info what? using infomaster to develop a rehabilitation ... · develop a rehabilitation and...
TRANSCRIPT
© 2014 HDR, Inc., all rights reserved.
Info What? Using InfoMaster to
Develop a Rehabilitation and
Replacement Plan
Amanda Lei - City of San Jose
Alex Palmatier - HDR
Forecasting and Budget
Lessons Learned
Background
Data Management
Prioritizing CCTV Work
Repair, Rehabilitation or Replacement?
Background Sewer System, Assessment Approaches and InfoMaster
Service Area Approximately 180 square miles
Approximately 2,200 miles of sewer
Approximately 985,000 residents
16 Lift Stations
90% clay pipe
85% 10-inch or smaller
45,000 Manholes
Average system age approximately 45 years
Median age is 52
City of San Jose
On-Call/As-Needed (project design/corrective maintenance)
Inspections Completed by Contractors and City Crews
2010 Started SSCA Pilot program
Currently Completed System Sample Set - 371 miles (usable data)
Remainder Focused on High SSO Areas
Historical CCTV Program
For sewer mains with diameter smaller than 18-inches:
o Inspecting mains within 200 feet of Water Bodies within two (2) years of CD Effective Date
o Repair or replace mains that receive a PACP rating of 5 within one (1) year of determination
o Repair or replace mains that receive a PACP rating of 4 within five (5) years of determination
o Assess 70% of City’s collection system within ten (10) years of CD Effective Date
o Reporting of SSOs that discharge to a critical habitat area as defined by the Endangered Species Act
3rd Party Consent Decree
Historical Approach to Renewal Planning
• Budget Based on Last Year
• Little knowledge of system risks
Backward Looking
• Projects determined as problems arise during the year Reactive
• Do as many projects as you can afford each year
Budget Constrained
• Money is spent but overall risk may not have been reduced much
Ignores asset and system
risks
Represents a New Focus for Most Utilities San Jose Goal – Risk Based Planning
Risk-Based Renewal Planning
• Based on asset risk scores throughout system and long term forecasts of risk and cost
Forward Looking
• High risk assets slotted for renewal before failure occurs Proactive
• Budget could be determined based on agreed risk targets for system
Risk or Budget Constrained
• High risk assets addressed first
• Budget may rise or fall to meet risk targets
Focused on risk
management
Analytical and Optimization Software Package by Innovyze
Built On and Runs Inside ArcGIS Infrastructure
Out-of-the-Box
Improves Data Collection and Validation
Smarter Decision Making
InfoMaster What?
Workflow Diagram What Does it Do?
Likelihood of Failure
• Pressure Changes
• Roughness Hydraulic
Model
• Age
• Material Infrastructure
Data
• Soil Type
• Railroads/Fault Lines GIS Data
• Break History
• Repairs/Lining CMMS & Work
Orders
Consequence of Failure
• Pipe/Valve Criticality
• Flow Delivered Hydraulic Model
• Hospitals, Schools, etc
• Power, Industry, etc. Critical Facilities
• Population Density
• Street Paving GIS Data
• Traffic Analysis
• Community Relations Other
Rehabilitation Engine Budget Scenarios Rehabilitation Costs
Prioritized Capital Plan
Calculation of
Risk
Multiple Calculation
Options
GIS
Data Management
CCTV Viewer
Risk Profile
Capital Planning
InfoMaster Why?
Data Management
Data
o Where is it coming from?
• Cond Assessment CCTV
• CIP Project CCTV
• O&M CCTV
o Where does it go?
• Storage
• Access
o How do we maintain it?
• Formats
• Quality Control
Challenges
Use GIS as Database of Record
o Data stored in SDE or GDB
o Hyperlink to videos for viewing
Combine Multiple Data Sources
o CCTV – PACP based
• Can accept anything tabular
o Single Inspection “Instance” or Multiple
o Can Use Any GIS Data
Solutions
Review of the Standards
o Update to require PACP database as a deliverable
o Develop a frequency interval for data submission
o Develop and implement a QA criteria and QC protocol
o Develop requirements for the use of specific codes
• Sag issues
Data Collection
Database Setup
o Project
o System
Mapping Asset Attribute
o Based on ESRI Data Model
• Connects to SDE or GDB
Importing CCTV Data
o Imports PACP Database
Creating the InfoMaster Workspace
Importing GIS
Data Validation
Defect Mapping
Defect Scoring
Rehab Plan
Importing CCTV
Missing MH
Missing Pipe
Pipe Length Errors
Validation Errors
Profile View Visualizing the Data
Map View Visualizing the Data
Prioritizing CCTV Work Cost Effective, Risk Based Assessment
Which Segments to CCTV Next?
Bidding Out the CCTV Work (spec language, data formats, point repairs)
Challenges
Develop Risk Profile
o Determine Likelihood of Failure (LoF) Criteria
o Determine Consequence of Failure (CoF) Criteria
o Use CCTV History
o Reliability Analysis
• Life Expectancy
• Failure Condition
Solutions
Based on Pipe Attributes
Environmental Conditions
o Soil type
o Land use
History
o Tasks
o Incidents
Likelihood of Failure
Pipe Attribute SSO
Risk =
(Pipe Attribute Spill Count
/ System Spill Count)
(Pipe Attribute Count /
System Count)
Determine a Correlation Between
Categories and Failure
o Identifies if they should be used and how to
score
LoF Evaluation
Cleaning Frequency
Age
Diameter
Material
Asset Type
Slope
Groundwater Table Location (with respect
to pipe)
Pipe Depth
Easement Type
Restaurant Proximity
LoF Attributes Evaluated
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0"-5" 6" 7"-10" 12"-21" 23"-45" >45"
Att
rib
ute
SS
O/S
top
pag
e R
isk
Diameter
Pipe Attribute SSO Risk
Pipe Attribute StoppageRisk
Pipe AttributeStoppage&SSO Risk
Average Risk
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
AB
S
AC
P
BR
I
CI
CIP CP
CP
L
DIP
HD
PE
PE
PV
C
PV
L
RC
P
VC
P
Att
rib
ute
SS
O/S
top
pag
e R
isk
Pipe Material
Pipe Attribute SSO Risk
Pipe Attribute StoppageRisk
Pipe AttributeStoppage&SSO Risk
Average Risk
Physical Attributes
o Diameter
o Flow
Critical Facilities
Health Impacts
Regulatory
Political
Consequence of Failure
Creeks (200 ft)
Schools (200 ft)
Hospitals
Railroad (intersect)
Right-of-Way
Storm Drain (200 ft)
Parks (200 ft)
Diameter
Length
Capacity
Land Use
Major Roads
CoF Criteria Identified by:
o Staff
o Available Data
o CD Requirements
Several Different Combinations Evaluated
for Risk Profile
CoF Evaluation
Use LoF and CoF Criteria to Generate Risk
Profile
Risk Profile
Combine Different Parameters, Risk Distributions and Risk Weights Risk Profile Comparison
Created 4 Different Combinations of LoF and CoF
Focused on SSO Likelihood
Consequence Looked at Combinations of Health and CD Components
Risk Profile Comparisons
LOF COF
Install Date Creeks (200')
Diameter Schools (200')
Material Hospitals
Slope Railroad (Intersect)
Depth Right of Way
Storm Drain (200')
Parks (200')
Roadways
Diameter/Length/Capacity
0
10
20
30
40
50
60
70
0
50
100
150
200
250
300
350
400
450
Co
mp
lete
/Pla
nn
ed C
CT
V (
Mi)
Fu
ture
CC
TV
(M
i)
Normalized Risk
FUTURE CCTV CCTV COMPLETE PLANNED
0
10
20
30
40
50
60
70
80
0
50
100
150
200
250
300
350
400
450
150-
200
200-
250
250-
300
300-
350
350-
400
400-
450
450-
500
500-
550
550-
600
600-
650
650-
700
700-
750
750-
800
800-
850
850-
900
900-
950
950-
1000
Co
mp
lete
/Pla
nn
ed C
CT
V (
Mi)
Fu
ture
CC
TV
(M
i)
Normalized Risk
FUTURE CCTV CCTV COMPLETE PLANNED
0
10
20
30
40
50
60
70
0
100
200
300
400
500
600
Co
mp
lete
/Pla
nn
ed C
CT
V (
Mi)
Fu
ture
CC
TV
(M
i)
Normalized Risk
FUTURE CCTV CCTV COMPLETE PLANNED
LoF CoF
Diameter
Creek (200')
Age
Storm Drain
(200')
Material
Population
Density
Asset Type
Cleaning
Frequency
0
10
20
30
40
50
60
70
80
90
0
100
200
300
400
500
600
700
800
900
50-1
00
100-
150
150-
200
200-
250
250-
300
300-
350
350-
400
400-
450
450-
500
500-
550
550-
600
600-
650
650-
700
700-
750
750-
800
800-
850
850-
900
900-
950
950-
1000
Co
mp
lete
/Pla
nn
ed C
CT
V (
Mi)
Fu
ture
CC
TV
(M
i)
Normalized Risk
FUTURE CCTV CCTV COMPLETE PLANNED
Repair, Rehabilitation or Replacement? Making Consistent, Defensible Decisions
o No Overall Condition Based Rehab Plan
o No Methodical Decision Tool
• Used “Expert” Method
o Creating the Decision Trees
• Repair Methods Used
• Decision Points
• Defect Lengths Determination
• Is Our Decision Model Reflective of Reality?
Challenges
Create Standards
o Data Collection and Defect Coding
o Defect Repair Methods
o Acceptable Number of Defects/Pipe
o At what point do we replace?
Solutions
Which discrete defects are fixed and how? Defect Repair Methods
How are pipeline repairs determined? Rehabilitation Methods
Cost Tables
Typical Decisions
Hydraulic
capacity
OK?
Part of
an existing
project?
Defects that
increase risk of
SSO?
Maintenance costs
versus capital costs?
Pipe smaller
than current
standards?
Number of
defects per
100 feet?
Length of
continuous
defects?
Started with 3 Existing Decisions
o Reviewed agency characteristics and goals
• System age and size
• R&R goals
Processed Existing 322 Miles of San Jose CCTV
Development Approach
Yields of Miles of Construction from 322 CCTV Miles
Yields and Costs Used for Forecasting
Review of Results
Agency Yield Cost (millions)
A 31% $85
B 10% $35
C 38% $70
What is Currently Being Used?
Are There Any Decisions from Previous Models that are Applicable to San Jose?
What Should Our Yield Be?
o What codes are driving decisions?
Developing San Jose’s Model
Data – MWLS Codes
o Are they truly sags?
What to Do with 6” Pipe
Major Issues to Address
Yield from initial 322 miles of San Jose CCTV: Approximately 32% construction yield
• 42.4 Miles of Rehab
» Allows 6” pipe to be lined
• 56.5 Miles of Repair
• 0.6 Miles of Replacement
• 3.6 Miles of Evaluate Sags
Emphasis away from replacement
Approximately $40 Million in Backlogged Projects
Results
Forecasting and Budget Short Term and Long Term CIP
Now That We Have Information – What’s Next?
Has some already been built?
o How do we reconcile?
How Do We Forecast Future CIP
o Is 32% a realistic yield?
From Budgeting to Project Delivery, How Will InfoMaster be Used?
o Impacts to budget
o Planning window
• 5 or 10 years
• CD impacts
Challenges
Estimating Future Yield and Cost Forecasting
$38,400,737
$-
$5,000,000
$10,000,000
$15,000,000
$20,000,000
$25,000,000
$30,000,000
$35,000,000
$40,000,000
$45,000,000
0 1 2 3 4 5 6 7 8 9 10
Est
imat
ed C
ost
Year
Model 2 Cost Forecast at 5% Yield Decrease Per Year
Initial Cost
Forecast Cost
Sanitary Sewer Condition Assessment (SSCA) Pilot
o Randomly selected ~50 miles of pipe to CCTV
o Identified to determine approximate condition of entire system
o Based on statistical sample of:
• Material
• Diameter
• Age
Results of San Jose Model
o 10% Construction Yield
Statistical Sample Set
Lessons Learned Issues, The Future, and Questions
Data Compatibility
o SDE vs GDB
• Numeric vs Text
Defect Length and Count of Defects
o PACP vs Count
o Continuous Defects
CapPlan Limitations
Asset Based vs Project Approach
Lessons Learned - Software
Where are we in implementation?
Completed Model
Validated Decision Process
o Compared to previous projects
Begin Rollout
o Setup on City servers
o Data mapping
o Staff training
Project Status
Questions?
Forecast with SSCA Yield Percentage Statistical Sample Set
0 1 2 3 4 5 6 7 8 9 10
Initial Cost $47,308,210
Forecast Cost $14,563,747 $9,970,632 $6,811,930 $4,653,571 $3,165,471 $2,171,385 $1,480,737 $1,018,555 $675,107 $447,983
$-
$5,000,000
$10,000,000
$15,000,000
$20,000,000
$25,000,000
$30,000,000
$35,000,000
$40,000,000
$45,000,000
$50,000,000
Co
st
CIP 10 Year Cost Forecast
Straight-line Yield
$-
$5,000,000
$10,000,000
$15,000,000
$20,000,000
$25,000,000
$30,000,000
$35,000,000
$40,000,000
$45,000,000
$50,000,000
0 1 2 3 4 5 6 7 8 9 10