 111 RM H 1+ R M 148300 -94769.49 . 360961
276
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NET B/C RANK TYPE SHF S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------57 C3/S5 L 1- I S 916618.9 1. 255845E+07 14. 70084 58 C2 H 1- G,ES L 2143008 2. 804779E+07 14.08805 59 PCl H 1- I L 6305523 8.044111E+07 13.75724 60 PC2 H 1- I H 445950 4888704 11.96245 61 C~ L 1- O,C,H L 4.817625E+07 5.055421E+08 11. 4936 62 PCl L 1- I L 1. 191433E+07 1. 2236E+08 11. 26996 63 URH H 1- I L 3220450 3. 127442E+07 10.71119 64 C3/S5 L 1- I H 3193137 2. 8928E+07 10. 05943 65 Cl H 1- I S 211791. 5 1806800 9. 531032 66 PCl L 1- G,ES S 96347.66 698739. 1 8. 252269 67 C2 H 1- I H 3549821 2. 573751E+07 8.250367 68 RH L 1- I H 945936.9 6291938 7.651541 69 URH L 1- G,ES S 655163. 6 4191031 7. 396923 70 C3/S5 H 1- O,C,H H 413600 2014498 5. 870642 71 RH L 1- R S 295590. 1 1256503 5. 250831 72 Cl L 1- S,PA H 1913755 7832591 5.092787 73 PCl H 1- O,C,H H 2962663 1.042215E+07 4.51783 74 PCl L 1- I S 271845.6 907450.6 4. 338111 75 RH L 1- I S 790650 2104964 3.66232 76 URH L 1+ S,PA S 1345049 2565293 2.907211 77 Sl,2,4 L 1- O,C,H L 1. 959992E+08 3. 671397E+08 2. 873169 78 Cl L 1- O,C,H S 1166888 2183128 2.870898 79 URH H 1+ O,C,H S 968920. 5 1693279 2.747594 80 URH H 1+ I S 58150 96603.09 2.661274 81 URH H 1+ I H 692700 560151. 1 1. 808649 82 URH H 1+ O,C,H H 5380440 4328500 1. 804488 83 Cl H 1- I L 1474445 1183235 1. 802495 84 Cl L 1- I H 2152566 535646 1.248641 85 Cl L 1- G,ES H 3047144 180924. 3 1. 059375
--------------------------------------------------------------------------RAm:: NO. 85 = POINT OF MAXIHUH EFFICIENCY (OPTIMAL RETROFIT)
--------------------------------------------------------------------------86 RH H 1+ S,PA H 348600 -29761. 75 .9146249 87 PC2 L 1- O,C,H L 4. 745484E+07 -4116480 .9132548 88 C2 H 1- R H 877220.7 -254000. 3 .710449 89 PC2 L 1- I L 1. 28628E+07 -4343112 .662351 90 C2 H 1- I L 1850720 -770793.5 · 583517 91 PCl L 1+ S,PA L 2865108 -1219087 .5745057 92 RH L 1- I L 756338.6 -336659.7 .5548625 93 PCl H 1+ I L 1. 948893E+07 -8719790 .5525773 94 RH L 1- R L 994562.4 -454456. 1 .5430593 95 RH H 1+ O,C,H L 815260.6 -389862. 3 .5217943 96 RH L 1- S,PA H 662462 -350496.7 .4709181 97 C3/S5 H 1+ G,ES S 102181. 5 -56041 · 45155~ 98 URH L 1+ R H 2. 648288E+07 -1. 485004E+07 .4392588 99 PCl L 1+ I L 1. 713839E+07 -9838918 .4259136 100 RH L 1- S,PA S 165849.3 -98345.31 .40702 101 RH L 1- G,ES S 65800 -39379.31 .4015302 102 RH H 1+ I L 1114111 -668141. 6 .4002918 103 URH L 1+ I S 3025033 -1874261 .3804163 104 PCl L 1+ S,PA H 175694.4 -109340.7 .3776656 105 URH L 1+ S,PA L 2646517 -1663822 .3713163 106 PCl H 1+ O,C,H H 3726468 -2353558 .3684214 107 URH L 1+ I H 1. 606371E+07 -1. 018992E+07 .3656557 108 PCl H 1+ I S 771674.5 -491066.6 .3636352 109 PCl L 1+ I H 2. 834568E+07 -1.80585E+07 .362919 110 URH L 1+ O,C,H H 2. 425889E+07 -1. 550113E+07 · 3610123 111 RH H 1+ R H 148300 -94769.49 .360961
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
277
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT HET B/C RANK TYPE SMF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------112 URM L 1+ I L J!.979259 -3190767 • 3591884 113 URM L 1+ G,ES s 191739.1 -125101. 2 . 3J!.751!47 11J!. PC1 H 1+ I M 1. 2861!41E+07 -8J!.J!.6089 . 3J!.3J!.531 115 C1 H 1+ R L 9353738 -6240021 . 3328848 116 URM L 1+ S,PA M 6281155 -J!.200608 .3312364 117 URM L 1+ O,C,H s 9636134 -6476701 • 3278735 118 C1 H 1+ R M 3275712 -221J!.38J!. • 3239992 119 C1 H 1+ S,PA M 2285920 -1549340 . 3222249 120 URM L 1+ O,C,H L 1.05613E+07 -7202475 . 3180314 121 C2 H 1- S,PA M 533506 -364J!.60. 5 • 3168578 122 URM L 1+ R s 9781850 -6745117 • 3104456 123 C2 L 1- O,C,H M 1. 067941E+07 -7369101 • 3099712 124 RM L 1+ O,C,H L 3708046 -2609538 • 2962498 125 URM L 1+ R L 5517682 -3895923 • 2939204 126 RM H 1+ I M 2289660 -1632J!.22 . 287046 127 RM H 1+ G,ES s 568626.8 -407380. 3 . 2838237 128 C2 H 1- O,C,H L 1. 56092E+07 -1. 12J!.679E+07 • 2793466 129 PC2 L 1- S,PA M J!.52736.7 -326389. 3 • 27907J!.7 130 C1 L 1+ S,PA M 1334670 -965676. 1 . 2764668 131 S1,2,J!. L 1- R M 1675600 -136.q.J!.69 . 2725052 132 C1 L 1- I L 2052605 -1501935 . 2682787 133 C1 L 1- R M 976096 -715263.8 . 267199J!. 13J!. RM L 1+ S,PA L 3136398 -2304397 • 2657409 135 C3/S5 L 1+ S,PA s 375060 -276.q.J!.3. 1 • 2629364 136 C1 H 1+ I s 494078 -365269. 1 . 2607057 137 RM L 1+ I L 1357468 -101011J!. . 2558841 138 PC1 L 1+ O,C,H M 1. 047522E+07 -7819682 . 2535067 139 C2 H 1- S,PA s 288145 -215963.8 . 2505032 140 C2 L 1- R M 576351!.7 -442512.5 • 232222 141 RM H 1+ O,C,H s 1048276 -810203. 1 • 227109 142 W,S3 H 1+ S,PA s 2107525 -1635702 . 2238752 143 C2 H 1- O,C,H M 1735686 -1355646 . 2189565 11!4 PC1 L 1+ I s 6893252 -539282J!. • 2176662 145 RM L 1+ G,ES M 327846.8 -257603.2 . 2142576 146 RM L 1+ R M 2716589 -2138470 . 2128108 147 PC1 L 1+ O,C,H s 4398993 -31l,67591 . 2117308 148 C2 H 1- O,C,H s 277295.9 -219472.2 . 2085273 149 C2 L 1- S,PA M 487502. 1 -386090.8 . 2080223 150 RM L 1+ I M 8635192 -6868200 . 2046269 151 RM H 1+ I s 592598.5 -472J!.28 . 2027857 152 RM L 1+ R s 1364577 -1092718 • 1992256 153 S1,2,4 H 1+ G,ES L 5198806 -4163369 . 1991683 154 RM L 1+ S,PA M 6098496 -4940683 . 1898523 155 C3/S5 L 1+ I L 3916800 -3180227 . 1880547 156 RM H 1+ G,ES M 222450 -180662.2 . 1878526 157 C3/S5 L 1+ I M 2075329 -1689510 . 1859075 156 C2 L 1- I M 21!47552 -2001222 • 1823577 159 PC1 L 1+ G,ES M 581l,702.4 -478342. 1 • 181905 160 PC2 L 1- I M 1037577 -850776.4 . 1800354 161 PC1 H 1+ O,C,H L 3722783 -3060288 . 1779571 162 RM L 1+ G,ES s 1099413 -904501!. 5 • 1772841 163 C2 L 1- G,ES L 1. 33681E+07 -1. 106193E+07 . 172512'! 161! C1 L 1+ S,PA s 556026. 5 -1!61352. 3 . 1702692 165 C3/S5 L 1+ O,C,H M 1137566 -9J!.5019. 2 • 1692621 166 C2 H 1+ G,ES M 1974754 -1641048 • 1689859 167 C2 H 1- G,ES s 1!50684 -376045 . 1656127 168 RM L 1+ O,C,H M 1. 1!36191 E+07 -1. 199017E+07 . 1651407 169 C2 L 1- O,C,H s 2163954 -1807376 . 1647807
277
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NET B/C RANK TYPE SHF S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------112 URH L 1+ I L ~979259 -3190767 .3591884 113 URH L 1+ G,ES S 191739.1 -125101. 2 .31J.75447 11~ PCl H 1+ I H 1. 286441E+07 -8446089 .31J.34531 115 Cl H 1+ R L 9353738 -621J.0021 .332881J.8 116 URH L 1+ S,PA H 6281155 -4200608 • 3312361J. 117 URH L 1+ O,C,H S 9636131J. -6476701 .3278735 118 Cl H 1+ R H 3275712 -2214384 • 3239992 119 Cl H 1+ S,PA H 2285920 -1549340 .322221J.9 120 URH L 1+ O,C,H L 1.05613E+07 -7202475 · 3180311J. 121 C2 H 1- S,PA H 533506 -361J.1J.60. 5 • 3168578 122 URH L 1+ R S 9781850 -6745117 .3101J.1J.56 123 C2 L 1- O,C,H H 1. 06791J.1E+07 -7369101 .3099712 124 RH L 1+ O,C,H L 370801J.6 -2609538 .29621J.98 125 URH L 1+ R L 5517682 -3895923 .2939201J. 126 RH H 1+ I H 2289660 -1632422 .287046 127 RH H 1+ G,ES S 568626.8 -407380. 3 .2838237 128 C2 H 1- O,C,H L 1. 56092E+07 -1. 124679E+07 • 27931J.86 129 PC2 L 1- S,PA H ~52736.7 -326389. 3 .279071J.7 130 Cl L 1+ S,PA H 1334670 -965678. 1 · 2761J.668 131 Sl,2,4 L 1- R H 1675600 -13641J.89 · 2725052 132 Cl L 1- I L 2052605 -1501935 .2662787 133 Cl L 1- R H 976096 -715283.6 .2671994 134 RH L 1+ S,PA L 3136398 -2304397 .2657409 135 C3/S5 L 1+ S,PA S 375060 -27641J.3. 1 .2629364 136 Cl H 1+ I S 494078 -365269. 1 .2607057 137 RH L 1+ I L 1357468 -1010114 · 2558841 138 PCl L 1+ O,C,H H 1. 047522E+07 -7819682 .2535067 139 C2 H 1- S,PA S 288145 -215963.8 .2505032 140 C2 L 1- R H 576351J..7 -442512.5 • 232222 141 RH H 1+ O,C,H S 1048276 -810203. 1 • 227109 142 W,S3 H 1+ S,PA S 2107525 -1635702 .2238752 143 C2 H 1- O,C,H H 1735686 -1355646 .2189565 144 PCl L 1+ I S 6893252 -5392824 .2176662 145 RH L 1+ G,ES H 327846.8 -257603.2 .211J.2576 IIJ.6 RH L 1+ R H 2716589 -21361J.70 .2128108 IIJ.7 PCl L 1+ O,C,H S 1J.398993 -31J.67591 .2117308 148 C2 H 1- O,C,H S 277295.9 -2191J.72.2 .2085273 149 C2 L 1- S,PA H 467502. 1 -386090.8 .2080223 150 RH L 1+ I H 8635192 -6868200 .201J.6269 151 RH H 1+ I S 592598.5 -1J.72428 .2027857 152 RH L 1+ R S 136~577 -1092718 • 1992256 153 Sl,2,IJ. H 1+ G,ES L 5198806 -1J.163369 .1991683 151J. RH L 1+ S,PA H 60981J.96 -1J.91J.0683 · 1898523 155 C3/S5 L 1+ I L 3916800 -3180227 · 188051J.7 156 RH H 1+ G,ES H 222~50 -180662. 2 · 1878526 157 C3/S5 L 1+ I H 2075329 -1689510 · 1859075 158 C2 L 1- I H 2447552 -2001222 • 1823577 159 PCl L 1+ G,ES H 581J.702.1J. -1J.7831J.2. 1 .181905 160 PC2 L 1- I H 1037577 -850776.4 · 1800354 161 PCl H 1+ O,C,H L 3722783 -3060288 · 1779571 162 RH L 1+ G,ES S 10991J.13 -901J.501J.. 5 • 1772841 163 C2 L 1- G,ES L 1. 33681E+07 -1. 106193E+07 .1725127 161J. Cl L 1+ S,PA S 556026. 5 -1J.61352.3 · 1702692 165 C3/S5 L 1+ O,C,H H 1137566 -91J.5019. 2 • 1692621 166 C2 H 1+ G,ES H 197~751J. -161J.I01J.8 • 1689859 167 C2 H 1- G,ES S 1J.50684 -37601J.5 .1656127 168 RH L 1+ O,C,H H 1. 1J.36191E+07 -1. 199017E+07 · 16511J.07 169 C2 L 1- O,C,H S 2163951J. -1807376 · 161J.7807
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
278
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT RET B/C RAHX TYPE SHF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------170 C2 L 1- s 241881.8 -202225.5 . 163949 171 PC1 H 1+ O,C,H s 159828.2 -133817.3 . 1627428 172 C1 L 1+ I s 353823.6 -296295.4 . 1625902 173 C1 L 1+ S,PA L 2.77972E+07 -2. 33449E+07 . 1601708 174 C2 H 1+ G,ES L 5116480 -4302497 . 1590904 175 RH L 1+ O,C,H s 2. 611807E+07 -2. 197621E+07 . 1585822 176 C1 H 1+ I H 1078560 -909767. 1 . 1564984 177 S1,2,4 L 1- O,C,H s 333840 -281653.2 . 1563228 178 C1 L 1+ R L 2. 103142E+07 -1. 778067E+07 . 1545664 179 RH H 1+ O,C,H H 1852890 -1568529 . 153469 180 RH L 1+ I s 8932788 -7564231 . 1532061 181 C2 H 1+ I L 2.700477E+07 -2.297141E+07 . 1493573 182 C1 L 1+ O,C,H s 2957125 -2531031 . 1440906 183 C1 H 1+ O,C,H L 1. 192852E+07 -1. 022711E+07 . 1426338 184 C1 L 1+ O,C,H M 1. 306418E+07 -1. 123243E+07 . 1402118 185 C2 H 1+ S,PA H 1486820 -1278662 . 140002 186 C2 H 1+ R H 1386287 -1193822 . 1388351 187 C3/S5 L 1+ O,C,H s 976821. 5 -841894. 3 . 1381288 188 S1,2,4 H 1+ S,PA H 1093250 -943597.5 . 1368878 189 C2 H 1+ I H 1. 787947E+07 -1. 550706E+07 . 1326889 190 C1 L 1+ R s 245992. 5 -213450. 8 . 1322876 191 C1 H 1+ O,C,H H 5629096 -4901586 . 129241 192 RH L 1+ S,PA s 1822780 -1589642 . 1279026 193 C2 L 1+ G,ES s 2238572 -1952776 . 1276687 194 C1 L 1+ R H 2746304 -2400145 . 1260454 195 W,S3 H 1+ S,PA H 4007506 -3512770 . 1234524 196 C1 L 1+ I H 3808845 -3338784 . 1234131 197 C3/S5 L 1+ I s 678866.9 -595737.6 . 122453 198 PC2 L 1- S,PA L 7433443 -6523476 . 1224154 199 C1 H 1+ O,C,H s 122996.3 -107999.6 . 121928 200 C2 H 1+ 0, C,H L 2. 568632E+07 -2.262906E+07 . 1190231 201 C2 H 1+ I s 4354812 -3839912 . 1182371 202 PC1 L 1+ S,PA s 247650 -218392 . 1181425 203 C2 H 1+ O,C,H H 1. 453975E+07 -1. 284915E+07 . 1162746 204 C1 L 1+ 0, C,H L 6.616771E+07 -5.873827E+07 . 1122821 205 C2 L 1+ G,ES L 1. 719561E+07 -1. 530423E+07 . 1099923 206 C2 H 1+ G,ES s 1112813 -990558 . 1098612 207 S1,2,4 L 1- G,ES L 4.368387E+07 -3.889677E+07 . 1095852 208 C2 L 1+ S,PA L 5.609033E+07 -5.004423E+07 . 1077923 209 PC2 L 1- O,C,H s 1149912 -1026848 . 1070208 210 PC2 L 1- R H 1225345 -1096359 . 105265 211 RH L 1+ R L 6041498 -5407977 . 1048616 212 C2 H 1+ O,C,H s 3741178 -3352773 . 1038189 213 PC1 L 1+ O,C,H L 2160909 -1937929 . 1031882 214 C1 L 1+ G,ES L 7.014597E+07 -6.298104E+07 . 1021432 215 S1,2,4 H 1+ S,PA L 2608251 -2341911 . 1021144 216 W,S3 H 1+ I s 1.034847E+07 -9293048 . 1019882 217 C2 L 1+ I L 6.602995E+07 -5.936998E+07 . 1008629 218 PC2 L 1- O,C,H H 3234423 -2909910 . 1003309 219 C2 L 1+ R L 9.428126E+07 -8.495198E+07 9.895168E-02 220 PC2 L 1- G,ES H 1236368 -1114267 9.875776E-02 221 C2 L 1+ O,C,H L 1.206058E+08 -1. 08745E+08 9.834338E-02 222 W,S3 H 1+ I L 2. 19518E+07 -1. 981781E+07 9.721232E-02 223 C2 L 1+ S,PA H 5. 161151E+07 -4.666639E+07 9. 581443E-02 224 C2 L 1+ I H 1. 213618E+08 -1. 097387E+08 9.577276E-02 225 S1,2,4 H 1- I s 156621 -141745. 1 9.49801!8E-02 226 W,S3 H 1+ I H 1. 728747E+07 -1. 564554E+07 9.497826E-02 227 W,S3 L 1+ S,PA H 1. 801923E+07 -1. 631851E+07 9.438358E-02
278
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NET B/C RAHX TYPE SHF S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------170 C2 L I- I S 241881.8 -202225.5 · 163949 171 PCl H 1+ O,C,H S 159828.2 -133817.3 · 1627428 172 Cl L 1+ I S 353823.6 -296295.4 · 1625902 173 Cl L 1+ S,PA L 2. 77972E+07 -2. 33449E+07 .1601708 174 C2 H 1+ G,ES L 5116480 -4302497 · 1590904 175 RH L 1+ O,C,H S 2.611807E+07 -2. 197621E+07 .1585822 176 Cl H 1+ I H 1078560 -909767. 1 · 1564984 177 Sl,2,4 L 1- O,C,H S 333840 -281653.2 · 1563228 178 Cl L 1+ R L 2. 103142E+07 -1. 778067E+07 .1545664 179 RH H 1+ O,C,H H 1852890 -1568529 · 153469 180 RH L 1+ I S 8932788 -7564231 · 1532061 181 C2 H 1+ I L 2.700477E+07 -2. 297141E+07 · 1493573 182 Cl L 1+ O,C,H S 2957125 -2531031 .1440906 183 Cl H 1+ O,C,H L 1. 192852E+07 -1. 022711E+07 · 1426338 184 Cl L 1+ O,C,H H 1. 306418E+07 -1. 123243E+07 · 1402118 185 C2 H 1+ S,PA H 1486820 -1278662 · 140002 186 C2 H 1+ R H 1386287 -1193822 .1388351 187 C3/S5 L 1+ O,C,H S 976821. 5 -841894. 3 · 1381288 166 Sl,2,4 H 1+ S,FA H 1093250 -943597.5 .1368878 189 C2 H 1+ I H 1. 787947E+07 -1. 550706E+07 · 1326889 190 Cl L 1+ R S 245992.5 -213450. 8 · 1322876 191 Cl H 1+ O,C,H H 5629096 -4901586 · 129241 192 RH L 1+ S,PA S 1822780 -1589642 · 1279026 193 C2 L 1+ G,E5 S 2238572 -1952776 · 1276687 194 Cl L 1+ R H 2746304 -2400145 · 1260454 195 W,53 H 1+ S,FA H 4007506 -3512770 · 1234524 196 Cl L 1+ I H 3808845 -3338784 .1234131 197 C3/55 L 1+ I S 678866.9 -595737.6 · 122453 198 PC2 L 1- S,FA L 7433443 -6523476 .1224154 199 Cl H 1+ O,C,H S 122996.3 -107999.6 .121928 200 C2 H 1+ 0, C,H L 2. 568632E+07 -2. 262906E+07 · 1190231 201 C2 H 1+ I 5 43511-812 -3839912 · 1182371 202 PCl L 1+ 5,PA S 211-7650 -218392 · 118111-25 203 C2 H 1+ O,C,H H 1. 453975E+07 -1. 284915E+07 · 1162746 204 Cl L 1+ 0, C,H L 6. 616771E+07 -5. 873827E+07 · 1122821 205 C2 L 1+ G,ES L 1. 719561E+07 -1. 530423E+07 · 1099923 206 C2 H 1+ G,ES S 1112813 -990558 · 1098612 207 51,2,4 L 1- G,ES L 11-. 368387E+07 -3. 889677E+07 · 1095852 208 C2 L 1+ S,PA L 5.609033E+07 -5.004423E+07 · 1077923 209 FC2 L 1- O,C,H 5 1111-9912 -1026848 · 1070208 210 PC2 L 1- R H 1225345 -1096359 · 105265 211 RH L 1+ R L 6041498 -5407977 · 1048616 212 C2 H 1+ O,C,H S 3711-1178 -3352773 · 1038189 213 PCl L 1+ O,C,H L 2160909 -1937929 · 1031882 214 Cl L 1+ G,ES L 7.014597E+07 -6. 298101l-E+07 .1021432 215 51,2,11- H 1+ 5,PA L 2608251 -2341911 · 1021144 216 W,53 H 1+ I S 1.034B1!-7E+07 -9293048 · 1019882 217 C2 L 1+ I L 6.602995E+07 -5. 936998E+07 · 1008629 218 PC2 L 1- O,C,H H 3234423 -2909910 · 1003309 219 C2 L 1+ R L 9.1I-28126E+07 -8. 495198E+07 9. 895168E-02 220 PC2 L 1- G,ES H 1236368 -11111-267 9. 875776E-02 221 C2 L 1+ O,C,H L 1.206058E+08 -1. 08711-5E+08 9. 8311-338E-02 222 W,S3 H 1+ I L 2. 19518E+07 -1. 981781E+07 9. 721232E-02 223 C2 L 1+ 5,PA H 5.161151E+07 -II-. 666639E+07 9. 581443E-02 224 C2 L 1+ I H 1. 213618E+08 -1. 097387E+08 9. 577276E-02 225 Sl,2,II- H I- I 5 156621 -141711-5.1 9.1I-9801l8E-02 226 W,S3 H 1+ I H 1. 72871l7E+07 -1. 564551lE+07 9. 497826E-02 227 W,53 L 1+ S,PA H 1. 801923E+07 -1. 631851E+07 9. 438358E-02
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
279
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT HET B/C RANK TYPE SMF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------228 51,2,4 L 1+ R L 5.959718E+07 -5.416019Et07 9. 122894E-02 229 PC2 L 1+ S,PA H 2048270 -1861961 9.095924E-02 230 C2 L 1+ G,ES H 8681598 -7900719 8.994647E-02 231 51,2,4 H 1+ O,C,H H 1444349 -1315088 8. 949418E-02 232 51,2,4 L 1+ I H 2404113 -2190546 8.883396E-02 233 51,2,4 H 1+ I s 152700 -139174 8.857916E-02 234 51,2,4 L 1- I s 153550 -140177.9 . 0870866 235 51,2,4 L 1- G,ES s 133350 -121845. 1 8.627595E-02 236 C2 L 1+ O,C,H H 2.079338E+08 -1. 900712E+08 8.590549E-02 237 W,S3 H 1+ O,C,H L 1. 014906E+07 -9279128 8. 571557E-02 238 C2 L 1+ S,PA s 5429938 -4964784 8.566469E-02 239 C2 L 1+ I s 3. 168373E+07 -2.898352E+07 a. 522388E-02 240 C2 L 1+ R H 7.837408E+07 -7. 172003Et07 8.490112E-02 241 51, 2, 4 L 1+ G,ES H 1371110 -1257293 8. 301096E-02 242 PC2 L 1+ O,C,H H 411300 -377311 8.263788E-02 243 C2 L 1+ O,C,H s 9.317464Et07 -8. 549868Et07 8.238257E-02 244 W,S3 H 1+ O,C,H H 8630769 -7936829 8.040308E-02 245 PC1 L 1+ G,ES s 229313 -211319.4 . 0784676 246 51,2,4 L 1+ I s 376985.8 -347472 7.828898E-02 247 W,S3 L 1+ S,PA s 1. 981228E+07 -1. 827212E+07 7.773789E-02 248 S1,2,4 H 1+ O,C,H L 2.025327Et07 -1.868247Et07 7.755788E-02 249 W,S3 H 1+ S,PA L 1. 990906Et07 -1. 838145Et07 7.672921E-02 250 W,S3 L 1+ I L 1. 743207E+07 -1. 610034E+07 7. 639533E-02 251 S1,2,4 L 1+ S,PA H 8438691 -7799844 . 0757045 252 S1,2,4 L 1+ O,C,H H 1. 693967E+07 -1. 567895E+07 . 074424 253 W,S3 L 1+ I H 2.812389E+07 -2.603567E+07 7.425078E-02 254 RH L 1+ G,ES L 1125990 -1042584 7.407372E-02 255 W,S3 H 1+ O,C,H s 8163653 -7561055 7.381472E-02 256 S1,2,4 L 1+ G,ES s 2667 -2470. 321 7.374536E-02 257 S1,2,4 H 1+ G,ES s 213360 -197996.4 7.200809E-02 258 W,S3 L 1+ G,E5 H 417424.9 -388604. 1 . 0690443 259 W,53 L 1+ I s 1.270248E+07 -1. 183214E+07 6.851731E-02 260 C2 L 1+ R s 1. 257442Et07 -1. 172047E+07 6.791193E-02 261 S1,2,4 L 1+ S,PA L 6.369128E+07 -5.953046Et07 6. 532795E-02 262 51,2,4 L 1+ O,C,H L 2.637323E+08 -2.478426E+08 6.024933E-02 263 W,S3 H 1+ G,E5 s 976913. 5 -918232.9 6.006736E-02 264 W,S3 L 1+ O,C,H 5 9.246719E+07 -8.696536Et07 5.950041E-02 265 PC2 L 1+ O,C,H s 350949.3 -330163.9 5.922643E-02 266 W,S3 H 1+ R L 3.261202E+07 -3.068341Et07 5.913823E-02 267 PC2 L 1+ O,C,H L 4.240817E+07 -3.992695Et07 5.850824E-02 268 51,2,4 L 1+ O,C,H s 1392567 -1311116 5.848975E-02 269 S1,2,4 H 1+ R H 2000640 -1886829 5.688719E-02 270 W,S3 H 1+ R 5 1.060439Et07 -1. 000539E+07 5.648597E-02 271 51,2,4 L 1+ G,E5 L 6. 011759E+07 -5.687161Et07 5. 399389E-02 272 W,S3 L 1+ R H 7. 351413Et07 -6.970098Et07 5. 186953E-02 273 PC2 L 1+ I 5 231023.6 -219700.5 . 0490126 274 W,S3 ;:. 1+ O,C,H H 3.450453E+07 -3.282542E+07 4.866354E-02 275 W,S3 L 1+ 5,PA L 4.83555E+08 -4.611396E+08 4.635543E-02 276 PC2 H 1+ O,C,H H 526464 -502466.8 4. 558188E-02 277 W,S3 L 1+ R s 1.202492Et08 -1. 148553Et08 4.485622E-02 278 W,S3 L 1+ O,C,H L 1. 231206Et07 -1. 17603Et07 4.481425E-02 279 51, 2, 4 L 1+ R H 1920648 -1839888 4.204835E-02 280 W,S3 L 1+ R L 4361000 -4180111 4. 147874E-02 281 W,S3 L 1+ G,ES s 1356810 -1300858 4. 123796E-02 282 S1,2,4 L 1+ S,PA s 44710 -42879. 1 4.095061E-02 283 W,53 H 1+ R H 5270788 -5174563 1. 825623E-02 284 W,53 H 1+ G,ES H 481422.6 -473429.9 1. 660221E-02
279
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NET BIC RANK TYPE SMF 5 TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------228 51,2,4 L 1+ R L 5.959718E+07 -5. 416019E+07 9. 122894E-02 229 PC2 L 1+ S,PA H 2048270 -1861961 9.095924E-02 230 C2 L 1+ G,ES H 8681598 -7900719 8. 994647E-02 231 51,2,4 H 1+ O,C,H H 1444349 -1315088 8. 949418E-02 232 51,2,4 L 1+ I H 2404113 -2190546 8. 883396E-02 233 51,2,4 H 1+ I 5 152700 -139174 8. 857916E-02 234 51,2,4 L 1- I 5 153550 -140177.9 .0870866 235 51,2,4 L 1- G,ES 5 133350 -121845. 1 8. 627595E-02 236 C2 L 1+ O,C,H H 2.079338E+08 -1. 900712E+08 8. 590549E-02 237 W,S3 H 1+ O,C,H L 1. 014906E+07 -9279128 8.571557E-02 238 C2 L 1+ S,PA S 5429938 -496l1-784 8. 566469E-02 239 C2 L 1+ I S 3. 168373E+07 -2. 898352E+07 8. 522388E-02 240 C2 L 1+ R H 7. 837l1-08E+07 -7. 172003E+07 8.lI-90112E-02 241 Sl,2,lI- L 1+ G,ES H 1371110 -1257293 8.301096E-02 242 PC2 L 1+ O,C,H H 411300 -377311 8. 263788E-02 243 C2 L 1+ O,C,H 5 9. 31746l1-E+07 -8. 549868E+07 8. 238257E-02 244 W,S3 H 1+ O,C,H H 8630769 -7936829 8.040308E-02 245 PCl L 1+ G,ES S 229313 -211319.4 .0784676 246 51,2,4 L 1+ I S 376985.8 -347472 7. 828898E-02 2l1-7 W,S3 L 1+ S,PA S 1. 981228E+07 -1. 827212E+07 7. 773789E-02 248 51,2,4 H 1+ O,C,H L 2.025327E+07 -1. 868247E+07 7. 755788E-02 249 W,S3 H 1+ S,PA L 1. 990906E+07 -1. 838145E+07 7. 672921E-02 250 W,53 L 1+ I L 1. 743207E+07 -1. 610034E+07 7. 639533E-02 251 51,2,4 L 1+ S,PA H 8438691 -7799844 .0757045 252 51,2, 4 L 1+ O,C,H H 1. 693967E+07 -1. 567895E+07 .074424 253 W,S3 L 1+ I H 2. 812389E+07 -2.603567E+07 7. 425078E-02 254 RH L 1+ G,ES L 1125990 -1042584 7.407372E-02 255 W,S3 H 1+ O,C,H 5 8163653 -7561055 7. 381472E-02 256 51,2,4 L 1+ G,ES 5 2667 -2470. 321 7. 374536E-02 257 51,2,4 H 1+ G,ES S 213360 -197996.4 7.200809E-02 258 W, S3 L 1+ G,ES H 417424.9 -388604. 1 .0690443 259 W, 53 L 1+ I 5 1. 270248E+07 -1. 183214E+07 6. 851731E-02 260 C2 L 1+ R S 1. 257442E+07 -1. 172047E+07 6. 791193E-02 261 Sl,2,4 L 1+ S,PA L 6. 369128E+07 -5. 953046E+07 6. 532795E-02 262 51,2,4 L 1+ O,C,H L 2. 637323E+08 -2. 478426E+08 6.024933E-02 263 W,S3 H 1+ G,ES S 976913.5 -918232.9 6.006736E-02 26l1- W,S3 L 1+ O,C,H S 9. 246719E+07 -8. 696536E+07 5. 950041E-02 265 PC2 L 1+ O,C,H S 350949.3 -330163.9 5. 922643E-02 266 W, S3 H 1+ R L 3. 261202E+07 -3.068341E+07 5. 913823E-02 267 PC2 L 1+ O,C,H L 4. 240817E+07 -3. 992695E+07 5. 850824E-02 268 Sl,2,4 L 1+ O,C,H S 1392567 -1311116 5. 848975E-02 269 Sl , 2,4 H 1+ R H 2000640 -1886829 5. 688719E-02 270 W,S3 H 1+ R S 1.060439E+07 -1. 000539E+07 5. 648597E-02 271 Sl,2,4 L 1+ G,ES L 6.011759E+07 -5. 687161E+07 5. 399389E-02 272 W,S3 L 1+ R H 7.351413E+07 -6. 970098E+07 5. 186953E-02 273 PC2 L 1+ I S 231023.6 -219700.5 .0490126 274 W,S3 ;:. 1+ O,C,H H 3. 450453E+07 -3. 282542E+07 4. 866354E-02 275 W,S3 L 1+ S,PA L 4. 83555E+08 -4. 611396E+08 4. 635543E-02 276 PC2 H 1+ O,C,H H 526464 -502466.8 4. 558188E-02 277 W,S3 L 1+ R S 1.202492E+08 -1. 148553E+08 4. 485622E-02 278 W, S3 L 1+ O,C,H L 1. 231206E+07 -1. 17603E+07 4. 481425E-02 279 51,2,4 L 1+ R H 1920648 -1839888 4.204835E-02 280 W,S3 L 1+ R L 4361000 -4180111 4. 147874E-02 281 W,S3 L 1+ G,ES S 1356810 -1300858 4. 123796E-02 282 Sl,2,4 L 1+ S,PA S 44710 -42879. 1 4.095061E-02 283 W,S3 H 1+ R H 5270788 -5174563 1. 825623E-02 284 W,S3 H 1+ G,ES H 481422.6 -473429.9 1. 660221E-02
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX D
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT BASED ON AN EXPECTED VALUE ANALYSIS
APPENDIX D
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT BASED ON AN EXPECTED VALUE ANALYSIS
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
281
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT BASED ON AN EXPECTED VALUE ANALYSIS
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NPV OF PV/C RANK TYPE SMF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------1 PC1 H 1- S,PA s 133350 2.412518E+07 181. 9162 2 RM H 1- G,ES s 404670 3. 111164E+07 77.8815 3 PC1 H 1- O,C,H s 148050 5156207 35.82747 lj. URM H 1- R M 689800. 1 2. 365563E+07 35.29345 5 PC1 H 1- G,ES s 171080 5296898 31. 96153 6 C1 L 1- G,ES L 711!-9980 1.9081!-57E+08 27.69178 7 URM L 1- S,PA s 1968781 3.81!-I!-657E+07 20.52811 8 RM L 1- R L 991!-562.1!- 1. 901965E+07 20. 12361!-9 C3/S5 L 1- S,PA s 232180 1!-369893 19.82111!-10 URM L 1- S,PA M 9809525 1. 651!-11!-8E+08 17.86267 11 C3/S5 H 1- O,C,H L 6515080 1. 002391E+08 16. 3857 12 RM L 1- O,C,H L 3048508 I!-.647191!-E+07 16.21!-1!-16 13 C3/S5 L 1- O,C,H L 1.1!-0I!-658E+07 1. 87053E+08 11!-.31662 11!- URM H 1- O,C,H L 8230580 1.01!-91!-l!-5E+08 13. 75056 15 URM L 1- S,PA L 1.682362E+07 1. 989876E+08 12.82787 16 RM H 1- I s 232815. 1 2452166 11. 53268 17 C3/S5 L 1- S,PA M 1791062 1. 742152E+07 10.72692 18 PC1 H 1- I s 24561!-1. 2 18761!-08 e. 638811!-19 C3/S5 H 1- S,PA M 72531!-0 5351829 8. 378373 20 RM H 1+ S,PA M 348600 21!-98188 8. 16631!-6 21 C1 H 1- G,ES M 2991776 2. 104633E+07 8.034729 22 URM L 1- O,C,H L 5.300257E+07 3.497782E+08 7. 599269 23 C1 H 1- O,C,H M 703120 4496438 7. 39498 24 C3/S5 L 1- S,PA L 6170303 3.882024E+07 7.291465 25 RH H 1- I H 2356257 1. 474596E+07 7. 258214 26 URM L 1- R L 3656768 2.255551E+07 7. 168155 27 RH H 1- O,C,H s 82250 473583.5 6.757851!-28 PC1 L 1- O,C,H M 966159.8 5262047 6.446353 29 URM H 1- O,C,H s 2001!-596 1. 059906E+07 6. 287379 30 URM H 1- R s 17330 89374. 12 6. 157192 31 C1 L 1- R L 8528352 1!-.058413E+07 5. 758731 32 C1 H 1- S,PA L 1.31824E+07 5.068889E+07 lj., 845195 33 RH L 1- O,C,H s 133551!-3 4965930 4. 718285 34 URM H 1- S,PA M 1683886 6007278 4.567509 35 C1 H 1- R M 1364-880 4-631747 4. 39352 36 URH H 1- O,C,H M 2.801784-E+07 9. 485271E+07 1!-.38544 37 PC1 H 1- O,C,H L 665175. 1 2215990 lj., 331438 38 RM L 1- O,C,H M 1509190 45304-26 4.001893 39 URM L 1- R s 4262868 1. 267594E+07 3.973571 1!-0 URH L 1- O,C,H s 2.422262E+07 7.056948E+07 3. 913371 41 URH L 1- G,ES s 655163.6 1500557 3.290354 1!-2 RH H 1- I L 535089 1204542 3. 251105 43 PC1 H 1- I H 7027794 1. 52248E+07 3. 16637 lj.lj. RH L 1- I H 945936.9 1988497 3. 102146 45 URH L 1- O,C,H H 7. 811875E+07 1. 638732E+08 3.097745 1!-6 C1 L 1- O,C,H L 1. 449005E+08 2.943038E+08 3.031075 1!-7 PC1 L 1- I H 7040823 1. 384212E+07 2. 965981 48 C1 H 1- I s 211791. 5 376199.5 2.776273 49 URH H 1- G,ES H 517350. 1 901!-654.2 2.749017 50 RH L 1- R s 295590. 1 459673.2 2.555103 51 PC1 H 1- I L 6305523 9721785 2. 51!-1769 52 URH L 1- R H 7.871!-2E+07 1. 163907E+08 2. 503526 53 URH L 1- I s 4581639 661!-5443 2.494104 54 URH L 1- G,ES L 2760186 3659853 2.3981!-02 55 URH L 1- I H 1. 768281E+07 2.li-22209E+07 2. 36961 56 C3/S5 L 1- R H 3973621 46618111- 2.17319
281
BUILDING CLASSES PRIORITIZED FOR OPTIMAL RETROFIT BASED ON AN EXPECTED VALUE ANALYSIS
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NPV OF PV/C RAm:: TYPE SHF S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------1 PCl H 1- S,PA S 133350 2. 412518E+07 181. 9162 2 RH H 1- G,ES S 404670 3. 111164E+07 77.8815 3 PCl H 1- O,C,H S 148050 5156207 35.82747 J!. URH H 1- R H 689800. 1 2. 365563E+07 35. 293J!.5 5 PCl H 1- G,ES S 171080 5296898 31. 96153 6 Cl L 1- G,ES L 7149980 1.908457E+08 27.69178 7 URH L 1- S,PA S 1968781 3. 844657E+07 20.52811 8 RH L 1- R L 994562.4 1. 901965E+07 20. 12364 9 C3/S5 L 1- S,PA S 232180 4369893 19.82114 10 URH L 1- S,PA H 9809525 1. 654148E+08 17.86267 11 C3/S5 H 1- O,C,H L 6515080 1. 002391E+08 16. 3857 12 RH L 1- O,C,H L 3048508 4. 647194E+07 16.24416 13 C3/S5 L 1- O,C,H L 1.404658E+07 1. 87053E+08 14.31662 14 URH H 1- O,C,H L 8230580 1.049lJ.lJ.5E+08 13.75056 15 URH L 1- S,PA L 1. 682362E+07 1. 989876E+08 12.82787 16 RH H 1- I S 232815. 1 2452166 11.53268 17 C3/S5 L 1- S,PA H 1791062 1. 742152E+07 10.72692 18 PCl H 1- I S 245641. 2 1876408 8. 638814 19 C3/S5 H 1- S,PA H 725340 5351829 8. 378373 20 RH H 1+ S,PA H 348600 2498188 8. 166346 21 C1 H 1- G,ES H 2991776 2. 104633E+07 8.034729 22 URH L 1- O,C,H L 5.300257E+07 3. 497782E+08 7. 599269 23 C1 H 1- O,C,H H 703120 lJ.lJ.96438 7.39498 24 C3/S5 L 1- S,PA L 6170303 3. 882024E+07 7.291465 25 RH H 1- I H 2356257 1. 474596E+07 7. 258214 26 URH L 1- R L 3656768 2. 255551E+07 7.168155 21 RH H 1- O,C,H S 82250 473583.5 6.757854 28 PCl L 1- O,C,H H 966159.8 5262047 6.lJ.lJ.6353 29 URH H 1- O,C,H S 2004596 1. 059906E+07 6. 287379 30 URH H 1- R S 17330 89374. 12 6. 157192 31 Cl L 1- R L 8528352 4.058413E+07 5. 758731 32 Cl H 1- S,PA L 1. 31824E+07 5.068889E+07 4.845195 33 RH L 1- O,C,H S 1335543 4965930 4.718285 34 URH H 1- S,PA H 1683886 6007278 4. 567509 35 Cl H 1- R H 1364880 11631747 4. 39352 36 URH H 1- O,C,H H 2.801784E+07 9.485271E+07 4. 385lJ.lJ. 37 PCl H 1- O,C,H L 665175. 1 2215990 4. 331438 38 RH L 1- O,C,H H 1509190 4530426 4.001893 39 URH L 1- R S 4262868 1. 267594E+07 3.973571 110 URH L 1- O,C,H S 2. 422262E+07 7.056948E+07 3.913371 41 URH L 1- G,ES S 655163.6 1500557 3.290354 42 RH H 1- I L 535089 1204542 3.251105 43 PCl H 1- I H 7027794 1. 52248E+07 3. 16637 44 RH L 1- I H 945936.9 1988497 3. 102146 45 URH L 1- O,C,H H 7.811875E+07 1. 638732E+08 3.097745 46 Cl L 1- O,C,H L 1. 449005E+08 2. 943038E+08 3.031075 47 PCl L 1- I H 7040823 1. 384212E+07 2. 965981 48 Cl H 1- I S 211791. 5 376199.5 2.776273 49 URH H 1- G,ES H 517350.1 904854.2 2.749017 50 RH L 1- R S 295590. 1 459673. 2 2.555103 51 PCl H 1- I L 6305523 9721785 2.5'1-1789 52 URH L 1- R H 7. 8742E+07 1. 183907E+08 2. 503526 53 URH L 1- I S 4581639 6845 lJ.lJ. 3 2.494104 54 URH L 1- G,ES L 2760188 3859853 2.398402 55 URH L 1- I H 1. 768281E+07 2. 422209E+07 2. 36981 56 C3/S5 L 1- R H 3973621 4661814 2.17319
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
282
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT HPV OF PV/C RANK TYPE SMF' s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------57 C3/S5 L 1- O,C,H M 1.046626E+07 1. 187371E+07 2. 134475 58 PC1 L 1- I L 1. 191433E+07 1. 307485E+07 2.097406 59 URM L 1+ S,PA s 1345049 1416644 2.053229 60 PC1 L 1- G,ES s 96347.66 99975. 16 2.03765 61 PC1 L 1- I s 271845.6 216758.6 1. 797359 62 URM H 1- I M 4232282 3275492 1. 77393 63 URM H 1+ O,C,H s 968920. 5 695781. 9 1. 7181 64 PC2 L 1- S,PA M 452736.7 285414. 8 1. 630421 65 C2 H 1- I M 3549821 2093964 1.589879 66 PC2 H 1- I M 445950 252411. 6 1. 566009 67 URM H 1+ I s 58150 32787.63 1. 563846 68 URM L 1- I L 1.703497E+07 8675466 1. 509274 69 RM L 1- I s 790650 361100.9 1.456714 70 C3/S5 L 1- O,C,H s 2085890 919483 1. 440811 71 C1 L 1- S,PA M 1913755 831058. 3 1. 434255 72 URM H 1- I s 665887. 3 261977.5 1. 393426 73 C1 L 1- O,C,H M 1. 128398E+07 3861888 1. 342245 74 S1,2,4 L 1- R M 1875600 620623.8 1. 330893 75 C1 H 1+ R L 9353738 1092280 1. 116775 76 C2 H 1- S,PA M 533506 52093. 25 1. 097643 77 C3/S5 L 1- I s 916618.9 66242.06 1. 072268 78 RM L 1- S,PA M 662462 45252. 13 1. 068309
--------------------------------------------------------------------------RANK HO. 78 : POIHT OF MAXIMUM EFFICIENCY (OPTIMAL RETROFIT) --------------------------------------------------------------------------79 URM H 1+ O,C,H M 5380440 -157610.5 . 9707068
80 RM L 1- S,PA s 165849. 3 -12520. 36 . 9245076 81 PC1 H 1- O,C,H M 2962663 -230987. 5 . 9220339 82 URM H 1+ I M 692700 -79899.31 . 8846552 83 RM H 1+ O,C,H L 815260.6 -95749.81 . 8825531 84 C1 H 1- O,C,H L 3.213579E+07 -4616678 . 8563384 85 W, S3 H 1+ S,PA s 2107525 -363462.3 . 8275407 86 RM L 1+ O,C,H L 3708046 -742102. 5 . 799867 87 RM L 1- G,ES s 65800 -18034.23 . 7259236 88 C2 H 1- G,ES L 2143008 -615834. 5 . 7126308 89 C3/S5 L 1- I M 3193137 -1066151 .6661116 90 C2 L 1- O,C,H L 4.817625]i:+07 -1. 807687E+07 . 6247765 91 URM H 1- I L 3220450 -1345424 . 5822249 92 C1 L 1- R M 976096 -421886.7 . 5677816 93 PC2 L 1- I M 1037577 -528170.2 .4909581 94 C1 H 1+ R M 3275712 -1759771 . 4627821 95 C1 L 1- G,ES M 3047144 -1648661 . 4589487 96 C2 H 1- S,PA s 288145 -157178.4 . 4545162 97 C1 L 1- O,C,H s 1166888 -654291. 8 . 4392849 98 C1 H 1- I L 1474445 -840341. 2 .4300627 99 C1 H 1+ S,PA M 2285920 -1356567 . 4065553 100 C1 L 1- I M 2152566 -1354724 . 3706469 101 C2 H 1- R M 877220.7 -553336. 8 . 369216 102 C1 H 1+ I s 494078 -328208.6 . 3357149 103 C3/S5 H 1- O,C,H M 413600 -280558.3 . 3216676 104 PC2 L 1- I L 1. 28628E+07 -9242656 . 2814429 105 RM L 1- I L 756338.6 -566385. 3 . 2511486 106 C2 L 1- R M 576354.7 -438300.6 . 2395298 107 C2 H 1- O,C,H M 1735686 -1323993 . 237193 108 PC2 L 1- O,C,H L 4.745484E+07 -3.692571E+07 . 221877 109 S1,2,4 L 1- O,C,H L 1. 959992E+08 -1. 541467E+08 . 2135341 110 C3/S5 H 1+ G,ES s 102181. 5 -81603.68 . 2013851 111 C2 H 1- I L 1850720 -1484007 . 1981462
282
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NPV OF PV/C RANK TYPE SMF' S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------57 C3/S5 L 1- O,C,H H 1.046626E+07 1. 187371E+07 2. 134475 58 PCt L I- I L 1. 191433E+07 1. 307485E+07 2.097406 59 URH L 1+ S,PA S 1345049 1416644 2.053229 60 PCl L 1- G,ES S 96347.66 99975. 16 2.03765 61 PCl L I- I S 271845.6 216758.6 1. 797359 62 URH H I- I H 4232282 3275492 1. 77393 63 URH H 1+ O,C,H S 968920. 5 695781. 9 1. 7181 64 PC2 L 1- S,PA H 452736.7 285414. 8 1. 630421 65 C2 H 1- I H 3549821 2093964 1.589879 66 PC2 H I- I H 445950 252411. 6 1.566009 67 URH H 1+ I S 58150 32787.63 1.563846 68 URH L I- I L 1.703497E+07 8675466 1. 509274 69 RH L I- I S 790650 361100.9 1.456714 70 C3/S5 L 1- O,C,H S 2085890 919483 1. 440811 71 Cl L 1- S,PA H 1913755 831058. 3 1. 434255 72 URH H 1- I S 665887. 3 261977.5 1.393426 73 Cl L 1- O,C,H H 1. 128398E+07 3861888 1. 342245 74 Sl,2,4 L 1- R H 1875600 620623.8 1.330893 75 Cl H 1+ R L 9353738 1092280 1. 116775 76 C2 H 1- S,PA H 533506 52093. 25 1. 097643 77 C3/S5 L I- I S 916618.9 66242.06 1. 072268 78 RH L 1- S,PA H 662462 45252. 13 1. 068309
--------------------------------------------------------------------------RANK HO. 78 : POINT OF MAXIMUM EFFICIENCY (OPTIMAL RETROFIT)
--------------------------------------------------------------------------79 URH H 1+ O,C,H H 5380440 -157610.5 .9707068 80 RH L 1- S,PA S 165849. 3 -12520.36 .9245076 81 PCl H 1- O,C,H H 2962663 -230987. 5 .9220339 82 URH H 1+ I H 692700 -79899.31 .8846552 83 RH H 1+ O,C,H L 815260.6 -95749.81 .8825531 84 Cl H 1- O,C,H L 3. 213579E+07 -4616678 .8563384 85 W, S3 H 1+ S,PA S 2107525 -363462.3 .8275407 86 RH L 1+ O,C,H L 3708046 -742102. 5 .799867 87 RH L 1- G,ES S 65800 -18034.23 .7259236 88 C2 H 1- G,ES L 2143008 -615834.5 .7126308 89 C3/S5 L 1- I H 3193137 -1066151 .6661116 90 C2 L 1- O,C,H L 4. 817625E+07 -1. 807687E+07 .6247765 91 URH H I- I L 3220450 -1345424 .5822249 92 Cl L 1- R H 976096 -421886.7 .5677816 93 PC2 L 1- I H 1037577 -528170.2 .4909581 94 Cl H 1+ R H 3275712 -1759771 .4627821 95 Cl L 1- G,ES H 3047144 -1648661 .4589487 96 C2 H 1- S,PA S 288145 -157178.4 .4545162 97 Cl L 1- O,C,H S 1166888 -654291. 8 .4392849 98 Cl H 1- I L 1474445 -840341. 2 .4300627 99 Cl H 1+ S,PA H 2285920 -1356567 .4065553 100 Cl L 1- I H 2152566 -1354724 .3706469 101 C2 H 1- R H 877220.7 -553336. 8 .369216 102 Cl H 1+ I S 494078 -328208.6 .3357149 103 C3/S5 H 1- O,C,H H 413600 -280558.3 .3216676 104 PC2 L I- I L 1. 28628E+07 -9242656 . 2814429 105 RH L I- I L 756338.6 -566385. 3 .2511486 106 C2 L 1- R H 576354.7 -438300.6 .2395298 107 C2 H 1- O,C,H H 1735686 -1323993 .237193 108 PC2 L 1- O,C,H L 4. 745484E+07 -3. 692571E+07 .221877 109 Sl,2,4 L 1- O,C,H L 1. 959992E+08 -1. 541467E+08 .2135341 110 C3/S5 H 1+ G,ES S 102181. 5 -81603.68 .2013851 111 C2 H I- I L 1850720 -1484007 .1981462
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
283
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT HPV OF PV/C RANK TYPE SMF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------112 RM H 1+ I L 1114111 -920081. 1 .1741568 113 C1 L 1+ R L 2. 103142E+07 -1. 75178E+07 . 1670651 114 C2 L 1- O,C,H M 1. 067941E+07 -8944096 . 1624915 115 C2 H 1- O,C,H s 277295.9 -232904.9 . 1600853 116 RM H 1+ R M 148300 -126523 . 1468442 117 URM L 1+ R M 2.648288E+07 -2. 271378E+07 . 1423222 118 RM H 1+ I s 592598.5 -509018.8 . 1410394 119 C1 L 1+ S,PA M 1334670 -1147447 . 1402766 120 RM L 1+ S,PA M 6098496 -5293188 . 1320502 121 RM H 1+ I M 2289660 -2023354 . 116308 122 C1 L 1+ O,C,H L 6.616771E+07 -5.932476E+07 . 1034183 123 C1 L 1- I L 2052605 -1845914 . 100697 124 C2 H 1- O,C,H L 1. 56092E+07 -1.409464E+07 9. 702982E-02 125 RH H 1+ O,C,H s 1048276 -953639. 1 9.027858E-02 126 URM L 1+ O,C,H H 2.425889E+07 -2.245963E+07 7.416923E-02 127 RM H 1+ G,ES s 568826.8 -528801. 2 7.036521E-02 128 C1 L 1+ O,C,H M 1. 306418E+07 -1. 214815E+07 7. 011772E-02 129 URM L 1+ S,PA H 6281155 -5856220 6.765241E-02 130 URM L 1+ O,C,H s 9636134 -8991664 6. 688053E-02 131 PC2 L 1- O,C,H M 3234423 -3022644 6.547671E-02 132 URM L 1+ G,ES s 191739. 1 -179332.6 6.470521E-02 133 URM L 1+ S,PA L 2646517 -2477762 6. 376482E-02 134 PC1 H 1+ I L 1. 948893E+07 -1. 826058E+07 6. 302792E-02 135 URM L 1+ I s 3025033 -2835107 6.278468E-02 136 PC1 L 1+ S,PA L 2865108 -2689581 6. 126357E-02 137 C2 L 1- S,PA M 487502. 1 -458262. 1 5.997935E-02 138 URM L 1+ R s 9781850 -9228940 5.652404E-02 139 PC1 L 1+ I L 1. 713839E+07 -1. 618202E+07 5. 580263E-02 140 C2 L 1- I M 2447552 -2316456 5. 356226E-02 141 PC1 H 1+ I s 771674.5 -730517.8 5. 333436E-02 142 PC1 H 1+ O,C,H M 3726468 -3531132 5.241871E-02 143 PC1 L 1+ I M 2.834568E+07 -2.688545E+07 5. 151503E-02 144 RM L 1+ S,PA L 3138398 -2977256 5. 134522E-02 145 C2 L 1- O,C,H s 2163954 -2053431 5. 107446E-02 146 C1 L 1+ I s 353823.6 -336107.5 5.007033E-02 147 URM L 1+ I M 1. 606371E+07 -1. 526656E+07 4.962434E-02 148 RM L 1+ O,C,H s 2. 611807E+07 -2.488121E+07 . 0473566 149 RM L 1+ I M 8635192 -8235773 4.625474E-02 150 C3/S5 L 1+ S,PA s 375060 -358407.7 4.439913E-02 151 URM L 1+ I L 4979259 -4758497 . 0443363 152 PC1 H 1+ I M 1. 286441E+07 -1. 23252E+07 . 0419151 153 PC1 L 1+ S,PA M 175694.4 -168400. 3 4. 151579E-02 154 RM L 1+ I s 8932788 -8583598 3.909085E-02 155 C2 H 1- G,ES s 450684 -433729. 3 3.762005E-02 156 URM L 1+ O,C,H L 1. 05613E+07 -1. 018181E+07 3. 593181E-02 157 PC2 L 1- O,C,H s 1149912 -1109673 3.499306E-02 158 URM L 1+ R L 5517682 -5329728 3.406392E-02 159 RM L 1+ G,ES s 1099413 -1064926 3. 136867E-02 160 RM L 1+ R s 1364577 -1322320 3.096682E-02 161 RM L 1+ R M 2716589 -2635097 2.999799E-02 162 PC1 L 1+ O,C,H H 1. 047522E+07 -1. 016644E+07 2.947706E-02 163 RM L 1+ I L 1357468 -1318046 2.904081E-02 164 C1 L 1+ S,PA L 2.77972E+07 -2.700563E+07 . 0284767 165 C1 L 1+ O,C,H s 2957125 -2874005 2.810835E-02 166 C1 H 1+ O,C,H M 5629096 -5478707 2. 671644E-02 167 C2 L 1- G,ES L 1. 33681E+07 -1. 30 1806E+07 2.618502E-02 168 PC1 L 1+ I s 6893252 -6717085 2.555646E-02 169 PC2 L 1- S,PA L 7433443 -7243628 2. 553528E-02
283
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT HPV OF PV/C RAHK TYPE SMF S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------112 RH H 1+ I L 1114111 -920081. 1 .1741568 113 C1 L 1+ R L 2. 103142E+07 -1. 75178E+07 · 1670651 114 C2 L 1- O,C,H H 1. 067941E+07 -8944096 .1624915 115 C2 H 1- O,C,H S 277295.9 -232904.9 · 1600853 116 RH H 1+ R H 148300 -126523 · 1468442 117 URH L 1+ R H 2. 648288E+07 -2. 271378E+07 · 1423222 118 RH H 1+ I S 592598.5 -509018.8 · 1410394 119 Cl L 1+ S,PA H 1334670 -1147447 · 1402766 120 RH L 1+ S,PA H 6098496 -5293188 · 1320502 121 RH H 1+ I H 2289660 -2023354 · 116308 122 Cl L 1+ O,C,H L 6. 616771E+07 -5. 932476E+07 · 1034183 123 Cl L 1- I L 2052605 -1845914 · 100697 124 C2 H 1- O,C,H L 1. 56092E+07 -1.409464E+07 9. 702982E-02 125 RH H 1+ O,C,H S 1048276 -953639. 1 9.027858E-02 126 URH L 1+ O,C,H H 2. 425889E+07 -2. 245963E+07 7.416923E-02 127 RH H 1+ G,ES S 568826.8 -528801. 2 7.036521E-02 128 Cl L 1+ O,C,H H 1. 306418E+07 -1. 214815E+07 7. 011772E-02 129 URH L 1+ S,PA H 6281155 -5856220 6. 765241E-02 130 URH L 1+ O,C,H S 9636134 -8991664 5. 588053E-02 131 PC2 L 1- O,C,H H 3234423 -3022644 6. 547671E-02 132 URH L 1+ G,ES S 191739.1 -179332.6 6. 470521E-02 133 URH L 1+ S,PA L 2646517 -2477762 6. 376482E-02 134 PCl H 1+ I L 1. 948893E+07 -1. 826058E+07 6. 302792E-02 135 URH L 1+ I S 3025033 -2835107 6. 278468E-02 136 PCl L 1+ S,PA L 2865108 -2689581 6. 126357E-02 137 C2 L 1- S,PA H 487502. 1 -458262. 1 5. 997935E-02 138 URH L 1+ R S 9781850 -9228940 5. 652404E-02 139 PCl L 1+ I L 1. 713839E+07 -1. 618202E+07 5. 580263E-02 140 C2 L 1- I H 2447552 -2316456 5. 356226E-02 141 PCl H 1+ I S 771674.5 -730517.8 5. 333436E-02 142 PCl H 1+ O,C,H H 3726468 -3531132 5. 241871E-02 143 PCl L 1+ I H 2. 834568E+07 -2. 688545E+07 5. 151503E-02 144 RH L 1+ S,PA L 3138398 -2977256 5. 134522E-02 145 C2 L 1- O,C,H S 2163954 -2053431 5. 107446E-02 146 Cl L 1+ I S 353823.6 -336107.5 5.007033E-02 147 URH L 1+ I H 1. 606371E+07 -1. 526656E+07 4. 962434E-02 148 RH L 1+ O,C,H S 2.611807E+07 -2. 488121E+07 .0473566 149 RH L 1+ I H 8635192 -8235773 4. 625474E-02 150 C3/S5 L 1+ S,PA S 375060 -358407.7 4. 439913E-02 151 URH L 1+ I L 4979259 -4758497 .0443363 152 PCl H 1+ I H 1. 286441E+07 -1. 23252E+07 .0419151 153 PCl L 1+ S,PA H 175694.4 -168400.3 4.151579E-02 1511- RH L 1+ I S 8932788 -6583598 3.909085E-02 155 C2 H 1- G,ES S 450684 -433729. 3 3. 762005E-02 156 URH L 1+ O,C,H L 1. 05613E+07 -1. 018181E+07 3. 593181E-02 157 PC2 L 1- O,C,H S 1149912 -1109673 3. 499306E-02 158 URH L 1+ R L 5517682 -5329728 3.406392E-02 159 RH L 1+ G,ES S 1099413 -1054926 3.136867E-02 160 RH L 1+ R S 1364577 -1322320 3.096682E-02 161 RH L 1+ R H 2716589 -2635097 2. 999799E-02 162 PCl L 1+ O,C,H H 1. 047522E+07 -1. 016644E+07 2. 947706E-02 163 RH L 1+ I L 1357468 -1318046 2.904081E-02 164 C1 L 1+ S,PA L 2. 77972E+07 -2.700563E+07 .0284767 165 C1 L 1+ O,C,H S 2957125 -2874005 2. 610835E-02 166 C1 H 1+ O,C,H H 5629096 -5478707 2. 671644E-02 167 C2 L 1- G,ES L 1. 33681E+07 -1. 301806E+07 2. 618502E-02 168 PCl L 1+ I S 6893252 -6717085 2. 555646E-02 169 PC2 L 1- S,PA L 7433443 -7243628 2. 553528E-02
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
284
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT HPV OF PV/C RAHX TYPE SMF s TYPE SIZE COST, $ BEHEFIT, $ RATIO
------------------------------------------------------------------------------170 RM L 1+ G,ES M 327846. 8 -319579 2. 521874E-02 171 C2 L 1- I s 241881. 8 -235947. 1 2.453553E-02 172 C1 H 1+ I M 1078560 -1052614 2.405616E-02 173 PC1 L 1+ O,C,H s 4398993 -4293696 2. 393674E-02 174 RM H 1+ O,C,H M 1852890 -1809095 2. 363606E-02 175 C1 L 1+ S,PA s 556026. 5 -543256. 1 2. 296726E-02 176 C2 H 1+ G,ES H 1974754 -1931619 2. 184311E-02 177 51,2,4 H 1+ G,ES L 5198806 -5086394 2. 162276E-02 178 RM H 1+ G,ES H 222450 -217666.4 2. 150406E-02 179 C3/S5 L 1+ I H 2075329 -2031170 2. 127801E-02 180 C3/S5 L 1+ I L 3916800 -3836442 2.051621E-02 181 C2 H 1+ I H 1. 787947E+07 -1. 751315E+07 2.048804E-02 182 PC1 L 1+ G,ES H 584702.4 -572807. 1 2.034414E-02 183 PC1 H 1+ O,C,H L 3722783 -3648593 1. 992879E-02 184 C2 H 1+ G,ES L 5116480 -5014731 1.988659E-02 185 RM L 1+ O,C,H H 1. 436191E+07 -1. 407635E+07 1. 988306E-02 186 C3/S5 L 1+ O,C,H H 1137566 -1115329 1. 954809E-02 187 C2 H 1+ S,PA H 1486820 -1458509 1. 904145E-02 188 C2 H 1+ R H 1386287 -1360358 1. 870421E-02 189 PC1 H 1+ O,C,H s 159828.2 -156916.9 . 0182149 190 C2 H 1+ I L 2.700477E+07 -2.651528E+07 1.812626E-02 191 RM L 1+ S,PA s 1822780 -1792370 1. 668339E-02 192 S1,2,4 L 1- O,C,H s 333840 -328326.6 1. 651508E-02 193 C1 H 1+ O,C,H L 1. 192852E+07 -1. 173194E+07 1. 647972E-02 194 S1,2,4 H 1+ S,PA H 1093250 -1075450 1. 628148E-02 195 C3/S5 L 1+ O,C,H s 976821. 5 -961385.6 1. 580214E-02 196 W,S3 H 1+ S,PA H 4007506 -3945158 1. 555784E-02 197 c2 L 1+ G,ES s 2238572 -2203829 1. 552032E-02 198 c2 H 1+ I s 4354812 -4287400 1. 547989E-02 199 C1 L 1+ I H 3808845 -3749892 1. 547805E-02 200 PC2 L 1- R H 1225345 -1207020 1.495488E-02 201 C1 L 1+ R s 245992.5 -242350.3 1. 480604E-02 202 C2 H 1+ O,C,H H 1. 453975E+07 -1. 432891E+07 1. 450078E-02 203 C2 H 1+ O,C,H L 2.568632E+07 -2.532229E+07 1. 417202E-02 204 PC2 L 1- G,ES H 1236368 -1218919 1. 411353E-02 205 C1 L 1+ R H 2746304 -2708181 1. 388162E-02 206 C3/S5 L 1+ I s 678866.9 -669469 1. 384349E-02 207 c2 L 1+ S,PA L 5.609033E+07 -5.531548E+07 1. 381436E-02 208 C1 H 1+ O,C,H s 122996. 3 -121347.7 1. 340353E-02 209 C2 H 1+ O,C,H s 3741178 -3691677 1. 323146E-02 210 c2 H 1+ G,ES s 1112813 -1098271 1. 306794E-02 211 PC1 L 1+ S,PA s 247650 -244432.8 1. 299109E-02 212 C2 L 1+ G,ES L 1. 719561E+07 -1. 697262E+07 1. 296794E-02 213 51, 2, 4 L 1- G,ES L 4-.368387E+07 -4. 312204E+07 1.286134E-02 214 S1,2,4 H 1+ S,PA L 2608251 -2574769 1. 283696E-02 215 RM L 1+ R L 604-14-98 -5964656 1. 271905E-02 216 51, 2, 4 H 1- I s 156621 -154667.3 1. 247423E-02 217 C1 L 1+ G,ES L 7.014597E+07 -6.928248E+07 1. 230995E-02 218 W,S3 H 1+ I s 1. 034847E+07 -1. 022357E+07 1.206915E-02 219 c2 L 1+ O,C,H L 1.206058E+08 -1. 191556E+08 1. 202399E-02 220 c2 L 1+ I L 6.602995E+07 -6.525954E+07 1. 166759E-02 221 PC1 L 1+ O,C,H J. 2160909 -2135715 1. 165924E-02 222 W,S3 H 1+ I L 2. 19518E+07 -2. 170104E+07 1. 142329E-02 223 C2 L 1+ I H 1. 213618E+08 -1. 200064E+08 1. 116859E-02 224 51, 2, 4 L 1- I s 153550 -151846.8 1. 109237E-02 225 W, S3 H 1+ I H 1. 728747E+07 -1. 709605E+07 1. 107299E-02 226 W, S3 L 1+ S,PA H 1. 801923E+07 -1. 781979E+07 1. 106814E-02 227 C2 L 1+ S,PA H 5. 161151E+07 -5. 104175E+07 1. 103941!-E-02
284
------------------------------------------------------------------------------R/F 5TRUC OCCUP RETROFIT HPV OF PV/C RAm: TYPE SHF 5 TYPE SIZE C05T, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------170 RH L 1+ G,ES H 327846.8 -319579 2.52187IlE-02 171 C2 L 1- I S 241881. 8 -235947. 1 2. 453553E-02 172 Cl H 1+ I H 1078560 -1052611l 2.405616E-02 173 PCl L 1+ O,C,H 5 4398993 -4293696 2. 393671lE-02 171l RH H 1+ O,C,H H 1852890 -1809095 2. 363606E-02 175 Cl L 1+ S,PA S 556026. 5 -543256. 1 2. 296726E-02 176 C2 H 1+ G,ES H 1974754 -1931619 2.184311E-02 177 51,2,4 H 1+ G,ES L 5198806 -5086394 2.162276E-02 178 RH H 1+ G,ES H 222450 -217666.4 2.1501l06E-02 179 C3/S5 L 1+ I H 2075329 -2031170 2. 127801E-02 180 C3/S5 L 1+ I L 3916800 -38361!-lJ.2 2.051621E-02 181 C2 H 1+ I H 1. 787947E+07 -1. 751315E+07 2.01l8804E-02 182 PCl L 1+ G,ES H 584702.4 -572807. 1 2. o 31!-lJ.14E-02 183 PCl H 1+ O,C,H L 3722783 -3648593 1. 992879E-02 181l C2 H 1+ G,ES L 5116480 -5014731 1.988659E-02 185 RH L 1+ O,C,H H 1. 436191E+07 -1. 407635E+07 1. 988306E-02 186 C3/S5 L 1+ O,C,H H 1137566 -1115329 1. 954809E-02 187 C2 H 1+ S,PA H 1486820 -1458509 1. 904145E-02 188 C2 H 1+ R H 1386287 -1360358 1. 870421E-02 189 PCl H 1+ O,C,H 5 159828.2 -156916.9 .0182149 190 C2 H 1+ I L 2.700477E+07 -2. 651528E+07 1. 812626E-02 191 RH L 1+ 5,PA 5 1822780 -1792370 1. 668339E-02 192 Sl,2,4 L 1- O,C,H S 333840 -328326.6 1. 651508E-02 193 C1 H 1+ O,C,H L 1. 192852E+07 -1. 173194E+07 1. 647972E-02 1911 Sl,2,4 H 1+ S,PA H 1093250 -1075450 1. 628148E-02 195 C3/S5 L 1+ O,C,H S 976821. 5 -961385.6 1. 580214E-02 196 W,S3 H 1+ 5,PA H 4007506 -391l5158 1. 555781lE-02 197 C2 L 1+ G,ES S 2238572 -2203829 1. 552032E-02 198 C2 H 1+ I S 11354812 -1l287400 1. 51l7989E-02 199 Cl L 1+ I H 3808845 -371l9892 1. 51l7805E-02 200 PC2 L 1- R H 1225345 -1207020 1. 495488E-02 201 Cl L 1+ R 5 245992.5 -242350.3 1. 480604E-02 202 C2 H 1+ O,C,H H 1. 453975E+07 -1.1l32891E+07 1. 450078E-02 203 C2 H 1+ O,C,H L 2. 568632E+07 -2. 532229E+07 1. 417202E-02 201l PC2 L 1- G,ES H 1236368 -1218919 1. 411353E-02 205 Cl L 1+ R H 2746304 -2708181 1. 388162E-02 206 C3/S5 L 1+ I S 678866.9 -6691l69 1. 384349E-02 207 C2 L 1+ 5,PA L 5.609033E+07 -5. 53151l8E+07 1. 3811l36E-02 208 Cl H 1+ O,C,H 5 122996. 3 -121347.7 1. 31l0353E-02 209 C2 H 1+ O,C,H 5 37111178 -3691677 1. 323146E-02 210 C2 H 1+ G,ES 5 1112813 -1098271 1. 306794E-02 211 PCl L 1+ S,PA S 2117650 -2Jj.l!-Jj.32.8 1. 299109E-02 212 C2 L 1+ G,ES L 1. 719561E+07 -1. 697262E+07 1. 296791lE-02 213 Sl,2,Il L 1- G,ES L 1l.368387E+07 -Il. 312204E+07 1. 286131lE-02 211l 51,2,4 H 1+ S,PA L 2608251 -2574769 1. 283696E-02 215 RH L 1+ R L 6041498 -5961l656 1. 271905E-02 216 51,2,Il H 1- I S 156621 -151l667.3 1. 21l7423E-02 217 Cl L 1+ G,ES L 7.014597E+07 -6. 92821l8E+07 1. 230995E-02 218 W,S3 H 1+ I S 1. 034847E+07 -1. 022357E+07 1.206915E-02 219 C2 L 1+ O,C,H L 1.206058E+08 -1. 191556E+08 1. 202399E-02 220 C2 L 1+ I L 6.602995E+07 -6. 525954E+07 1. 166759E-02 221 PCl L 1+ O,C,H J. 2160909 -2135715 1. 165921lE-02 222 W,S3 H 1+ I L 2. 19518E+07 -2. 170101lE+07 1. 11l2329E-02 223 C2 L 1+ I H 1. 213618E+08 -1. 200064E+08 1. 116859E-02 224 51,2,4 L 1- I S 153550 -15181l6.8 1. 109237E-02 225 W, S3 H 1+ I H 1. 728747E+07 -1. 709605E+07 1. 107299E-02 226 W, S3 L 1+ 5,PA H 1. 801923E+07 -1. 781979E+07 1. 106811lE-02 227 C2 L 1+ 5,PA H 5.161151E+07 -5.104175E+07 1. 103944E-02
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
285
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT HPV OF PV/C RANK TYPE SMF s TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------226 C2 L 1+ R L 9.~26126E+07 -9. 32~702E+07 . 0109696 229 S1, 2,~ L 1- G,ES s 133350 -131904. 6 1. 063666E-02 230 S1,2,4 H 1+ I s 152700 -151061.2 1.073216E-02 231 S1,2,4 L 1+ I M 2404113 -2376766 . 0105422 232 C2 L 1+ G,ES M 6661596 -6591495 1. 037661E-02 233 W,S3 H 1+ O,C,H L 1. 014906E+07 -1. 004421E+07 1. 033054E-02 234 C2 L 1+ O,C,H M 2.079336E+06 -2.056241E+06 . 0101462 235 S1,2,4 L 1+ R L 5.959716E+07 -5. 900395E+07 9. 954055E-03 236 S1,2,4 H 1+ O,C,H M 1444349 -1430011 9.927134E-03 237 C2 L 1+ I s 3. 166373E+07 -3. 137139E+07 9.656057E-03 236 S1,2,4 L 1+ G,ES M 1371110 -1357651 9.615666E-03 239 C2 L 1+ S,PA s 5429936 -5376961 9.756471E-03 240 C2 L 1+ R M 7.637406E+07 -7. 761121E+07 9.733676E-03 241 C2 L 1+ O,C,H s 9. 317464E+07 -9. 227369E+07 9.669362E-03 242 W,S3 H 1+ O,C,H M 6630769 -6549126 9.45953E-03 243 S1,2,~ L 1+ I s 376965.8 -373439.9 9.406047E-03 244 W,S3 H 1+ S,PA L 1. 990906E+07 -1. 972462E+07 9. 254205E-03 245 W,S3 L 1+ S,PA s 1. 961228E+07 -1. 963356E+07 9.020526E-03 246 W,S3 L 1+ I L 1. 743207E+07 -1. 727625E+07 6.938744E-03 247 S1,2,4 L 1+ O,C,H M 1.693967E+07 -1. 676664E+07 6.915562E-03 246 PC2 L 1+ S,PA M 2046270 -2030132 6. 655247E-03 249 W, S3 H 1+ O,C,H s 6163653 -6091793 8.602431E-03 250 S1,2,4 H 1+ G,ES s 213360 -211466. 5 8.780818E-03 251 S1,2,~ L 1+ S,PA M 8438691 -8364737 8.763662E-03 252 PC1 L 1+ G,ES s 229313 -227309. 1 8.738739E-03 253 W,S3 L 1+ I M 2. 812389E+07 -2.788167E+07 8.612701E-03 254 RM L 1+ G,ES L 1125990 -1116727 8.226966E-03 255 W,S3 L 1+ G,ES M 417424.9 -414021. 9 8. 15239E-03 256 W,S3 L 1+ I s 1. 270248E+07 -1. 260061E+07 6.003756E-03 257 S1,2,4 L 1+ G,ES s 2667 -2645.786 7.953501E-03 256 PC2 L 1+ O,C,H M 411300 -408096.7 7.78817E-03 259 S1,2,4 H 1+ O,C,H L 2.025327E+07 -2.009616E+07 7.757465E-03 260 S1,2,4 L 1+ S,PA L 6. 369128E+07 -6. 319993E+07 7.71466E-03 261 C2 L 1+ R s 1. 257442E+07 -1. 247753E+07 7.705499E-03 262 S1,2,4 L 1+ O,C,H s 1392567 -1382840 6.984999E-03 263 W,S3 L 1+ O,C,H s 9.246719E+07 -9. 182174E+07 6.960354E-03 264 S1,2,4 H 1+ R M 2000640 -1986803 6.91637E-03 265 W,S3 H 1+ G,ES s 976913.5 -970166.2 6.906691E-03 266 W,S3 H 1+ R L 3. 261202E+07 -3. 238819E+07 6.863502E-03 267 S1,2,4 L 1+ O,C,H L 2.637323E+08 -2. 619612E+08 6.715575E-03 266 W,S3 H 1+ R s 1.060439E+07 -1. 053457E+07 6.584492E-03 269 W,S3 L 1+ R M 7. 351413E+07 -7. 306749E+07 6.075499E-03 270 S1,2,4 L 1+ G,ES L 6.011759E+07 -5. 976624E+07 5. 811168E-03 271 PC2 L 1+ O,C,H L 4.240817E+07 -4.216523E+07 5.726761E-03 272 PC2 L 1+ O,C,H s 350949. 3 -348942.4 5.718601E-03 273 W,S3 L 1+ O,C,H M 3.450453E+07 -3.430761E+07 5.707094E-03 274 W,S3 L 1+ S,PA L 4. 83555E+08 -4. 809422E+06 5.403343E-03 275 W,S3 L 1+ O,C,H L 1. 231206E+07 -1. 224564E+07 5. 394386E-03 276 PC2 L 1+ I s 231023. 6 -229797.4 5.307907E-03 277 W,S3 L 1+ R s 1. 202492E+08 -1. 1962E+06 5. 232279E-03 276 S1,2,4 L 1+ S,PA s 44710 -~83. 59 5.063945E-03 279 PC2 H 1+ O,C,H M 526464 -523854. 1 4.957451E-03 280 W,S3 L 1+ R L 4361000 -4339714 4.880969E-03 281 S1,2,~ L 1+ R M 1920646 -1911295 4.869487E-03 282 W,S3 L 1+ G,ES s 1356610 -1350353 4.759016E-03 283 W,S3 H 1+ R M 5270788 -5259647 2. 113666E-03 284 W,S3 H 1+ G,ES M 461422. 6 -480497.9 1. 920762E-03
285
------------------------------------------------------------------------------R/F STRUC OCCUP RETROFIT NPV OF PV/C RANK TYPE SMF S TYPE SIZE COST, $ BENEFIT, $ RATIO
------------------------------------------------------------------------------226 C2 L 1+ R L 9.~26126E+07 -9. 32~702E+07 .0109696 229 Sl, 2,~ L 1- G,ES S 133350 -131904.6 1. 063666E-02 230 S1,2,4 H 1+ I S 152700 -151061. 2 1.073216E-02 231 S1,2,4 L 1+ I H 2404113 -2376766 .0105422 232 C2 L 1+ G,ES H 6661596 -6591495 1. 037661E-02 233 W,S3 H 1+ O,C,H L 1. 014906E+07 -1. 004421E+07 1. 033054E-02 234 C2 L 1+ O,C,H H 2.079336E+06 -2.056241E+06 . 0101462 235 S1,2,4 L 1+ R L 5. 959716E+07 -5. 900395E+07 9. 954055E-03 236 S1,2,4 H 1+ O,C,H H 1444349 -1430011 9. 927134E-03 237 C2 L 1+ I S 3. 166373E+07 -3. 137139E+07 9. 656057E-03 236 S1,2,4 L 1+ G,ES H 1371110 -1357651 9. 615666E-03 239 C2 L 1+ S,PA S 5429936 -5376961 9. 756471E-03 240 C2 L 1+ R H 7. 637406E+07 -7. 761121E+07 9. 733676E-03 241 C2 L 1+ O,C,H S 9.317464E+07 -9. 227369E+07 9. 669362E-03 242 W,S3 H 1+ O,C,H H 6630769 -6549126 9. 45953E-03 243 S1,2,~ L 1+ I S 376965.8 -373439.9 9.406047E-03 244 W,S3 H 1+ S,PA L 1. 990906E+07 -1. 972462E+07 9. 254205E-03 245 W,S3 L 1+ S,PA S 1. 961228E+07 -1. 963356E+07 9.020526E-03 246 W,S3 L 1+ I L 1. 743207E+07 -1. 727625E+07 6. 938744E-03 247 S1,2,4 L 1+ O,C,H H 1. 693967E+07 -1. 676664E+07 6. 915562E-03 246 PC2 L 1+ S,PA H 2046270 -2030132 6. 655247E-03 249 W, S3 H 1+ O,C,H S 6163653 -6091793 8.602431E-03 250 Sl,2,4 H 1+ G,ES S 213360 -211466.5 8. 780818E-03 251 Sl,2,~ L 1+ S,PA H 8438691 -8364737 8. 763662E-03 252 PCl L 1+ G,ES S 229313 -227309. 1 8. 738739E-03 253 W,S3 L 1+ I H 2. 812389E+07 -2. 788167E+07 8. 612701E-03 254 RH L 1+ G,ES L 1125990 -1116727 8. 226966E-03 255 W,S3 L 1+ G,ES H 417424.9 -414021. 9 8. 15239E-03 256 W,S3 L 1+ I S 1. 270248E+07 -1. 260061E+07 6.003756E-03 257 S1,2,4 L 1+ G,ES S 2667 -2645.786 7. 953501E-03 256 PC2 L 1+ O,C,H H 411300 -408096.7 7. 78817E-03 259 Sl,2,4 H 1+ O,C,H L 2.025327E+07 -2.009616E+07 7. 757465E-03 260 S1,2,4 L 1+ S,PA L 6. 369128E+07 -6. 319993E+07 7. 71466E-03 261 C2 L 1+ R S 1. 257442E+07 -1. 247753E+07 7.705499E-03 262 S1,2,4 L 1+ O,C,H S 1392567 -1382840 6. 984999E-03 263 W,S3 L 1+ O,C,H S 9. 246719E+07 -9. 182174E+07 6. 960354E-03 264 Sl,2,4 H 1+ R H 2000640 -1986803 6. 91637E-03 265 W,S3 H 1+ G,ES S 976913.5 -970166.2 6.906691E-03 266 W,S3 H 1+ R L 3.261202E+07 -3. 238819E+07 6. 863502E-03 267 Sl,2,4 L 1+ O,C,H L 2. 637323E+08 -2. 619612E+08 6. 715575E-03 266 W,S3 H 1+ R S 1.060439E+07 -1. 053457E+07 6. 584492E-03 269 W,S3 L 1+ R H 7.351413E+07 -7. 306749E+07 6.075499E-03 270 S1,2,4 L 1+ G,ES L 6.011759E+07 -5. 976624E+07 5.811168E-03 271 PC2 L 1+ O,C,H L 4. 240817E+07 -4. 216523E+07 5. 726761E-03 272 PC2 L 1+ O,C,H S 350949. 3 -348942.4 5. 718601E-03 273 W,S3 L 1+ O,C,H H 3. 450453E+07 -3. 430761E+07 5.707094E-03 274 W,S3 L 1+ S,PA L 4. 83555E+08 -4. 809422E+06 5.403343E-03 275 W,S3 L 1+ O,C,H L 1. 231206E+07 -1. 224564E+07 5. 394386E-03 276 PC2 L 1+ I S 231023. 6 -229797.4 5.307907E-03 277 W,S3 L 1+ R S 1. 202492E+08 -1. 1962E+06 5. 232279E-03 276 S1,2,4 L 1+ S,PA S 44710 -~83. 59 5.063945E-03 279 PC2 H 1+ O,C,H H 526464 -523854. 1 4. 957451E-03 280 W,S3 L 1+ R L 4361000 -4339714 4. 880969E-03 281 S1,2,~ L 1+ R H 1920646 -1911295 4. 869487E-03 282 W,S3 L 1+ G,ES S 1356610 -1350353 4. 759016E-03 283 W,S3 H 1+ R H 5270788 -5259647 2.113666E-03 284 W,S3 H 1+ G,ES H 461422. 6 -480497.9 1. 920762E-03