lndlana utiliw vectren dsm action plan: commission final

FILED FEB 0 3 2006

lNDlANA UTILIW

REGULATORY COMMISSION Vectren DSM Action Plan: Final Report

Prepared for: Vectren Energy Delivev of Indiana, Inc.

Evansville, Indian a

Prepared by: Forefront Economics Inc.

and H. Gil Peach & Associates LLC

with contributions fiom: Mark E. Thompson

H. Gil Peach Luisa Freeman-Fawardin

Howard Reichmuth

December 9,2005

Forefront Economics Inc. and H. Gil Peach and Associates LLC. Vectren DSMAction Plan: Final Report. Report prepared for Vectren Energy Delivery of Indiana, Inc., Evansville, IN, Dec. 2005.

Vecfren DSM Action Plan: Final Report Table of Contents

TABLE OF CONTENTS

.. Table of Contents .................................................................................................................................................. n . . List of Tables and Figures .................................................................................................................................... 11 . .

List of Tables ...................................................................................................................................................... 11 . .

List of Figures .................................................................................................................................................... 11

I . Executive Summary .......................................................................................................................................... 2 I1 . Introduction ..................................................................................................................................................... 2

. . ........................................................................................................................................ Organization of Report 2 ........................................................................................................................................ III . Market Assessment 2

............................................................................................................................................................. Overview 2 ........................................................................................................................................................... Residential 2 ......................................................................................................................................................... Commercial 2

............................................................................................................................................................. Industrial 2 IV . Conservation Potential ................................................................................................................................... 2

............................................................................................................................................................. Overview 2 .............................................................................................................................................. Technical Potential 2

..................................................................................................................... Conservation Measure Assessment 2 V . Program Design Process .................................................................................................................................. 2

............................................................................................................................................................. Approach 2 .................................................................................................................. DSM Program Development Process 2 ................................................................................................................ Policy and Regulatory Considerations 2

..................................................................................................................................... Target Market Segments 2 ......................................................................................................................................... Technology Screening 2

DSM Program Designs ........................................................................................................................................ 2 VI . DSM Program Plans ...................................................................................................................................... 2

Program 1 . Small Buildings Energy Efficiency Program .................................................................................. 2 ................................................................................. Program 2 . General Services Energy Efficiency Program 2

......................................................................................... Program 3 . Customized Energy Efficiency Program 2

Program 4 . Hospitality Industry Energy Efficiency Program ............................................................................ 2 ........................................................................ Program 5 . Multi-Family Building Energy Efficiency Program 2

Program 6 . Innovative Energy Efficiency Technologies Research and Demonstration Program ...................... 2 ................................................................................................... Program 7 . Energy Efficient Builder Program 2

....................................................................... Program 8 . New Program Development and Regulatory Affairs 2 Program 9 . Public Education and Outreach Program ......................................................................................... 2

VII . Program Cost Effectiveness ......................................................................................................................... 2 .................................................................................................................................... Expected Program Costs -2

.................................................................................................................................. Expected Program Savings 2 .............................................................................................................................................. Avoided Gas Costs 2

.................................................................................................................................. Cost Effectiveness Results 2 VIII . Alternative Forecast and Policy Parameters ............................................................................................ 2

.................................................................................................................... Low-Income Area Policy Concerns 2

Prepared by Forefront Economics and H . Gil Peach & Associates Page i

Vectren DSMAction Plan: Final Re~ort Table o f Contents

Overall Policy Concerns . An Alternative Forecast ........................................................................................... 2 M . Program Evaluation ....................................................................................................................................... 2

First Steps ............................................................................................................................................................ 2 ........................................................... Evaluation for Program 1 . Small Buildings Energy Efficiency Program 2

Evaluation for Program 2 . General Services Energy Efficiency Program ......................................................... 2 Evaluation for Program 3 . Customized Energy Efficiency Program ................................................................. 2 Evaluation for Program 4 . Hospitality Industry Energy Efficiency Program .................................................... 2

................................................ Evaluation for Program 5 . Multi-Family Building Energy Efficiency Program 2 Evaluation for Program 6 . Innovative EE Technologies Research and Demonstration Program ...................... 2

........................................................................... Evaluation for Program 7 . Energy Eff~cient Builder Program 2

Evaluation for Program 8 . New Program Development and Regulatory Affairs ............................................... 2 Evaluation for Program 9 . Public Education and Outreach Program ................................................................. 2

Appendix A . Data Sources and References ........................................................................................................ 2 Primary (Vectren) ................................................................................................................................................ 2 Secondary ........................................................................................................................................................... -2 References ........................................................................................................................................................... 2

Appendix B . Methodology .................................................................................................................................... 2 Appendix C . Technology Characteristics and Assumptions ............................................................................. 2 Appendix D . Cost Effectiveness Methodology ................................................................................................... 2

Technology Cost Effectiveness ........................................................................................................................... 2 Program Cost Effectiveness ................................................................................................................................ 2

Appendix E . Summary of Natural Gas Program Rebates, Minnesota ............................................................ 2 Natural Gas Utility Conservation Rebates .......................................................................................................... 2

Appendix F . Energy Conservation Measure Assumptions ............................................................................... 2

Prepared by Forefront Economics and H . Gil Peach & Associates Page ii

Vectren DSMAction Plan: Final Report Table of Contents

LIST OF TABLES AND FIGURES

List of Tables

.................. Table 1 . Vectren North Total Usage. Technical Potential and Program Savings (millions of therms) 2 .................................. Table 2 . Vectren North Customers and Usage (unadjusted for weather) by Rate Schedule 2

Table 3 . Vectren North Total Annual Gas Use by End-Use ................................................................................... 2 .................................................................................................... Table 4 . Housing by Heating Fuel and Tenancy 2

Table 5 . New Gas Services. Construction. and Gas Market Share ......................................................................... 2 Table 6 . Number of Residential Customers by Segment ........................................................................................ 2 Table 7 . Average Weather-Normalized Usage. SF Residential by Meter Set Year ............................................... 2 Table 8 . Number of Residential Customers by Segment ........................................................................................ 2 Table 9 . Residential Survey Weights by Survey Quadrant ..................................................................................... 2 Table 10 . Appliance and End-Use Installation Rates from Residential Survey ..................................................... 2 Table 11 . Residential Sector Annual Gas Usage by End-Use ................................................................................ 2 Table 12 . Residential Sector Monthly Usage by End-Use ..................................................................................... 2 Table 13 . Average Use per Residential Customer (therms) ................................................................................... 2 Table 14 . Number of Businesses and Employment, Vectren North Service Area ................................................. 2 Table 15 . Age of Gas Buildings - Customized Regional Area of CBECS Survey ................................................ 2

............................................... Table 16 . Size of Gas Buildings - Customized Regional Area of CBECS Survey 2 Table 17 . Natural Gas End-Uses in Commercial Buildings - Customized Regional Area of CBECS Survey ...... 2

................................ Table 18 . Vectren North Non-Residential Premises and Loads by NAICS-Based Segments 2 ........................................................................................ Table 19 . Assumed Commercial Customer Distribution 2

.............................................................................. Table 20 . Commercial Sector Annual Gas Usage by End-Use 2 .................................................................................... Table 2 1 . Commercial Sector Monthly Usage by End-Use 2

Table 22 . Average Use per Commercial Customer by Segment (therms) .............................................................. 2 Table 23 . Assumed Industrial Customer Distribution ............................................................................................ 2

................................................................................... Table 24 . Industrial Sector Annual Gas Usage by End-Use 2 ........................................................................................ Table 25 . Industrial Sector Monthly Usage by End-Use 2

Table 26 . Average Use per Industrial Customer by Segment (therms) .................................................................. 2 Table 27 . Demand Side Planning Elements ............................................................................................................ 2 Table 28 . Summary of Technical Potential by Planning Element (millions of therms) ......................................... 2 Table 29 . DSM Technology Assessment, Small Buildings .................................................................................... 2 Table 30 . DSM Technology Assessment. Large Buildings .................................................................................... 2

................................................................................................... Table 3 1 . Net Savings by Measure (themslyear) 2 .......................................................................................................... Table 32 . Small Building Ranked Measures 2 .......................................................................................................... Table 33 . Large Building Ranked Measures 2

Table 34 . DSM Technologies for Natural Gas End-Uses and Customer Segments ............................................... 2 Table 35 . Sorting of DSM Technologies ................................................................................................................ 2 Table 36 . Recommended programs and Technology Groupings ........................................................................... 2 Table 37 . Measures and Incentives . Small Buildings Energy Efficiency Program ............................................... 2 Table 3 8 . Estimated Participation and Savings . Small Buildings Energy Efficiency Program ............................. 2 Table 39 . Estimated Five-Year Program Budget . Small Buildings Energy Efficiency Program .......................... 2 Table 40 . Measures and Incentives - General Services (GIs) Energy Efficiency Program ................................... 2 Table 4 1 . Estimated Participation and Savings - General Services Energy Efficiency Program ........................... 2

Prepared by Forefront Economics and H . Gil Peach & Associates Page iii

Vectren DSM Action Plan: Final Report Table of Contents

Table 42 . Estimated Five-Year Program Budget . General Services Energy Efficiency Program ........................ 2 Table 43 . Incentives . Customized Energy Efficiency Program ............................................................................. 2 Table 44 . Estimated Participation and Savings . Customized Energy Efficiency Program ................................... 2

................................. Table 45 . Estimated Five-Year Program Budget . Customized Energy Efficiency Program 2 ......................................... Table 46 . Measures and Incentives . Hospitality Industry Energy Efficiency Program 2

...................... Table 47 . Estimated Participation and Savings . Hospitality Industry Energy Efficiency Program 2 .................... Table 48 . Estimated Five-Year Program Budget . Hospitality Industry Energy Efficiency Program 2

Table 49 . Measures and Incentives . Multi-Family Building Energy Efficiency Program .................................... 2 Table 50 . Estimated Participation and Savings - Multi-Family Building Energy Efficiency Program ................. 2 Table 5 1 . Estimated Five-Year Program Budget . Multi-Family Building Energy Efficiency Program ............... 2

............................................................... Table 52 . Measures and Incentives . Energy Efficient Builder Program 2 Table 53 . Estimated Participation and Savings . Energy Efficient Builder Program ............................................. 2

.......................................... Table 54 . Estimated Five-Year Program Budget . Energy Efficient Builder Program 2 ........................................................................................ Table 55 . Public Education Budget Items and Amounts 2

............................................................................................................................ Table 56 . Total Program Budget 2 Table 57 . Program Staffing Assumptions ............................................................................................................... 2 Table 58 . Summary of Program Assumptions ........................................................................................................ 2 Table 59 . Total Program Savings ........................................................................................................................... 2 Table 60 . Real Levelized Avoided Cost per Them ............................................................................................... 2

.................................................................................................. Table 6 1 . Cost Effectiveness Results by Program 2 Table 62 . Global Assumptions Used in Cost Effectiveness Calculations .............................................................. 2

............................................................................................................... Table 63 . LIHEAP, Valuable but Failing 2 Table 64 . Layout for Non-Equivalent Control Group Evaluation Design .............................................................. 2 Table 65 . Layout for Non-Equivalent Control Group Evaluation Design with Double Pre-Test ........................... 2 Table 66 . Benefits and Costs by Cost Effectiveness Test ....................................................................................... 2 Table 67 . Avoided Cost of Gas Details .................................................................................................................. 2 Table 68 . Energy Conservation Measure Assumptions Used in Program Planning and Design ............................ 2

Prepared by Forefront Economics and H . Gil Peach & Associates Page iv

Vectren DSMAction Plan: Final Report Table of Contenfs

List of Figures

Figure 1 . DSM Action Plan Project Timeline ......................................................................................................... 2 Figure 2 . Organization of the Report and Relationship to Project Tasks ............................................................... 2

......................................................................................... Figure 3 . Total Vectren North Gas Sales by Rate Class 2 ............................................................................... Figure 4 . Total Vectren North Gas Sales by Energy End-Use 2

.................................................................................................................. Figure 5 . Vectren North End-Use Map -2 ............................ Figure 6 . Residential Housing Units Permitted for Construction, Vectren North Service Area 2

................................................................................................ Figure 7 . Residential Gas Usage by Housing Type 2 Figure 8 . Month-to-Month Normal Gas Usage by End-Use ................................................................................... 2 Figure 9 . Actual and Modeled Total Residential Gas Usage .................................................................................. 2

.......................................................................... Figure 10 . Trends in Employment, Vectren North Service Area -2 ............................................................................. Figure 1 1 . Commercial Monthly Gas Usage by Rate Schedule 2

.................................................................. Figure 12 . Commercial Monthly Normalized Gas Usage by End-Use 2 .............................................................................. Figure 13 . Actual and Modeled Total Commercial Gas Usage 2

.................................................................................. Figure 14 . Industrial Monthly Gas Usage by Rate Schedule 2 ....................................................................... Figure 1 5 . Industrial Monthly Normalized Gas Usage by End-Use 2

................................................................... Figure 16 . Daily Gas End-Uses - Average Industrial NlN3 Customer 2

.................................................................. Figure 17 . Daily Gas End-Uses - Average Industrial N140 Customer -2 Figure 18 . Daily Gas End-Uses - Average Transportation T600x Customer ......................................................... 2 Figure 19 . Daily Gas End-Uses - Average Transportation T700x Customer ......................................................... 2 Figure 20 . Actual and Modeled Total Industrial Gas Usage .................................................................................. 2

................................................................................................ Figure 2 1 . Residential Technical Potential Models 2 Figure 22 . Small Building Technical Potential Models .......................................................................................... 2 Figure 23 . Large Building Technical Potential Models .......................................................................................... 2

.................................................................................... Figure 24 . Cost Effectiveness Spectrum - Small Buildings 2 Figure 25 . Cost Effectiveness Spectrum - Large Buildings .................................................................................... 2 Figure 26 . The DSM Program Development Process ............................................................................................. 2 Figure 27 . Vectren Customers by Rate Class ......................................................................................................... 2 Figure 28 . End-Uses of Gas by Vectren Customers ............................................................................................... 2 Figure 29 . Conceptual Illustration of Technical, Market and Program Potential ................................................... 2 Figure 30 . The Winter of 2005-2006 (from CBPP Study) ...................................................................................... 2 Figure 3 1 . Trends in Real Income, Indiana Families with Children ....................................................................... 2

................................................................................................................. Figure 32 . Air and Water Temperatures 2

Prepared by Forefront Economics and H . Gil Peach &Associates Page v

Vectren DSMActwn Plan: Final Report Table of Contents

Prepared by Forefront Economics and H. Gil Peach & Associates Page vi

Vectren DSMAction Plan: Final Re~ort I. Executive Summarv

I. EXECUTIVE SUMMARY

This document presents a long-term Demand Side Management (DSM) Action Plan for the Vectren North gas

distribution company. The DSM Action Plan was prepared by Forefront Economics Inc. and H. Gil Peach and

Associates with consultation and review by a Project Advisory Board consisting of utility management and

interested parties. The design, implementation, oversight and cost effectiveness of natural gas DSM programs

are addressed in the DSM Action Plan. Findings from our analysis are shown in the table below.

Table 1. Vectren North Total Usage, Technical Potential and Program Savings (millions of therms)

The technical potential tells us that if the gas saving technologies identified in this report were applied across all

applicable customers, without regard to market or economic constraints, weather normalized annual gas usage

could be reduced to nearly half of current consumption. Annual DSM savings, shown in the table, provide a far

more realistic savings number for planning purposes. The DSM programs presented in this report are expected

to result in nearly 1 1 million therms of annual energy savings by the fifth year of operation, 1.4 percent of

current usage. At less than half the cost per them of delivered gas supply, the demand side resource is shown to

be highly cost effective. Net energy costs in the Vectren North service area are expected to be $37 million

lower if the DSM programs are implemented. Most of these benefits accrue to residential and small commercial

customers.

The approach taken in developing the set of recommended DSM programs for Vectren North's consideration

was generally as follows: (1) conduct a market assessment for determining gas usage and characteristics across

customer groups, (2) review a comprehensive list of DSM technologies for saving energy, (3) consider the

appropriateness of selected technologies for Vectren North's service territory in terms of markets, cost

effectiveness and accessibility to products, (4) group the highest potential technologies into logical sets for

marketing and outreach, (5) design program strategies to promote the technologies based on industry best

practices, (6) consider the cost effectiveness of the designed program, including costs to Vectren and to

participating customers, and (7) describe a final set of recommended program designs that make the most sense

for the utility and have a strong potential for delivering cost effective energy savings.

The final set of program designs is listed below:

1. Small Buildings Energy Efficiency Program 2. General Services Energy Efficiency Program 3. Customized Energy Efficiency Program

Prepared by Forefront Economics and H. Gil Peach & Associates Page I

Vectren DSM Action Plan: Final Report I. Executive Summary

4. Hospitality Industry Energy Efficiency Program 5. Multi-Family Building Energy Efflciency Program 6. Innovative Energy Efflciency Technologies Research and Demonstration Program 7. Energy Eff~cient Builder Program 8. New Program Development and Regulatory Affairs 9. Public Education and Outreach Program

All of these programs, with the exception of 6, 8 and 9, are expected to deliver measurable energy savings.

Programs 6, 8 and 9 are research and development and program support activities. The programs that deliver

energy savings were all subjected to cost effectiveness analysis as described in Section VII; all of these

programs showed positive cost effectiveness results.

Prepared by Forefront Economics and H. Gil Peach & Associates Page 2

Vectren DSMAction Plan: Final Report IL Introduction

11. INTRODUCTION

In June 2005, Vectren Energy ~e l ivery of Indiana, Inc. contracted with Forefront Economics Inc. and H. Gil

Peach and Associates to develop a Long-Term Demand Side Management (DSM) Action Plan for the Vectren

North service territory (Vectren North). Development of the Action Plan was undertaken with the consultation

and review of an Advisory Board consisting of representatives of Vectren North, the Citizens Action Coalition

of Indiana (CAC), the Indiana Gas Industrial Group (IGIG), and the Indiana Office of Utility Consumer

Counselor (OUCC). The Action Plan includes assessment of the potential for cost effective natural gas DSM

programs. The design, implementation, oversight and cost effectiveness of natural gas DSM programs are to be

addressed in the Action Plan. Vectren North's gas service territory encompasses 4 1 counties in central Indiana

surrounding Indianapolis (excluding Marion County) and the southeastern portion of Indiana.

The DSM planning project was started in June 2005 and was completed with delivery of this report to the

Advisory Board on December 9,2005. Draft reports of research results have been reviewed and discussed with

the project Advisory Board at various junctures and comments addressed. A timeline of key project milestones

is provided in Figure 1.

December 1,2004 Order Approvedfor b n g Term DSM r

March H, 2005 3 Cwultant pmpods submitted for selection r Sept 13,2005 DnhDSMActiDn Plan

DnR DSM Rcgram Aans, -d consavation 7

I I week 1 2 3 4 5 6 7 8 9 10 11 12 13 114 15 16 17 18 19 120 21 22 23 24 25 26 27 28 23 30

July 26,2005 Refimina Market A n e s w n t delivered

June 14,2005 Ro.ect Kckoff with Forefrmt Econmics L Nov. 8,2005

Review Meeb'n to finalize Plan Decmber 9,2005

Final DSM Ac6m Plan Delivered to Vectren North "L

Figure 1. DSM Action Plan Project Timeline

Organization of Report

This document presents a Long-Term DSM Plan for the Vectren North service territory. The DSM planning

project that resulted in the recommendations in this report consisted of three primary tasks:

1. Market Assessment 2. Conservation Potential 3. DSM Program Design

While each of these primary areas of activity can be viewed separately, they are related sequentially with each task building on the preceding tasks.


Vectren DSMAction Plan: Final Report LI. Introduction

Figure 2 illustrates the flow of work on the project and the sections within this report that correspond to the

findings for each task.

Section 111 Section IV Section IV

universe of DSM markets technologies and technologies 1 "r" utility-specific 'Zt0 r Section V

PACKAGE screened DSM

technologies into groupings

according to target markets

Section VI Section VJI

DSM programs program cnsts based on best- and benefits

practice marketing

techniques and policy

considerations

Figure 2. Organization of the Report and Relationship to Project Tasks

Section IX

programs for process

improvements and savings

impact


Vectren DSMAction Plan: Final Report 111. Market Assessment

111. MARKET ASSESSMENT

Overview

The market assessment begins by describing the Vectren North service territory in terms of households,

businesses and customer data. A description of the number of customers in the basic rate classes and a

corresponding energy usage model is created for estimating the gas sales to these customers in terms of basic

gas energy end-uses; such as, space heat, water heat, cooking, dryers and process energy. The energy end-use

estimates are developed in the form of energy usage models that then provide a starting base case for estimating

the technical potential and energy savings and cost effectiveness of a wide variety of demand side measures and

programs.

The gas energy use estimates are normalized to long-term weather conditions by using the energy usage models

applied to a typical or normal year. All energy use and end-use estimates reported here have been normalized to

the 30-year monthly temperature averages for Indianapolis. Though the energy use estimates reported here are

for a normal year, the models were developed using actual usage and weather data from June 2004 through early

August 2005. Usage data were obtained from monthly Revenue Ledger reports.

Customers and Loads by Segment The Vectren North service territory has about 570,000 customers distributed into the four basic rate classes as

presented in Table 2. Monthly gas sales by rate class are presented in Figure 3.

Table 2. Vectren North Customers and Usage (unadjusted for weather) by Rate Schedule

Source: Vectren North Monthly Revenue Ledger Reports, June 2004-May2005

It is evident in Table 2 and Figure 3 that most of Vectren gas sales are to residential customers and that the

industrial customers use only a small fraction of the annual sales.


Vectren DSMAction Plan: Final Report LU. Market Assessment

Figure 3. Total Vectren North Gas Sales by Rate Class

End-Use Energy-The monthly energy use of Figure 3 was decomposed into its constituent end-uses by

developing simple models for the end-use energy in each sector. When the monthly energy use of Figure 3 is

decomposed into end-uses, it appears as in Figure 4.

160

g '* 0 E IZ0 3- loo

s r 80 2 s r 60

I- - .2 - . 40 - 20

0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month

WH I3 Process El Cooking CDly rn SH

Figure 4. Total Vectren North Gas Sales by Energy End-Use

The monthly information shown in Figure 4 is aggregated into annual end-use estimates in Table 3. The

importance of estimating the utility energy sales by end-uses is that end-use energy bears a workable physical

relationship to a wide variety of engineering parameters. For example, the magnitude of space heat usage bears

a direct relationship to the overall furnace efficiency, while the magnitude of water heat usage bears a direct

relationship to gallon per day hot water use and hot water set temperature. Thus, the exercise of mapping the

various sector gas end-uses is also creating an end-use model of the utility. It is apparent in Table 3 and Figure

4 that about two-thirds of the gas use is for space heat and about one-fourth of the use is for water heat. Most of

these end-uses are in the residential sector.


Vectren DSMAction Plan: Final Report HI. Market Assessment

Table 3. Vectren North Total Annual Gas Use by End-Use

Source: Our analysis of monthly usage

A visual perspective of the distribution of end-uses in Table 3 is provided in Figure 5, an end-use map of

Vectren North gas and transportation sales. Figure 5 is proportioned so that each square equally represents

about 1.7 million therms per year; thus the larger the visual area, the larger the usage. The horizontal axis in

Figure 5 indicates the market sector, Residential (R), CommerciaVIndustrial (C), and Transport (T), and the

vertical axis represents the fraction of the sector in each end-use. In this figure, red is space heat, blue is water

heat, green is cooking or drying, and yellow is process energy.

1

0.9

0.8

0.7 3

0.6 b C

0.5 O E 0.4 2 0.3

0 2

0.1 R R R R R R R R R R R R R R R R C C C C C C C T T T

Sector

Figure 5. Vectren North End-Use Map

In Figure 5, note the figure is stratified from left to right in terms of average customer size. This stratification

constrains the small-scale residential and commercial users, with small forced-air furnaces and residential-scale

gas water heaters, to occupy the leftmost two-thirds of the graph. The rightmost portion of the graph shows the

end-uses of the larger gas customers, whose space and water heat is via boilers. Thus the graph also generally

indicates the type of gas-using appliances matched to the end-uses.

It is quite apparent in Figure 5 that residential-scale space heat and hot water heat are the dominant end-uses,

while space and water heat are the most significant end-uses for the commercial sector and parts of the industrial

sector as well.

Prepared by Forefront Economics and H. Gil Peach &Associates Page 7

Vectren DSM Action Plan: Final Report m. Market Assessment

Residential

Service Area Description The market assessment presented in this section begins with a high-level view of residential housing in the

Vectren North service area, followed by a detailed analysis of residential gas loads. As shown in Table 4 below,

there were nearly 870,000 occupied housing units in thk Vectren service area. Over half of households use

utility gas (Vectren) for their primary space heating fuel. Electricity is the second most common heating fuel

and is used in one of four homes. Nearly three of every four homes are owner occupied.

Table 4. Housing by Heating Fuel and Tenancy

Source: 2000 Census Data for Counties in Vectren North Service Area

Residential construction estimated from housing permit data from the Vectren service area is shown in Figure 6.

Single family construction trended higher from 1994 through 1999 before leveling off at around 14,000 units a

year. In 2004 single family construction hit a peak of nearly 14,700 units. After peaking at 3,600 living units in

1997, multi-family construction has been remarkably flat at around 3,000 units a year.

16000

14000

12000

10000

8000

6000

4000

2000

0

1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

-t Single F a d y -+- Muhi-Fady

Figure 6. Residential Housing Units Permitted for Construction, Vectren North Service Area


Vectren DSMAction Plan: Final Report LLI; Market Assessment

Data shown in Figure 6 are based on monthly permit data lagged to approximate the timing of construction and

better align temporally with actual gas service installations. In Table 5 below, total construction and gas service

installations are presented. Gas is estimated to serve over 80 percent of single family construction and

approximately one-third of new multi-family construction. Market share in multi-family can swing significantly

from year-to-year (as evidenced in 200 1) due to the influence of a few large projects, a shift in the proportion of

low- and high-end multi-family units constructed, andfor differences in the timing of estimated construction and

gas service installation.

Table 5. New Gas Services, Construction, and Gas Market Share

Source: Connects from Vectren North CIS Data. Units Built Estimated from Housing Permit Data for Vectren North Service Area.

Customer Description A market segmentation strategy was adopted to describe the residential customer class in greater detail. The

segments were also selected to better describe cost effective DSM opportunities which can vary significantly by

type of housing and vintage of construction.

Table 6. Number of Residential Customers by Segment

Source: Vectren North CIS Data

Existing Construction New Construction Total

Residential customers are segmented by vintage of construction and type of housing. There are typically many

important differences between older and newer homes that have large impacts on energy use and conservation

potential. Differences in the thermal integrity of the building shell and appliance penetration rates, for example,

can lead to large differences in annual usage between older and newer homes. Existing construction is defined

as all homes with gas service installed prior to 2001. New construction consists of all homes connected to gas in

2001 and after.

Single Family


Multi-Family Total

(thousands) 446.5 36.3

482.8

3 1.2 5.2

36.4

477.7 41.5

519.2

Vectren DSMAction Plan: Final Report LTI. Market Assessment

We looked for a clear line of demarcation in selecting a vintage for defining the existing housing stock from

new construction. Residential building codes were not found to provide such a line, as the last significant code

change for energy usage was found to be a move from 65 to 80 percent AFUE furnaces nearly 15 years ago. We

chose to define new construction as homes built in 2001 and after to allow for current usage data for a full three

years of consumption (2001-2004). At the request of the Advisory Board, we ran usage analysis by year of

construction to see if any clear pattern presented in use per customer by vintage. Results are shown in the table

below.

Table 7. Average Weather-Normalized Usage, SF Residential by Meter Set Year

Source: Our Analysis of Monthly Vectren North CIS Usage Data

Meter data from each home was included from August 15,2004, through July 15,2005, to estimate the weather

normalized models used in the analysis. Our sample of existing homes ended with 1998 to provide greater

separation between construction practices of new and existing homes. However, as a review of the results

indicates, there is no clear trend in average usage for the years examined and, therefore, no clear logic for using

one year over another as the definition for new construction in this analysis.

Annual Therms

955

896

83 7

982

942

939

Meter Set Year

1996

1997

1998

200 1

2002

2003

The number of living units per building, single family and multi-family, also enter into the segmentation

approach. Single and multi-family units exhibit many differences that impact gas consumption and conservation

potential. These differences include size of unit, appliance penetration, building shell integrity and lifestyle

attributes. The housing type was determined from the unit number portion of the service address. Premises with

non-missing unit numbers were classified as multi-family while units with no unit number were classified as

single family buildings.'

Homes in Sample

46

98

76

470

264

265

A large share (86 percent) of residential customers fell into the single family existing segment. Single family

new construction makes up about 7 percent of all customers. Multi-family is a relatively small segment of the

residential class accounting for a total of 7 percent of all customers, mostly (6 percent) in the existing vintage.

Multi-family new construction only accounts for one percent of all residential customers.

Frequency tables of unit number were examined for entries unrelated to unit number such as "NA", "None", or "BOD (beware of dog) that could bias the classification. These sorts of entries were not found in the data.


Vectren DSMAction Plan: Final Report U% Market Assessment

Gas Usage Analysis The number of customers by segment is shown in Table 8. Customer counts represent the mean monthly

population levels from mid-2004 through mid-2005.

Table 8. Number of Residential Customers by Segment

Source: Vectren North CIS Records, June 2004-May 2005

Segment Existing Single Family Existing Multi-Family New Construction Single Family New Construction Multi-Family Total

In this report the partition of residential customers by housing age and type is estimated from service installation

records. The current estimates show that most of the housing stock is single family and that most of the single

family is existing stock.

Number of Customers 446,546 31,154 36,347

5,192 519,239

Our analysis of customer usage also took advantage of a residential survey Vectren fielded in July. A report was

issued by the market research firm dated August 22,2005, describing the survey results, including appliance

installation rates. Since the results in the report were not weighted to the Vectren North service area, we also

asked for and were provided SAS datasets with survey results. Using premise and respondent counts within

each of the four quadrants surveyed, weights were calculated to allow the results from the stratified sample

design to be expressed for the Vectren North area. The results are presented in the table below.

Table 9. Residential Survey Weights by Survey Quadrant

Source: Vectren North Residential Survey Results, August 2005

Appliance installation rates calculated from the weighted survey results are shown in Table 10. The number of

respondents was sufficient for results in all segments except new multi-family. Customer survey data provide

more current and detailed data on end-uses than the Census data presented in Table 4.

Prepared by Forefront Economics and H. Gil Peach & Associates Page I1

Vectren DSMActwn Plan: Final Report LU. Market Assessment

Table 10. Appliance and End-Use Installation Rates from Residential Survey

Source: Vectren North Residential S w e y Results, August 2005

Gas use for space heating is present in over 90 percent of single family homes with little difference between the

existing and new construction segments. Gas water heating is also highly prevalent in single family housing

with gas, but more so in new construction than existing. Water heating with gas is found in about 50 percent of

multi-family gas customers. Gas cooking and hearth products (logs, fireplaces, inserts, and stoves) are more

popular in new construction.

Figure 7 shows that most of the residential gas usage is associated with the existing single family stock and that

there is a conspicuous winter peak usage.

100

90

" s 80 9 5 2~ 70

zg 60 "8 50 g! g% 40 gE 30 ulc a t 20

10


Month

I m SF Eisting MF Existing SF New w MF New I Figure 7. Residential Gas Usage by Housing Type

Prepared by Forefront Economics and H. GiC Peach & Associates Page 12

Vectren DSMAction Plan: Final Report H1. Market Assessment

Overall the residential sector is projected to use about 460 million therms per year, about 63 percent of total

utility gas sales. This projection is reported for a "normal" weather year with the same average numbers of

customers as May 2004 through May 2005. The final result of the end-use disaggregation is presented in Table

11.

Table 11. Residential Sector Annual Gas Usage by End-Use

Source: Our analysis of monthly usage data, Vectren North CIS

Table 11 reports the annual gas usage, therms per year, for the most significant of the residential gas usage

categories. This annual gas usage is distributed on a month-to-month basis as shown in Figure 8.

100 - 90 c e 80 zg 70 9 2 60 s g ;; l c 8 : 30

E 20 - 10 0 Jan Feb Mar Apr May Jun Jtd A q Sep Oct Nov Dec

Month a Cooking CDry Other Wn rn SH

Figure 8. Month-to-Month Normal Gas Usage by End-Use

It is readily apparent in both Table 11 and Figure 8 that the predominant residential gas end-uses are space heat

(SH) and water heat (WH). It is also interesting to note the apparent minor role played by the end-uses:

cooking, clothes dryer, and other. These end-uses are quite small relative to space heat and hot water. For

reference and illustration, these monthly end-use results are presented by month in Table 12.


Vectren DSM Action Plan: Final Report LTI. Market Assessment

Table 12. Residential Sector Monthly Usage by End-Use


Average use per customer for each of the four segments is summarized in the table below. It is interesting to

note that existing and new homes have the same total annual load. In both cases new homes have slightly lower

space heat usage that is offset by higher base usage. This finding is consistent with the appliance installation

rates in Table 11 which show higher installation rates of water heating in new construction than existing.

Additional water heat load appears to be the primary reason for the greater base load in new construction.

Table 13. Average Use per Residential Customer (therms)


All projections of usage to a normal year require the use of some sort of model, however simple. The usage

model for the residential sector consists of an assembly of submodels for each end-use in terms of monthly

temperature. In the course of this analysis, the model underlying this projection was trued to the actual recorded

usages and temperatures for the test year May 2004 through May 2005. This true-up provides a reality check on

the total of all the end-uses estimated by the model, though it does not provide a check on any particular end-

use. Figure 9 shows that modeled and actual total residential gas usage agree well.

Figure 9 also shows that the total residential usage bears a close correlation to the average monthly temperature.

Physically, this is identically the same relationship long noted between heating degree days and gas usage. But


Vectren DSMAction Plan: Final Report I . . Market Assessment

in this model, the Mean Month Temperature is used instead of deg days as the temperature variable because it is

an absolute reference, not indirect as in a degree day, and because it is more compatible with the structure of the

engineering models that are the submodels of the overall residential sector end-use model.

180

160 a3 p 140

3 a 120

0 0 10 20 30 40 50 60 70 80

Mean Month Temperature (Fo)

+ Data + Model Total

Figure 9. Actual and Modeled Total Residential Gas Usage

Note in Figure 9 the steep, but predictable, change in usage with temperature. This slope provides a check point

for the cumulative effect of assumptions regarding heat loss. Usage model parameters (such as, furnace

efficiency, distribution efficiency, and shell thermal losses) all act together to determine the heat loss at any

given temperature. The cumulative effect of even small errors in these parameters can become a significant over

or under statement of gas use. Hence there is a need to calibrate the model against a real world check point.

This temperature slope of gas use versus temperature is one of the two real world check points for each of the

models.

In Figure 9, the low usage values, base load, occurring at the highest temperatures, is the other check point.

These low consumption and high temperature periods represent usages that are not space heat. Taken together,

these check points have significant resolving power, sufficient to separate out the base load from the space heat

end-use and to provide a close limit on the base load and the total monthly residential gas usage.

A full discussion of the sector usage model and the end-use submodels, as well as test year and normal year, is

presented in the Methodology section of the Appendix.

Commercial

Service Area Description For this analysis the commercial population has been examined from two separate perspectives. First, the

population has been classified by the 2005 commercial rates. This classification facilitates reconciliation of

results with the general ledger records, but it does not align clearly with commercial business types and gas


Vectren DSMAction Plan: Final Report LTI. Market Assessment

usage. Second, the population is classified into ten primary business types. This latter method yields more

insight into commercial gas end-uses by business type.

As with residential, business attributes in the Vectren service area are first described then followed by a

discussion of commercial customers and usage. County-specific data were also used to better relate secondary

data to the Vectren service area. Table 14 shows the number of businesses and employment by commercial

segments.

Table 14. Number of Businesses and Employment, Vectren North Service Area

Source: Woods and Poole 2005 State Profile and STATS Indiana Website

Number of Employment** Percent Employees per Businesses* Business

There are an estimated 53,000 businesses in the Vectren service area employing a total of nearly one million

Construction Manufacturing Wholesale Trade Retail Trade Transportation, Warehousing Information* Professional, Technical Services Health Care, Social Services Arts, Entertainment, Recreation Accommodation, Food Services Other Private (not Above) Total

workers, an average of just over 18 employees per busines2 About 19 percent of total employment is in

manufacturing. Retail trade is the largest non-manufacturing sector accounting for 16 percent of all

*Averaged 2001 and 2002 totals from Woods & Poole 2005 State Profile **2003 totals from STATS Indiana website (www.stats.indiana.edu)

6,129 3,007 2,675 9,004 1,670

772

4,065 4,764

807 4,29 1

15,636 52,817

employment.

As shown in Figure 10; non-manufacturing employment has increased steadily over the last 34 years,

71,774 182,43 8 30,220

153,740 38,825 13,893

41,852 ppppp

90,034 19,264 82,660

23 8,62 1 963,321

increasing an average of 1.5 percent annually. Manufacturing employment, on the other hand, has been mostly

flat with periods of upward and downward trend. Overall, manufacturing employment has declined from a high

7.5 18.9 3.1

16.0 4.0 1.4

4.3 9.3 2.0 8.6

24.8 100.0

of 246,000 in 1969 to an estimated 209,000 in 2004, a drop of 0.5 percent annually since 1969.

11.7 60.7 11.3 17.1 23.2 18.0

10.3 18.9 23.9 19.3 15.3 18.2

2 Business counts do not include proprietors and government workers. Employment estimates are likely to be understated by an unknown amount due to non-disclosure requirements of the U.S. Bureau of Economic Analysis. These requirements protect businesses that could be identified in the data when a small number of f m s operate in a reporting jurisdiction. Also, certain businesses are exempt from reporting adding to the potential for under reporting.

Employment data in this chart are presented to show long-term trend. Due to differences in the reporting methods of data sources, the data may not agree with other sources presented in this paper.


Vectren DSM Action Plan: Final Re~ort I . . Market Assessment

1 m

WX)

800

7W - '9

E €-xl z e t

300

2W

1W

0 C 8 - O O C D - O D C O , c r . , , , m m , m m g ~ g $ ; z s I E L I L " L " " L , , , - , B , B

- Manufacturing - Primte NorrManufactwing

Figure 10. Trends in Employment, Vectren North Service Area

Commercial Building Stock Descriptive information on the commercial building stock in the Vectren North service area is useful for better

understanding the nature of the DSM opportunity. We considered the likely distribution of commercial building

stock by building type, age and square feet in our assessment of the applicability of conservation technologies.

Unfortunately, we were unable to identi8 internal or secondary sources to describe the specific service area.

We turned instead to the 1999 Commercial Buildings Energy Consumption Survey (CBECS), a national survey

of the building characteristics and energy end-uses in over 5000 commercial buildings. Although the CBECS

sample includes buildings from all 50 states and the District of Columbia, results at the Census division level are

the lowest geographic detail available.

The Census division that includes Indiana is the East North Central Division which also contains Ohio,

Michigan, Illinois and Wisconsin. We felt that including the entire division as a proxy for central and southern

Indiana was not appropriate. CBECS includes a climate zone field that we used to narrow the geographic space

to an area that is more likely to be representative. The CBECS survey includes HDD zones of less than 4000,

4000 to 5499, 5500 to 7000, and greater than 7000. Choice of an appropriate climate zone was complicated by

the fact that Indianapolis, with just over 5525 HDD a year, lies in the same climate zone as northern Indiana and

Chicago. We used a climate zone screen of 4000 to 5499 which includes southern Indiana and the Evansville

(HDD=4617) area. However, this left an insufficient number of buildings for analysis. To augment the sample

we included two adjacent Census divisions East South Central (Kentucky, Tennessee, Alabama and Mississippi)

and West North Central (Missouri, Kansas, Iowa, Nebraska, the Dakotas and Minnesota), keeping the climate

screen of 4000 to 5499 HDD. This resulted in the inclusion of commercial buildings markets such as Saint

Louis (HDD=4758), Louisville (HDD=4352), and Kansas City (HDD=5250). The climate zone screen

eliminated northern and southern markets such as Fort Wayne (HDD=6205) and Nashville (HDD=3677).


Vectren DSMAction Plan: Final Report LlI. Market Assessment

This filtering provided a sample of 261 commercial buildings with gas usage. Of these, there were sufficient

numbers (at least 20 buildings) in five of the 20 building activity types to report the data. Commercial building

stock data are reported in this report for the five building activity types and for all 26 1 gas buildings. The

average year built and distribution of buildings by vintage is shown in the table below.

Table 15. Age of Gas Buildings - Customized Regional Area of CBECS Survey

Source: Commercial Buildings Energy Consumption Survey, 1999

The average age of gas buildings is over 40 years and more than two-thirds of the stock was built before 1980.

Retail and service buildings tend to be older than most building types. Warehouses are the newest building

stock, with nearly half of all warehouses using gas constructed in 1980 and after.

Data on square footage of commercial buildings using gas is shown in Table 16. Size of commercial buildings

varies significantly by building type. The average size of commercial buildings is around 160,000 square feet.

Of the five building types shown in the table, services are by far the smallest building type followed by retail

buildings. Offices and warehouses are near the average size for all buildings. The distribution of building size

is typically heavily skewed with most buildings in the smaller sizes. For example, even though the average

office is over 160,000 square feet, over half of all office buildings are smaller than 50,000 square feet. In other

words, the average is much larger than the median of the distribution due to very large office buildings at the tail

of the distribution.

--

Prepared by Forefront Economics and H. Gil Peach & Associates Page I8

Vectren DSMAction Plan: Final Report ID. Market Assessment

Table 16. Size of Gas Buildings - Customized Regional Area of CBECS Survey


Activity

Ofice/F'rofessional Warehouse (Nonrefiig.)

Education

Retail (excl. Mall)

Service

All Building

The incidence of gas end-use by building type is shown in Table 17 below. It is no surprise that gas usage for

space and water heating have the highest saturation levels, averaging 80 and 70 percent respectively. The use of

gas for cooking and electricity generation is present in 29 and 5 percent of buildings, respectively. It is

interesting to note the presence of cooking loads in all building types, including nearly a quarter of all office

buildings using gas.

Table 17. Natural Gas End-Uses in Commercial Buildings - Customized Regional Area of CBECS Survey

2 a '

38

29

41

20

23

151


Customer Description As with residential customers, a segmentation strategy to group customers into segments with similar load and

conservation opportunities is desirable. The segmentation approach used for non-residential customers is shown

in Table 1 8.4

Average Square

164,151

164,897

67,104

44,350

15,826

163,490

4 Note that the data in Table 18 are NAICS-based segments and, hence, will not correspond to the data in Table 2 which shows usage by segments based on rate schedules.


Square Footage Category (percent)

1,000 or

Lns

0.0

0.0

0.0

0.0

0.0

0.0

1,001- 5,000

13.2

3.4

2.4

15.0

34.8

9.2

5,001- 10,OO

10.5

3.4

2.4

10.0

30.4

14.0

10,001- 25,000

21.1

13.8

9.8

30.0

13.0

25.7

25,001- 50,000

7.9

13.8

43.9

25.0

17.4

13.7

50,001- 100,000

7.9

20.7

26.8

5.0

4.3

10.3

100,001- 200,000

5.3

17.2

7.3

10.0

0.0

7.5

200,001- 500,000

28.9

20.7

7.3

5.0

0.0

6.2

500,001- 1

millon

2.6

6.9

0.0

0.0

0.0

11.1

Over 1

million ,

2.6

0.0

0.0

0.0

0.0

2.4

Vectren DSMAction Plan: Final Report LIl; Market Assessment

Table 18. Vectren North Non-Residential Premises and Loads by NAICS-Based Segments

The segmentation of customer data was based on an extract from the Vectren customer information system

(CIS) that included North American Industrial Classification System (NAICS) codes. This project benefited

2005 Load

from having current NAICS codes since Vectren recently paid a third party to append NAICS codes to their CIS

records. Customers were then segmented into the groups shown in Table 18 based on the codes.

in Millions of Therms, Actual Usage (unadjusted for weather), August 2004-July

Over three-fourths (78 percent) of all non-residential customers are classified as commercial, based on NAICS.

Segment

These commercial customers account for just over half of all non-residential loads. In terms of annual loads,

offices and schools are the largest commercial segments. Vectren has a large number of retail customers, but

Premises

they only account for four percent of total non-residential loads.

Commercial

In the meantime, the preliminary results discussed in the commercial and the industrial sections are based on

Percent

broadly defined rate-based definitions of segment. These broadly based definitions are valuable for assessing

Grocery

Hospitals

Lodging

Off'ice

Other Health

Restaurants

Retail

Schools

Wholesale and Warehouse

Unclassified and Other

Total Commercial

sector loads and disaggregation by major end-uses.

Load

2

1

1

19

6

6

11

3

6

22

77

947

335

3 13

9,149

2,624

2,775

5,330

1,629

2,845

10,539

36,486

Gas Usage Analvsis The commercial sector usage has been analyzed in terms of the 2005 commercial rates. In the commercial

Percent

Other Nun-Residential

sector, the monthly customer population varies slightly from month to month with an average monthly

3.3

13.3

3.1

45.8

14.4

16.7

19 31.1

22.4

60.5

230

population of about 50,000. The commercial customers are composed of four basic rate categories: NCM1,

1

3

1

11

3

4

4

7

5

14

53

Ag., Mining, Util., and Const.

Manufacturing

Total Other Non-Residential

Total Non-Residential


Source: Vectren North CIS Data

17

6

23

100

8,239

2,85 1

11,090

47,576

53 -2

149.5

202.7

432.3

12

35

47

100

Vectren DSMAction Plan: Final Report hX Market Assessment

NCM;?, NCM3 and NC40. The mean monthly populations of these categories are shown in Table 19 along with

the estimated number by major end-use.

Table 19. Assumed Commercial Customer Distribution

Source: Our analysis of monthly usage, Vectren North Revenue Ledger Reports

Note in Figure 11 that the rate NC40, which has the largest commercial customers, also has so few participants

that the aggregate usage is relatively small. The rate NCMl with about 35,000 participants appears to be space

heat dominated. The average heat loss rate is that of a large residence. This group likely consists of small office

and retail, where the principal gas use is space heating.

The rate NCM2, with about 1 1,000 participants, has the largest aggregate usage in the commercial sector. It is

characterized by an average heat loss rate equivalent to about five to ten houses and higher hot water use. This

rate is probably populated by medium-scale office and retail and by smaller restaurants.

40 1

I Jan Feb Mar Apr May Jun Jul Aug S e p Oct Nov D e c I

I Month I NCMl . NCM2 NCM3 k?d NC40 I

Figure 11. Commercial Monthly Gas Usage by Rate Schedule

The rate NCM3 has only about 2500 participants and these are characterized by an average heat loss rate

equivalent of 25 houses with significant hot water use as well. This rate is probably populated with schools,

smaller lodging, medium-sized restaurants and laundries.

The rate NC40 has only about 95 participants. The average participant is a large scale operation with an average

heat loss rate equivalent to about 60 to 100 houses also with significant hot water use. This rate is probably

populated with larger offices and schools and retail, large restaurants, smaller hospitals, and medium-scale

lodging.


Vectren DSM Action Plan: Final Report IIA Market Assessment

Notably, each of these rate categories has explicit billing information for the 2005 portion of the test year and

the model for the stock in that rate category has been separately reconciled to the usage record. Overall the

commercial sector, defined by rate schedule, uses about 2 13 million therms per year, about 29 percent of total

utility gas sales. This projection is reported for a ccnormal" year which has the same level of customers as May

2004 through May 2005, but has the 30-year average temperatures instead of the actual temperatures. The final

result of the end-use disaggregation is presented in Table 20.

Table 20. Commercial Sector Annual Gas Usage by End-Use

Percent of Tota


Table 20 reports the annual gas usage, therms per year, for the most significant of the commercial gas rates.

This annual gas usage is distributed on a month-to-month basis as shown in Figure 12.

45 - 40 + 35 E

2 30 t 3

c 25 g 0 m 1 20 5 8 r 15

$0 - 3 5 I-


Month

Cooking CDly rn Other WH SH

Figure 12. Commercial Monthly Normalized Gas Usage by End-Use

It is readily apparent in both Table 20 and Figure 12 that the predominant commercial gas end-uses are space

heat (SH) and water heat (WH). For reference and illustration, these monthly end-use results are presented in

Table 2 1.


Vectren DSMActwn Plan: Final Report LU Market Assessment

Table 21. Commercial Sector Monthly Usage by End-Use

Commercial Sector End-Uses (thermslmonth) Cooking I Clothes Dryer I Other I Water Heat I Furnace

1 arch April

May June

January February

July August September October


1,077,188 1,042,440 1,077,188 1,042,440

November December Total Annual

In addition to the rate schedule based analysis discussed above, loads were also modeled by the segmentation

strategy discussed earlier. The results augment the rate schedule based analysis and provide average usage by

segment, useful for assessing DSM opportunities by building prototype.

1,077,188 972,944

1,077,188 1,077,188 1,042,440 1,077,188

Weather normalized use per customer is shown in Table 22 for base, heat and total load. Base loads are those

loads which are not correlated with temperature including water heat, cooking and process 10ads.~ Heating load

refers to the temperature dependent portion of total load, typically a gas furnace or other space heating device.

16,591 16,056 16,591 16,056

1,042,440 1,077,188

12,683,019

Typical of commercial loads, the average use per customer shown in Table 22 reveal large spreads in annual

usage between segments. On the low end, grocery and retail customers use just under 4,000 therms a year.

Hospitals are by far the commercial segment with the highest use per customer, averaging 10 times the load of

typical retail and grocery customers.

16,59 1 14,985

16,591 16,591 16,056 16,591

Sector Total

It is also interesting to note the heating load as a percentage of total load. Segments with larger heat related

loads tend to be comprised a smaller commercial customers with load patterns typical of residential customers.

Retail, grocery, and schools all have high percentages (around 70) of heat load. Of these, schools are the only

ones that do not appear to fit the category of small commercial.

783,551 758,275 783,551 758,275

213.139.888

16,056 16,591

195,343

While restaurants and buildings with food preparation end-uses do not appear to have large total gas

consumption, cooking end-uses represent the third highest usage category, as shown earlier in Table 20, at 6

percent. Since there are many opportunities for increasing the energy efficiency of food preparation equipment,

783,551 707,723

783,551 783,551 758,275 783,551

5 Table 22 is used for identifying segments with high or low average usage per customer. Use Table 18 to compare total loads between segments.

5,365,270 4,861,304 4,633,637 4,158,673

758,275 783,551

9,225,676


5,487,635 5,001,111

24,137,057 13,043,037 2,767,306

0 4,090,5 18 4,034,106 3,994,406 4,412,557

3 1,72 1,524 28,836,848

0 0 0 0

4,705,384 5,309,873

56,054,474

11,158,491 23,3 17,114

134,981,377

Vectren DSMAction Plan: Final Re~ort m. Market Assessment

restaurants and other hospitality establishments have been target markets for DSM programs in several

jurisdictions, in spite of their relatively lower total gas usage profile.

Table 22. Average Use per Commercial Customer by Segment (therms)

Segment Grocery

Hospitals Lodging

Off~ce

Other Health

Restaurants

Retail

Schools

Unclassified and Other 1 2,438 1 3,747 1 6,185 1 60.6 Source: Our Analysis of Usage Data, Vectren North CIS Usage

2,827

4.481

Wholesale and Warehouse

All projections of usage to a normal year require the use of a usage model, however simple. In the commercial

Base

1,150 26,488

6,361

2.451

1,026

6,073

3,595 1 4,784 1 8,379 1 57.1

sector, there is a usage model for each of the segments. Each usage model consists of an assembly of submodels

Total

3,855

40,870

10,531

5.404

Heat

2,705

14,382

4,170

2.953

3,041

2.062

for each end-use in terms of monthly temperature. In the course of this analysis, each usage model underlying

Percent Heat

70.2

35.2

39.6

54.6

2,880

14,321

this projection was trued to the actual recorded temperature and usages for that segment. This true-up provides

5,868

6.543

a reality check on the total of all the end-uses estimated by the model, though it does not provide a check on any

51.8

31.5

3,906

20,394

particular end-use. Figure 13 shows that modeled and actual total commercial gas usage agree well.

73.7

70.2

40

Q 35 & m .2 30 3 =

25 m (3 20

3 5 15 3 € L L rn at 10 rnx a t 5

0 0 10 20 30 40 50 60 70 80

Mean Month Temperature (Fa)

+ Data -& Model Total -

Figure 13. Actual and Modeled Total Commercial Gas Usage


Vectren DSMAction Plan: Final Report L17. Market Assessment

Figure 13 also shows that the total commercial gas usage bears a close correlation to the average monthly

temperature. Physically, this is identically the same relationship long noted between heating degree days and

gas usage. But in this model, the Mean Month Temperature is used instead of degree days as the temperature

variable because it is an absolute reference, not indirect as in a degree day, and because it is more compatible

with the structure of the engineering models that are the submodels of the overall commercial sector end-use

model.

Note in Figure 13 the steep, but predictable, change in usage with temperature. This slope provides a check

point for the cumulative effect of assumptions regarding heat loss, furnace efficiency, distribution efficiency,

shell thermal loss and infiltration.

In Figure 13, the low usage values occurring at the highest temperatures, are check points for the sum of usages

that are not space heat. Taken together, these check points and the temperature slope have significant resolving

power, sufficient to separate out the space heat end-use and to provide a close limit on the base load and the total

monthly commercial gas usage.

A full discussion of the sector usage model and the end-use submodels, as well as test year and normal year, is


Industrial

For this analysis the industrial population has been classified by the 2005 industrial rates. This classification

facilitates reconciliation of results with the general ledger records. This definition of the industrial sector

includes the industrial commodity gas rates NIN3 and NI40, as well as, the gas transportation customers. While

there are some industrial gas commodity sales, most of the industrial gas use is by transportation customers.

In the industrial sector, the monthly industrial customer population for Vectren North varies slightly from month

to month with an average monthly population of about 900 customers. In this sector, the industrial customers

are assumed to be composed of four basic rate categories: NIN3, NI40, T600x and T700x. For this analysis, the

mean monthly populations of these categories in the first six months of 2005 are shown in Table 23.

Table 23. Assumed Industrial Customer Distribution

End-Use Total Furnace Water Heat

Source: Our analysis of monthly usage, Vectren North Revenue Ledger Reports I


m 3 136 136 136

Cooking Clothes Drver

NI40 48 48 48

27 27

T600x 358 358 358

T700x 3 70 370 3 70

10 10

72 72

74 74

Vectren DSM Action Plan: Final Report LtI Market Assessment

Note in Figure 14 that the rate T700x, which has the largest industrial customers, has the majority of the gas

usage. While this rate has about 370 customers, most of the gas use is by less than 50 of the largest

transportation customers.

Figure 14. Industrial Monthly Gas Usage by Rate Schedule

The rate T600x with about 350 customers appears to be space heat dominated. The average heat loss rate is that

of about 25 residences. This group likely consists of larger buildings including lodging.

The rate NI40, with about 50 participants, has a high average heat loss rate, equivalent to about 50 to 75

residential homes. It also appears to have low usage in late December as if the facility was shut down for part of

the month. This rate is probably populated by education institutions.

The rate NIN3 has only about 135 participants and these are characterized by an average heat loss rate

equivalent of 25 houses with significant hot water use as well. This rate is probably populated with schools,

smaller hotels lodging, medium sized restaurants and laundries.

Notably, each of these rate categories has explicit billing information for the first six months of 2005, the test

year, and the model for the average customer in that rate category has been separately reconciled to the usage

record. Overall the industrial sector is projected to use about 64 million therms per year, about 9 percent of total

utility gas sales. This projection is reported for a "normal" year which has the same level of customers as May

2004 through May 2005, but has the 30-year average temperatures instead of the actual temperatures. The final

result of the end-use disaggregation is presented in Table 24.


Vectren DSMAction Plan: Final Re~ort m. Markef Assessment

Table 24. Industrial Sector Annual Gas Usage by End-Use 1 8 6

End-Use Furnace WaterISteam Process

Total 1 1 64,117,878 1 100.0 I Source: Our analysis of monthly usage, Vectren North Revenue Ledger Reports

ThermsNear 9.291.196

clothes Dryer Cooking.

Table 24 reports the annual gas usage, therms per year, for the most significant of the industrial gas rates. This

Percent of Total 14.5

15,928,753 37,165,425

annual gas usage is distributed on a month-to-month basis as is shown in Figure 15. It is readily apparent in

58.0 724,708

1.007.795

both Table 24 and Figure 15 that the predominant industrial gas end-uses are process related.

1.1 1.6

Figure 15. Industrial Monthly Normalized Gas Usage by End-Use

This process energy is quite specific and diverse and actually includes water heating, steam, drying, etc., but for

the purposes of this analysis all these process-coordinated uses will be classified into two end-uses: direct

process energy and indirect process energy via boilers. The partition into direct and indirect process energy is

based on assumptions from the U.S. Department of Energy industrial gas usage statistics.

For reference and illustration, the monthly end-use results are presented in Table 25.


Vectren DSMActwn Plan: Final Report ILT. Market Assessment

Table 25. Industrial Sector Monthly Usage by End-Use h

Industrial Sector End-Uses (thermslmonth) Cooking 1 Clothes Dryer 1 Process 1 Water/Steam 1 Furnace


Also as perspective, the monthly gas end-uses are presented for a single average building of each of the four

commercial rates in Figure 16 through Figure 19.

200

180

5: 160

140 m

120 ; 100 t 0, 80

60 m 3 40

20


Month

I Cooking Ed CDry l3 Other . WH SH I Figure 16. Daily Gas End-Uses - Average Industrial NIN3 Customer

Note in Figure 16 that the average customer in rate NIN3 has a usage profile quite similar to customers in the

larger commercial rate categories, NCM3 and NC40.


Vectren DSMAction Plan: Final Report ILL Market Assessment

500

I Jan Feb Mar Apr May Jun Jul Aug S e p Oct Nov D e c l

I Month I Cooking fa CDry Other WH SH

Figure 17. Daily Gas End-Uses - Average Industrial M40 Customer

The average site in rate NI40 has a high space heat signature as if it were a large complex of buildings. Also

there is an anomalously low space heat usage for December through January suggesting that facilities in this rate

are not fully used in late December. It walks and talks like a school.

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month

w Cooking s CDry E9 Other WH w SH

Figure 18. Daily Gas End-Uses - Average Transportation T600x Customer

Average transportation customers in rate T600x are moderately sized and are characterized by a relatively high

base load. The total base load for this rate is reasonably well known, but the partition of the base load into its

sub end-uses is based on assumptions. It is probable that the end-use mix for the 350 customers in this rate is

quite diverse.


Vectren DSMAction Plan: Final Report Market Assessment

450

400

350 m e 300 1 250 e g 200 & 150 m ," 100

50


Month

Cooking W CDry IZl Other rn WH rn SH

Figure 19. Daily Gas End-Uses - Average Transportation T700x Customer

Transportation customers in rate T700x vary significantly in size, with most of the energy going to a few of the

largest customers. Therefore, this average may not be physically representative of any of the customers in this

rate. Note in this figure the minimal space heat energy. This may not be space heat per se, but temperature

sensitivities of the larger processes.

Industrial customer usage was also analyzed by segment, defined by NAICS codes discussed earlier in this

section. These results are shown in Table 26 below.

Table 26. Average Use per Industrial Customer by Segment (therms)

Customers in the manufacturing segment have a much higher average usage than the Agricultural, Mining,

Segment

Ag., Mining, Util., and Const.

Manufacturing

Utilities, and Transportation (AMUT) segment. The two industrial segments are each characterized by low

space heating loads relative to the commercial segments discussed earlier, indicative of the higher process loads

found in this sector.

All projections of usage to a normal year require the use of a usage model, however simple. In the industrial

sector, there is a usage model for each of the four industrial rate categories. And each rate usage model consists

of an assembly of submodels for each end-use in terms of monthly temperature. In the course of this analysis,

each rate usage model underlying this projection was reconciled to the actual recorded usages for that rate and

temperatures for the 2005 portion of the test year, May 2004 through May 2005. The reconciliation provides a

reality check on the total of all the end-uses estimated by the model, though it does not provide a check on any

particular end-use. Figure 20 shows that modeled and actual total residential gas usage agree well.

Base

4,141

35,069


Heat

2,890

19,622

Total

7,031

54,691

Percent Heat

41.1

35.9

Vectren DSM Action Plan: Final Re~ort LI% Market Assessment

8

a p 4.0 6 3 E $ Z 5 0 8 4

pg 3 g e 2 m.c a t 1

0

0 10 2 0 . 30 40 50 60 70 80

Mean Month Temperature (FO)

+ Data -B- Model Total

Figure 20. Actual and Modeled Total Industrial Gas Usage

Figure 20 also shows that the total industrial gas usage bears a close correlation to the average monthly

temperature. Physically, this is identically the same relationship long noted between heating degree days and

gas usage. But in this model, the Mean Month Temperature is used instead of degree days as the temperature

variable because it is an absolute reference, not indirect as in a degree day, and because it is more compatible

with the structure of the engineering models that are the submodels of the overall industrial sector end-use

model.

Note in Figure 20 the steep, but predictable, change in usage with temperature. This slope provides a check

point for the cumulative effect of assumptions regarding heat loss, furnace efficiency and distribution efficiency.

In Figure 20, the low usage values, occurring at the highest temperatures, are check points for the sum of usages

that are not space heat. Taken together, these check points and temperature slope have significant resolving

power, sufficient to separate out the space heat end-use and to provide a close limit on the base load and the total

monthly industrial gas usage.

A full discussion of the sector usage model and the end-use submodels, as well as, test year and normal year, is


&?pared by Forefront Economics and IZ Gil Peach & Associates Page 31

Vectren DSMAction Plan: Final Report N. Conservation Potential

IV. CONSERVATION POTENTIAL

The work presented in this section is based on modeling results at the segment and rate schedule level of detail.

Consequently, the conservation potential assessment is based on the same segmentation scheme discussed

earlier in this report. This allows us to present results related to specific technologies.

Overview

This market assessment portion of the work will set forth the energy end-uses by customer segment: residential,

commercial, industrial, and transport. Model results by customer segment allow us to discuss the demand side

management "technical potential" proceeding fiom these segments and end-uses. The technical potential is

derived by assuming that all customers in each sector use the most efficient available gas technology. In this

analysis, technical potential is restricted to meeting existing gas end-uses with gas employed more efficiently.

As such, this estimate of technical potential has been restricted to measures that reduce the amount of gas

needed to meet end-use loads. The technical potential derived in this analysis does not contemplate fuel

switching.

However, there are real world efficiency effects that can increase gas usage without fuel switching, particularly

commercial and residential lighting efficiencies and gas-fired combined heat and power. In terms of technical

potential, these effects can be large. At this stage of the analysis, these effects have not been included in the

technical potential. Later stages of this report will discuss in detail the magnitude of savings and the cost

effectiveness of a full range of individual measures and packages of measures.

It became apparent fiom a review of the modeling work by segment and rate schedule described earlier that

many segments shared common load characteristics across four "planning elements". These planning elements

are small buildings, large buildings, process energy and restaurants. In the analysis of the end-use energy per

average customer at the sector level, it is apparent that about 70 percent of the total gas energy sales are to small

buildings, approximately residential scale. These gas sales are to a functionally homogenous array of gas forced

air furnaces and tank-style water heaters. These energy sales are designated here as the small buildings planning

element.

The sector level analysis also showed that about 23 percent of the total gas energy sales are to large buildings,

several to many times residential size. Most of these sales serve the space heat and water heat end-uses, but the

gas-using equipment and relevant controls are significantly different than that used in small buildings. This gas-

using equipment is a functionally homogenous array of boilers, whole building controls and energy management

practices. These energy sales are designated here as the large building planning element.



As a preliminary to demand side management planning, this end-use information, along with other

demographics, has been organized into logical demand side planning elements presented in Table 27. These

planning elements are derived by re-grouping various residential, commercial and industrial sector customers

around common building types and systems.

Table 27. Demand Side Planning Elements

'lanning

I Commercial cooking appliances, hot Restaurant 16 ' water fixtures 1 2.500

Small Buildings

Large Buildings Process Enerw

Source: Our analysis of Vectren North usage, sector and technology data

Target Energy Use (million thermsIyear)

The small building planning element is composed of the entire residential sector and rate NCM1, about 35,000

customers, of the commercial sector. This population is characterized by residential scale buildings heated by

small forced air furnaces and with water heat in residential scale gas hot water heaters.

515

176 47

The large building planning element is composed of the whole commercial sector excluding rate NCM1, the

industrial rates NIN3 and NI40, and the 350 smaller transportation customers. This population is characterized

by large scale buildings with a predominance of boilers and reasonably complex building controls.

Target Technology

The process energy planning element is composed of the largest 350 industrial transport customers. This

population is characterized by its diversity of uses and by the likelihood that the process energy cannot be

manipulated without special care for the process.

Applicable Population

Residential forced air heaters, duct work, residence thermal integrity Boilers, building controls, showerheads, building thermal integrity Diverse

The restaurant planning element is drawn fiom the commercial and industrial rate schedules. This population is

characterized by a wide range of specifically different gas energy intensive operations underlain by the use of

similar gas using equipment in fundamentally similar food preparation processes.

525,000

1 1,000 350

Technical Potential

The technical potential for each of the planning elements was derived by applying all the efficiency measures at

once so that interactions between efficiency measures and load reduction measures are properly accounted for.

In later stages of the program planning, various measures may be considered individually, but foi estimating the

total technical potential, all the measures were applied as a package.


Vectren DSM Action Plan: Final Report N. Conservation Potential

In this analysis technical potential for a planning element will be shown as in Figure 21. This figure illustrates

the derivation of technical potential; it is for the residential sector only, not the whole small building planning

element.

90

- 80 4 70

L o, 60 C

%f 50 .I 0 - = % g 40 rr 1 3 0 5 20 3 t- 10

0 0 10 20 30 40 50 60 70 80

Mean Month Temperature (F3

-Base Case -Wth Measures -Wth Solar

Figure 21. Residential Technical Potential Models

Figure 2 1 represents the building energy use models for a single average building in the residential sector. In an

energy use model of this sort, the line designated as the model specifies the monthly gas usage given a particular

average monthly outdoor temperature. The model is then typically used to estimate total normal annual energy

use by evaluating the model at each of the average monthly temperatures in a normal year. The blue line is the

current performance model for the residential sector. When we express this information in terms of an average

residential customer, we find that the warm weather base load is about 20 therms per month. As it gets colder,

gas usage increases to about 160 therms per month at 30°F.

The red line shows what happens as the house is insulated better and more efficient gas space and water heaters

are used. This more efficient building shows a lower base load due to better showerheads, a tank-less water

heater with low standby losses. And it shows significantly lower temperature sensitivity due to a more efficient

space heater and refinements to the building shell. Finally, the green line shows the gas usage model with solar

energy preheating the hot water and the house properly sited for passive solar gain.

Small Buildings There is a well developed community of interest and capability directed at residential space heat and water

heating efficiency. In most retrofit programs, heating efficiency is approached in the same treatment from its

three logical avenues: better furnace and distribution efficiency, lower thermal and infiltration losses, and better

controls. The water heating savings potential proceeds from lower flow fixtures, better furnace efficiency, and

lower tank standby losses.

One of the largest components of the potential is latest 90+ efficiency furnaces coupled to a leak tested duct

system. The next largest component is the improved thermal shell of the structure. Ultimately, all the diverse


Vecfren DSM Action Plan: Final Report N. Conservation Potential

improvements to small buildings energy use resolve into a change in base load and a change in the temperature

slope. Figure 22 shows the model of the aggregated small building use and the model of the same population

with all technical savings options employed.

100

90 Q -

80 u, p 3 = 70

3z 60

< 3 ~ 50

40 Q E a L 30 " Q $ = 20 n! t

10

0 20 30 40 50 60 70 80


-3- Base +Tech

Figure 22. Small Building Technical Potential Models

Figure 22 shows the effect of applying to every building a 90+ furnace, improved building shell, flow efficient

hot water fxtures, and even solar water heat and a passive solar space heating component. This reasonably

aggressive application of efficiency technology leads to a technical potential with a 54 percent reduction in gas

energy use.

Large Buildings The population of this planning element will have larger buildings with more complex controls than typical

residential applications. Usually, there will be a boiler. Often there will be a designated energy manager. This

type of situation has been the objective of energy management contractors because there are large enough

energy flows to create significant dollar savings.

The largest elements of savings for this group is associated with improved boiler efficiency and improved

controls. The thermal integrity of the shell in this group is subject to improvement especially with respect to

infiltration. Ultimately, all the diverse improvements to large buildings energy use will resolve into a change in

base load and a change in the temperature slope. Figure 23 shows the model of the aggregated large building

use and the model of the same population with all technical savings options employed.


Vectren DSM Action Plan: Final Re~ort IK Conservation Potential

- 30 0

5 25 z 2! - 20 ". Q, cng 75 m ., 2 z. u, 10 d - m 5 4- 0 I-

0 20 30 40 50 60 70 80


+Base -Tech

Figure 23. Large Building Technical Potential Models

Figure 23 shows the effect of applying to every building a 90+ boiler, improved building shell, flow efficient hot

water fixtures, and improved controls. This reasonably aggressive application of efficiency technology leads to

a technical potential with a 3 8 percent reduction in gas energy use. For the purposes of estimating technical

potential, the restaurant planning element has been included in the large building population.

Process Energy There has been no technical potential estimate for the process energy planning element. Most of the process

energy identified in the industrial sector gas usage models is by the top 100 or so gas transportation customers.

It is expected that these customers have highly specific and often proprietary processes not subject to "plug and

playy' program measures. But there may be some fundamental processes in this planning element, such as steam

production or controls, which may benefit from program offerings. Such application of applicable measures

would clearly have a technical potential, and this technical potential is not currently included in the overall

estimate of technical potential. If such measures were included they would in aggregate be a small number and

would not significantly change the overall estimate of technical potential.

A summary of the technical potential analysis by planning element is shown in Table 28. Our analysis of

technical potential shows that it is technically possible to cut usage in half. However, these estimates are not

realistic estimates of actual reductions because they are unconstrained by market, behavioral and budget

considerations.

- --

Prepared by oref front ~conornics and H. ~ i l ~ e a c h & Associates Page 37

Vectren DSMAction Plan: Final Report IK Conservation Potential

Table 28. Summary of Technical Potential by Planning Element (millions of therms)

I Total Load 1 Technical Potential I Planning Element Percent I Therms

Small Buildings Large Buildings

11 Total I 754 1 47% 1 351.1 11

Process Restaurants

Source: Our analysis of monthly usage data and applicable technologies

Conservation Measure Assessment

278.1 66.9

515 176 47 16

In order to evaluate technologies for their potential in gas DSM programs, it is necessary to compile detailed

information at the energy conservation measure (ECM) level of detail. We compiled this information through

an integrated approach that combined an extensive review of industry literature, the detailed analysis of Vectren

loads described earlier and our own expert opinion. Detailed assumptions at the ECM level are presented in

Table 29 for small buildings and Table 30 for large buildings. A discussion of the approach to measure analysis

follows these two tables.

54% 3 8%

Table 29. DSM Technology Assessment, Small Buildings

0% 3 8%


0.0 6.1


Table 30. DSM Technology Assessment, Large Buildings

Cost ~ffectiveness~ A primary consideration in measure screening is an estimate of measure cost effectiveness. This cost

effectiveness estimate is derived from subsidiary estimates of the measure cost (first cost and maintenance),

measure yield, measure life and an assumed discount rate. For the purposes of this screening exercise, the cost

effectiveness estimate will be expressed as a levelized cost (dollarsltherm) for the life of the measure. This form

of expression for the cost effectiveness, rather than an abstract costhenefit ratio, permits ready comparison of

the subject measure to the immediate or future marginal commodity cost of gas. At this stage of the analysis we

are more concerned with the rank ordering of measures by levelized costs than we are with a comparison to

avoided costs. A discount rate of five percent was used based on Vectren's weighted cost of capital.

Measure Savings-The screening relies on measure savings that are observable in real world billing histories.

Thus the measure savings used in this screening are the net observable savings after'and including the effects of

take back, measure interactions and background energy usage changes. Competent impact evaluations often

report savings at the measure level as in Table 3 1. .

6 Two types of cost effectiveness analysis are presented in this report. This section deals only with technology assessment using levelized cost. More comprehensive analysis is required at the program level. See Appendix D for a discussion of each type of cost effectiveness analysis.



Table 31. Net Savings by Measure (thermslyear)

The measure specific estimates in Table 3 1 were derived by regression from a year of billing and temperature

data for each site and they have been normalized to a long-term year. The table shows annual savings from a

residential gas efficiency program to about 150 therms per year. The table also shows quite a difference in the

savings estimates for a measure such as attic insulation or thermostats. These large differences may be due in

part to limited sample sizes for particular measures and they may be due to different participant selection and

other program circumstances. But they show, at least, that even with the best available quantification, measure

level savings can be quite variable.

In residential applications, these net savings are generally less than noted in competent engineering estimates.

For example, Ecotope 1999 estimated duct sealing would lead to savings of 135 therms per year. But this

example proceeded from a sample of pre-screened "duct losers" and did not include take back effects. The

Proctor and Blasnick work was on housing stock with a preponderance of basement ducts with notoriously

limited duct sealing potential. Here again, the savings due to a particular measure can be quite variable.

Blasnick(2), 1998 13 1 58 90 70

29

For this measure screening exercise, the residential measure savings estimates will be drawn predominantly

from the Proctor work in 1999 at SIGECO, the first column in Table 3 1. This work was done on a similar

housing stock and climate as expected at Vectren North. It could be argued that the savings to be expected from

Vectren North would be greater than those observed at SIGECO because it is slightly colder. No such

correction has been made to the savings estimates used in this analysis because it would be a small change to a

highly variable number. In practice, the best way to ensure high savings is to identify and enroll program

participants with probable efficiency problems.

Blasnick(l), 1998 187 9 1 189 70

54

Measure Average per Site Thermostat Attic Insulation Air Sealing Duct Sealing Wall Insulation CO Fix Furnace Check Other

The large building (commercial) measure savings estimates will rely on engineering ratios to characterize

changes in boiler efficiency and controls improvements, the largest measures in this category.

Proctor, 1997 172

52 47 44 204 151 28

Measure Costs-Cost information for this screening exercise includes the incremental costs of the measures

(which may differ depending on context new or existing) and the costs of site access. But the costs do not

include the program costs or the often underestimated cost of participant recruitment.


Vectren DSMActwn Plan: Final Report N. Conservation Potential

Incremental costs are the costs of acquiring and installing the measure over the base case alternative. Obviously

the costs can vary significantly with the size and complexity of the job. With careful design and execution, a

program may hope to beat these costs; but for the purposes of this screening, the assumed costs will serve to

place the measures in reasonable cost effectiveness categories.

For this screening, a prototypical residential building is assumed for the small building measures and a larger

prototype is assumed for the large buildings. In all cases the costs are measured in 2005 dollars.

Particularly important for costs applied to large buildings is the issue of multi-fuel cost sharing. The significant

measures related to controls and commissioning typically involve a comprehensive view of building energy use

which necessarily includes the control of electrical energy uses, as well as, gas. The benefits of controls and

commissioning are often strongly electric savings. There is an inherent multi-fuel cost share for these measures.

Therefore, the gas related costs for these measure types are assumed to be one-half the total cost.

Measure Definition-Measures are classified here into generic groups. For example, the measure group

classified as commissioning broadly means performance review and correction, and may involve billing review,

monitoring, interviewing, and analysis. The measures related to food service or laundry are, in reality, quite

diverse covering various fryers, ovens, washers, etc.; but in this exercise they have been classified together.

It is also important to note that the measures considered in the screening do not include combined heat and

power (cogeneration) measures. Generally, these measures will increase gas use while decreasing electric use.

There is a potential fuel switching issue here, even though in many cases, the combined headpower approach is

clearly the more efficient and environmentally benign approach. In terms of physical potential, combined

headpower is quite large and it has the possibility of dramatically changing the energy landscape for the better.

However, at this stage of the analysis, the issue is too large and important to include as if it were a typical DSM

measure.

Screening Savings and Full Costs-This screening exercise must necessarily distinguish between measures with

minimal overall savings and those which may carry the program in terms of savings. Accordingly, the measures

have been characterized by a five-year program savings expectation, thermslyr, resulting from five years of

highly aggressive program activity.

For the small buildings, this level of activity is quite aggressive (of the order of 20,000 units per year) and

contemplates the participation of one-third of the residential customers, those with the highest annual usage.

Participation levels of this magnitude would achieve about 35 percent of the identified technical potential in five

years. While unrealistically high, these calculations provide a measure of savings useful for screening and

bundling technologies for program planning.


Vectren DSM Action Plan: Final Report N. Conservation Potential

In the case of smaller buildings, the cost effectiveness of addressing the low usage customers can be severely

diminished because of lower savings expectations. But as the program proceeds, impact evaluations will

provide a guide as to the costs and benefits of including the lower usage customers in the program. For the large

buildings, the five year program horizon is even more aggressive than for small buildings and the screening

savings would amount to about 60 percent of the technical potential. For large buildings programs there are far

fewer participants (of the order of a few thousand), but each site is potentially a complex unique job.

Cost Effectiveness Ranking and Spectrum for Small Buildings-The small building measures are ranked by

cost effectiveness in Table 32 along with available savings.

Table 32. Small Building Ranked Measures

The information in Table 32 is then cast into a cost effectiveness spectrum as in Figure 24. In this figure, the

measures have been categorized into three general categories: furnace, shell and contols.


Screening Savings

thermslyr 1 ,O 19,375 3,307,500 2,446,500 6,116,250 3,500,000 2,143,750 4,587,188 4,43 1,000 4,638,156 2,038,750 4,077,500

245,000 682,500

5,300,750 5,950,000 1,706,250 1,63 1,000

8 15,500 1,223,250

185,500 1,260,000 5,460,000

71,750 13,251,875

171,500 1,050,000

47,250 262,500

1 1,4 17,000

Measure

Proper HVAC Sizing Low Flow Fixtures CO Remediation Energy Star Construction EE Water Heater with Energy Factor of 0.6 or Better Required Programmable Thermostats Wall Insulation (RO-Rl 1) AFUE of 65 to 92 SFe - 748 thermslyr Space Duct Seal Wall Insulation (R11-R19) Floor/Basement Insulation Tankpipe Wrap Gas Clothes Dryer (Energy Star) House Sealing using Blower Door HVAC Tune Up AFUE of 82 to 92 SFe - 593 thermslyr Space Ceiling Insulation (R11-R38) Ceiling Insulation (R30-R3 8) Ceiling Insulation (R19-R3 8) AFUE of 65 to 92 MFe AFUE of 82 to 92 SFn - 432 thermslyr Space AFUE of 65 to 82 SFe - 748 thermslyr Space AFUE of 82 to 92 MFe EE Windows AFUE of 65 to 82 MFe EE Water Clothes Washer AFUE of 82 to 92 MFn - 247 thermslyr Space Gas StoveIOven Solar Water Heater

Levelized Cost

$/therm $0.07 $0.12 $0.19 $0.3 1 $0.35 $0.42 $0.5 1 $0.52 $0.52 $0.53 $0.61 $0.65 $0.66 $0.69 $0.69 $0.70 $0.76 $0.76 $0.76 $0.83 $0.95 $0.95 $1.11 $1.16 $1.52 $1.61 $1.68 $1.71 $1.80

Vectren DSMAction Plan: Final Report W. Conservation Potential

The furnace category includes more efficient furnaces and maintenance/efficiency improvements to existing

furnaces. The shell category includes whole house insulation and air sealing measures. The controls category

includes the measures that reduce usage, such as, low flow fixtures and solar site for passive gain.

Note in Figure 24 that most of the cost effective savings are associated with furnace improvements and

insulation. The most cost effective measures are the flow reduction fixtures, proper furnace sizing and solar

orientation for passive gain.

&$.XI $.20-$.40.$40$.60$.-$.80 $.80 $l.CC- $1.20 $ 1 . 0 $1.62- $1.83- $1.03 $1.20 $1.40 $1.60 $1.80 $2.00

Levelized Cost Bin

Furnace SkII Controls

Figure 24. Cost Effectiveness Spectrum - Small Buildings

The most expensive measures are glazing replacements at about $1.16/therm and solar water heat at about

$1.70/therm. Both these measures have significant savings potential but are not immediately cost effective to

either the utility or the customer. However, in both these cases, there is an amenity benefit and they could be

accessed by a partial subsidy ostensibly to cover the energy only benefits of the measures.

Note also in Table 32 and Figure 24 that there is a wide range on the levelized costs for furnace replacements.

The cost effectiveness depends on the annual heat load on the furnace. A full range of furnace replacement

scenarios is developed in Table 32 and it generally shows that smaller multi-family (or single) residences are not

cost effective candidates for upgrade from 82 percent AFUE to 92 percent AFUE. In practice, the cost

effectiveness of a furnace replacement at a particular site can be readily estimated from an annual billing history.

Cost Effectiveness Ranking and Spectrum for Large Buildings-The large building measures are ranked in

Table 33 by cost effectiveness along with available savings.


Vectren DSMActwn Plan: Final Report N. Conservation Potential

Table 33. Large Building Ranked Measures

The information in Table 33 is then cast into a cost effectiveness spectrum as in Figure 25. In Figure 25, as in

Figure 24, the measures have been categorized into three general categories: furnace, shell, and usage reduction

or controls. The furnace category includes more efficient furnaces and maintenancelefficiency improvements to

existing furnaces. The shell category includes roof insulation and low-e windows. The controls category

includes the measures that reduce usage; such as, low flow fixtures, solar water heat and, more importantly, it

includes comprehensive building controls.

Note in Figure 25 that most of the cost effective savings are associated with furnace improvements and

commissioning/controls. The most cost effective measures are the flow reduction fixtures, high efficiency

windows in new construction, and commissioning and controls.

35,000,000

30.000.000

Y) rn C 25,000,000 .E 8 z .2? 20,000,000 3 0 , 15,000,000

5,000,000

0 $0$.20 $.%$.a $.40-$.W $.6D$.W $.80$1.00 $1.00 $1.20 $1.40-

$1.20 $1.40 $1.60

Leverized Cost Bin

W Furnace w S k l l Controls

Figure 25. Cost Effectiveness Spectrum - Large Buildings



As with the small buildings the window upgrades and solar water heating are in a cost effectiveness class by

themselves.

The furnace improvements in large buildings include new boiler technology with efficiencies in the range of 93

percent plus. An attribute of these new boilers is that they can come to temperature or produce steam much

more quickly than older boilers, and they also have finer control on the output. Where modern boilers replace

older ones, a smaller boiler is usually suitable and extra savings associated with boiler startup and control are

often available.

It is readily apparent that for large buildings the cost effectiveness situation is much different than for small

buildings. In large buildings, most of the cost effective savings potential is associated with commissioning and

proper control of the energy use. Insulation plays a much smaller role in these buildings.

-- -


Vectren DSMAction Plan: Final Report V. Program Design Process

V. PROGRAM DESIGN PROCESS

The purpose of this section is to describe the program design process, including the melding of market and

technology assessment into the DSM program development process. Program designs presented in this report

build off of the work presented in the market assessment and conservation potential sections. The program

designs presented in this section will provide the reader with important details regarding the portfolio of

recommended programs and the target market, end-uses and technologies addressed. Detailed program plans,

including program budgets, cost effectiveness analysis and evaluation plans, are presented in subsequent

sections of this report.

Approach

The approach taken in developing the DSM Program Plans contained in this report was three-fold:

1 . Identify DSM technologies and screen them for consideration in the design of programs 2. Review industry literature regarding gas DSM program approaches and identify best practices from other

leading gas companies 3. Apply marketing approaches to packaged DSM technologies and techniques identified in the Vectren

DSM technology and market assessment

DSM Program Development Process

The process that has been undertaken in the development of the Vectren DSM programs is presented in Figure

26.

Figure 26. The DSM Program Development Process

This series of six steps in program design is a fairly standard approach that has been used in the development of

programs for electric and gas utilities throughout the u.s.~ It balances the engineering and economic

(3)

PACKAGE screened DSM Technologies into groupings

according to end-use applications and delivery approaches

-

American Electric Power, UtiliCorp, PacifiCorp, People's Natural Gas, Northeast Utilities, Northwest Natural Gas and others have employed similar models in the past decade.

(2)

SCREEN DSM Technologies according to utility-specific

criteria

(1)

IDENTIFY universe of DSM Technologies


)

1 (6)

FINALIZE selected set of DSM Programs for promotion, develop

implementation schedules and plans

(4)

DESIGN DSM Programs based on available budgets, best-practice marketing techniques and policy

considerations

)

(5)

ASSESS program costs and benefits individually and in terns of overall impact on gas company

revenues

'

Vectren DSM Action Plan: Final Report V. Promam Desian Process

characteristics of specific end-use technologies with public policy and corporate objectives. The process then

considers the specific environmental and market characteristics of the service territory in which the technologies

would be promoted. The result is a set of DSM programs that have a high likelihood of success in terms of

customer adoption and achievement of program goals.

The following issues were considered in developing the draft DSM Program Plans presented in this report:

Target Markets Served-What are the potential numbers of customers in each market sector that might be targeted for implementation of the DSM technologies?

To turn groups of technologies into appropriate DSM Programs (Step 3, PACKAGE; and Step 4, DESIGN)

these additional contextual factors must be considered, among others:

Marketing Approach-What types of tools and methods should be used to promote the adoption of technologies? What combination of education, customized information (energy audits), arrangement services (identification of contractors), financing options and direct incentives are required to move the market sectors to take the desired actions?

Once the best programs from a technical perspective have been determined, a final step is performed in

considering the appropriate DSM programs to select as the final group to be implemented.

Regulatory Environment-What policies exist in the regulatory community that might need to be taken into consideration in packaging DSM technologies into programs? Are there some programs that may not be cost effective from a Total Resource Cost (TRC) perspective that should be included for other reasons (e.g., low-income programs, research and development programs, special customer service programs)?

Related Governmental and Market Programs-Finally, what existing governmental programs (federal or state) and other market programs (lending institutions) might be available to help support the implementation of programs?

Applying the above questions to the set of technically feasible technologies already identified involves a

screening process. Screening the universe of DSM technologies requires the application of a set of customized

criteria unique to the utility for which programs are to be developed. The criteria usually applied to these data

are:

Market Potential-The potential for a technology to be adopted in a given market as based on current market saturation, availability, price and other measures of market acceptance.

Economic Requirements-The economic criteria that would be required for the market to adopt the technology, usually a cost-benefit measure (such as a C/B ratio of 1 or better) or a set payback rate (such as five years or less).

Cost Effectiveness-Usually expressed as levelized costs, provides a measure of the resource cost for comparisons with other technologies. Other standard tests can also be applied. Refer to Appendix C for a detailed description of these standardized tests.

Policv and Reszulatorv Considerations

Vectren's interest in developing a portfolio of DSM programs for its customers could not be timelier. Recent

reports from national news media suggest increases in natural gas prices of 52 percent, mostly due to disruptions



to the supply chain caused by Hurricane ~atrina.' Even prior to the recent disaster, as early as 2003, the

American Council for an Energy Efficient Economy (ACEEE) summarized the prudence of offering DSM

programs to consumers and utility customers in its report on "Responding to the Natural Gas Crisis: America's

Best Natural Gas Energy Efficiency Programs" (ACEEE Report No. U035).

Natural gas custonzers are facing rapidly rising costs. This has signiJicant adverse efects on individual customers as well as the broader economy ... Inzproved energy eflciency is a concrete step that custonzers can take to ofset price increases, but decades of experience with natural gas customers suggests that they won't necessarily take such a step without facilitation via energy eflciency programs. Moreover, the natural gas price problem creates serious societal costs as well, which strengthens the rationale for afJmative government policies to address this problem through energy eflciency.

Energy companies can take the initiative thenzselves to ofer their customers programs, but they also need support @om their regulators to make such programs a reality. Regulatory support may conze @om a variety of mechanisms, which include program cost recovery through rates, financial incentives for meeting established performance targets, and perhaps some type of 'lost revenue' recovery or decoupling ofprofitsJi.om sales volume.

... There is little time to spare to create and expand programs to serve customers presently not served by efJiciency programs. Generally, financial incentive progranzs can be created and inzplemented rather quickly, while programs ofering technical assistance and related services take more time to develop and implement ... Utility companies, governments and related organizations should view natural gas efJiciency programs as both a near-term and long-term element in an overall strategy of helping natural gas customers manage their energy costs, as well as helping our economy deal with higher market energy prices. (p. 21)

In addition, fiom a policy perspective, current program design occurs in a post "911 1" context in which security

factors should be taken into account where possible. This is different fiom the earlier context of DSM program

design and in some cases it may lead to different kinds of program designs. Further, in the area of gas supply,

residential customers now compete with gas generation of electricity and also turn to electricity when they are

unable to pay their escalated gas bills. In the past few years as gas costs have risen and remained high, a

secondary effect has been an increase in the use of electricity when households cannot pay their gas bills. This

creates an increase in electric bills. The net effect at the household level is that energy bills are increasingly

interactive and can become difficult or impossible to pay. Both gas and electric utilities in much of the U.S. are

now experiencing payment problems unprecedented since the 1930s. These problems will intensify.

In the area of electricity supply, global warming (about which knowledge is non-controversial in scientific circles) is affecting us with an encroaching problem of physical limits. As an illustration of what this means, recently a travel magazine urged travelers to see Glacier National Park now, because soon there will be no glaciers. Already, tourist observation points in Glacier National Park and in the Swiss Alps no longer provide the view they once did. Similarly, when Scanada Consultants Ltd. recently co-sponsored a repeat of the Sverdrup polar expedition, the team reported people falling through the ice

8 EIA Short Term Energy Outlook, September 7,2005. The increase in gas price for the northeastern states is projected at 71%, for the nation as a whole it is 52%. Note that commodity cost of gas is only a portion of the gas bill, so that total bills will go up less than these amounts. Still, this year the size of the effect will be of "energy shock" proportions and this effect occurs in the context of long-term supply shortage.


Vectren DSMAction Plan: Final Report K Program Design Process

that had been rock solid about 100 years ago. The primary effect on electricity is in the projected depletion of hydro-generation resources, leading to scarcity and up-pricing in neighboring jurisdictions. This is a classic problem of physical limits. Accordingly, electricity price will continue to increase.

On the gas side, current industry publications generally accept that the depletion of current fields has occurred-roughly as projected 30 or 40 years ago. Accordingly, gas price will continue to increase. EIA projections of the proportion of gas supply that would have to be made up from liquefied natural gas (LNG) over the next twenty years seem overly optimistic. Given a projection of supply shortage and the unlikely possibility that LNG will prove out to the extent that is needed, gas price may be expected to continue to increase.

Just as with "911 1" security concerns, the physical limits problems for electricity and gas supply are new factors

that must play into the development of new DSM program designs and the formulation of rule modifications for

the assessment of DSM program designs. Supply problems that markets may not be fluid enough to solve and

security concerns were not major elements in prior cycles of DSM. Now, planning has to take these factors into

account.

Simply from a technical perspective, we need to move towards gas equipment functioning independent of

electricity supply. This may not be achievable in a practical sense from current program designs, but it is a

direction in which program design must now move. For example, we need to move towards gas heating that

will keep Indiana homes warm during an electric power failure. As a final policy concern, we need to look

critically to see if some of the things we mean to accomplish through programs are actually being accomplished.

For example, we generally try to install programmable thermostats. Most, if not all, thermostats available on the

market are difficult to use, especially by senior citizens, and in low light areas such as hallways. We may need

to improve equipment specifications to ensure that the benefits we project for some of our improvements are

easy to obtain through better product design.

Target Market Segments

Consideration of the best DSM programs to launch must first address the specific markets to which the

programs would be promoted, their needs and likely market behaviors. This section summarizes the market

segmentation analysis discussed in detail in the Market Assessment report. The groups of Vectren North

customers described below represent the specific target market segments to which DSM technologies and

programs will be addressed. The Vectren North service territory serves about 570,000 customers distributed

into the four basic rate classes as in Figure 27 and by end-use in Figure 28.

The types of end-uses of gas that dominate each customer sector differ, but are dominated by:

space heating (furnaces and boilers) and water heating in the residential and commercial sectors, and

process uses, water heat applications and space heating in the industrial sector.


Vectren DSMAction Plan: Final Report K Program Design Process

Residential 62%

Figure 27. Vectren Customers by Rate Class

Space Heat 65%

8%

Figure 28. End-Uses of Gas by Vectren Customers

In the previous section of this report conservation potential was addressed by each of four planning elements

(see Table 27). These planning elements are derived by re-grouping various residential, commercial, and

industrial sector customers around common building types and systems. In this section, we build off of the

planning element analysis to address the groups that DSM programs would most logically be targeted to and

what specific groups of technologies should be promoted to help increase efficient uses of natural gas within

each segment. A description of the planning elements follows:

The small building planning element is composed of the entire residential sector and about 35,000 customers in the commercial sector. This population is characterized by residential scale buildings heated by small forced air furnaces, and with water heat in residential scale gas hot water heaters.

The large building planning element is composed of the whole commercial sector and is characterized by large scale buildings, a predominance of boilers and reasonably complex building or facility controls.

The process energy planning element is composed of the largest 350 industrial transport customers. This population is characterized by its diversity of uses and by the likelihood the process energy cannot be manipulated without special care for the process.

The restaurant planning element is drawn from the commercial sector. This population is characterized by a wide range of specifically different gas energy intensive operations underlain by the use of similar gas-using equipment in hndamentally similar food preparation processes.

prepared by Forefront Economics and H. Gil Peach & Associates Page 51


Small Buildings There is a well developed community of interest and capability directed at residential space heat and water

heating efficiency. In most retrofit programs, heating efficiency is approached in the same treatment from its

three logical avenues: better furnace and distribution efficiency, lower thermal and infiltration losses, and better

controls. The water heating savings potential proceeds from lower-flow fixtures, better furnace efficiency, and

lower tank standby losses.

One of the largest components of the potential is the latest 90+ efficiency furnaces coupled to a leak tested duct

system. The next largest component is the improved thermal shell of the structure. Ultimately, all the diverse

improvements to small buildings energy use resolve into a change in base load and a change in the temperature

slope. The effect of applying to every building a 90+ furnace, improved building shell, flow-efficient hot water

fixtures, and even solar water heat and a passive solar space heating component leads to a technical potential

with a 54 percent reduction in gas energy use. Actual achievable potential is, of course, much less, but the high

technical potential defines a key area for effort.

Large Buildings The population of this planning element will have larger buildings with more complex controls than typical

residential applications. Usually, there will be a boiler. Often there will be a designated energy manager. This

type of situation has been the objective of energy management contractors because there are large enough

energy flows to create significant dollar savings.

The largest elements of savings for this group is associated with improved boiler eff~ciency and improved

controls. The thermal integrity of the shell in this group is subject to improvement especially with respect to

infiltration. Ultimately, all the diverse improvements to large buildings energy use will resolve into a change in

base load and a change in the temperature slope. The effect of installing a 90+ boiler, improved building shell,

flow-efficient hot water fixtures, and improved controls to every building defines the maximum technical

potential. This reasonably aggressive application of efficiency technology leads to a technical potential with a

38 percent reduction in gas energy use. Actual achievable potential is, of course, much less, but high technical

potential indicates an area for intensive effort. For the purposes of estimating technical potential, the restaurant

planning element has been included in the large building population.

Process Energy There has been no technical potential estimate for the process energy planning element.

Appliance installation rates reported in the Market Assessment section (see Table 10) were also used in

considering the likely numbers of customers to which each program would apply, and subsequently, what

reasonable level of participation might be expected in the customer segments. In the case of existing homes, the

DSM programs target the replacement of existing appliances in most cases. In the case of new housing,


Vectren DSM Action Plan: Final Report V. Program Design Process

estimating participation levels involves looking at the percentage of new households that have not adopted the

technologies of interest as being the group to which programs would be targeted.

As noted earlier, identification of technical potential provides the totality of energy savings that is technically

feasible, but not achievable for a variety of reasons. Market potential is a measure of the likely uptake of

technologies from natural market forces, program incentives or aggressive marketing (or other exogenous

factors such as energy price increases or effects of natural disasters). Program participation levels represent an

even lower measure of the likely number of customers one can anticipate will elect to install technologies and

sign up for services offered through Vectren's DSM programs. Participation levels typically start out low as

programs are launched but before advertising or word-of-mouth begins to disseminate information about their

offerings. They often follow an "S-shaped" curve, the standard market penetration model for the adoption of

products along a typical market cycle. Figure 29 illustrates the concept of the different levels of potential that

underlie the program planning process (percentages are merely illustrative).

Technical Potential (1 00%) Market Potential (85%) \

Program Potential, Aggressive (70%) /

Figure 29. Conceptual Illustration of Technical, Market and Program Potential

Projected participation rates in a program can be set fairly high, at the Aggressive Program Potential level, if one

assumes a robust program budget, fairly high incentive levels or a significant marketing and media campaign.

For this study, we assume a more conservative program potential in setting the participation rates, which in turn

is based on a reasonable program budget (within the range of gas programs in the region based on cited

references), leveraging of external partnerships and delivery agents (community-based organizations, trade

associations) rather than increasing internal staffing, and more modest incentive levels. This lower range or

target for achievement of Program Potential is reasonable given that this would be the first few years of program

implementation, and that, if successful, more aggressive budgets and program services might be added in the

future.

Technology Screening

This section describes the lists of DSM technologies considered for packaging and promotion through programs.

As described in the Conservation Potential section, a comprehensive list of DSM technologies applicable to


Vectren DSMActwn Plan: Final Report K Program Design Process

natural gas end-uses was developed based on existing programs of other utilities and from various industry-

published reports. The list of technologies identified for consideration is provided below in

Table 34, broken out by small and large building applicability.

Table 34. DSM Technologies for Natural Gas End-Uses and Customer Segments

Technology Characteristics Each natural gas DSM technology from

Table 34 is then considered in terms of a set of characteristics, such as:

Large Buildings (Large Commercial and Industrial)

solar water heater

EE boilers, various sizes commissioning relretro-commissioning window film low-e windows roof insulation energy efficient construction

clothes washer (Energy Star) gas range gas clothes dryer

tankfpipe wrap low flow fixtures EE water heater with energy factor of 0.6 or better demandtank-less water heater custom measures

End-Uses

1. Renewable Energy

2. Space Conditioning

3. Appliances

4. Water Heating

5. Process Uses

Sector-residential or commerciaVindustrial

Building type to which it applies-the market segments identified in the Assessment phase

Small Buildings (Residential and Small Commercial)

solar water heater EE furnace - AFUE of 90 or > EE furnace - AFUE of 90 or > EE furnace - AFUE of 82 to 90 EE windows programmable thermostats ceiling insulation (R11-R3 8) ceiling insulation (R30-R3 8) ceiling insulation (R19-R3 8) sealing using blower door wall insulation (RO-R1 1) wall insulation (R11-R19) proper HVAC sizing floorhasement insulation W A C airflow calibration Energy Star construction 'lothe' washer (Energy Star) gas range Gas clothes dryer Heat-rated gas fireplace tankfpipe wrap low flow fixtures EE water heater with energy factor of .6 or better Demandtank-less water heater Not applicable

Percent energy savings per unit

Per unit installed cost in dollars

Average life of equipment

Yearly operation and maintenance (O&M) costs, if any

Levelized costs per unit savings



Data for each technology were presented in an earlier section (see Table 29 and Table 30). These were derived

from various sources, including published reports and industry literature listed in the Data Sources and

References section of the Appendix. The technologies identified in

Table 34 were then "screened" based on a sorting process for a set of standard variables. This produces a table

of ordered technologies (one for small buildings and one for large buildings) based on levelized costs, with

secondary ordering by market barriers and the relative need for the program. These subjective factors are

described below.

In addition to these engineering and economic characteristics, two indices were created to address the likelihood

that each technology would be adopted based on existing market barriers and the particular need for the program

given situational factors within the service territory, governmental or regulatory policies, and Vectren's own

corporate interests. These two indices, while entirely subjective, provide an opportunity to apply a

"reasonableness" measure to the engineering and economic data used in the analysis.

Market Barriers-This index is measured using a score of one to five, with one representing low market barriers

and five representing many market barriers to adoption. An index of one implies that the technology is

commercially available, is reasonably priced and easily found in the market, is reliable and relatively easy to

have it installed and functioning. An index of five implies that the technology is new or relatively untested, is a

custom or special order product, is high in price, or has questionable reliability given existing experience in the

marketplace. Low rankings suggest that, in order for a technology to be implemented, it will require education

of end-users as to its benefits, training of market actors to promote the technology, and possibly incentives to

buy down the first cost.

Need for Program-This index is also scored from one, meaning low need, to five meaning a strong need. The

index can include such factors as circumstances unique to Vectren's service territory (as determined through the

market assessment and through discussions with the Advisory Board and Vectren management) or governmental

programs in existence in Indiana being promoted by state agencies or programs of interest to regulators. This

factor also would be the place to acknowledge particular needs of certain industrial groups prevalent in

Vectren's service territory that might make a targeted program most appropriate.

The economics, engineering characteristics, and subjective factors are used to sort the technologies into a list

that provides a preliminary prioritization for further consideration. The resulting list of technologies with cost

effectiveness, market barriers and the "need for program" indices is shown below in Table 35.


Vecfren DSMAcrion Plan: Final Report V. Program Design Process

Table 35. Sorting of DSM Technologies

These three sortings provide an ordered set of technologies that would require varying levels of education,

market interaction andlor incentives in order to achieve adoption. High scoring technologies would presumably

require less subsidy or promotion to achieve penetration, where lower scoring technologies (i.e., lower on the

list and those with high market barriers) might require more information or incentives. On the other hand, lower

scoring technologies are often packaged with other higher scoring technologies in a single program aimed at the

same end-use to improve their overall cost effectiveness. A final step is to verify some of the data identified

from secondary sources with that specific to the utility and service territory. Even without this step, experience

shows that this screening process usually produces a fairly consistent set of high potential technologies across

various utilities.


Small Buildings Technologies

Solar Siting

Low Flow Fixtures

Ceiling Insulation (R19-R38) $0.76 2 EE Windows $1.16 4

Gas Stove/Oven $1.71 4 2-3 *500/0-75% of controls benefit is electric savings so

Solar Water Heater $1.80 4 2-3

Large Technologies

Low Flow Fixtures

Energy Star Gas Stove

Levehd Cost

($/therm)

$0.09 $0.12

Levelized Cost

($/therm)

$0.22 $0.30

Market Barriers

@=low, +high)

5 1

Market Barriers

(1-5) 1 3

Need for Program

(1-5)

3 3-5

Need for Program

(1-5)

4-5

3

Vectren DSMAction Plan: Final Report V . Program Design Process

Technology Groupings The technology screening process points to a set of DSM programs that group the highest ranking technologies,

based on cost effectiveness, in packages targeted at end-uses within specific target segments. The most

favorable of the technologies identified through the screening process are then grouped in the next step

according to what segments of the target market they would be targeted to and how. For example,

weatherization type measures that scored well in the screening process can be grouped together into a low-cost

measures program.

Packaging of Technologies into Programs The following is a discussion of the technologies most appropriate to each customer sector. The specific

technologies to be promoted through programs are then described in Table 36 in the next section of the report.

Residential-The Technical Potential Study reveals that Vectren North is, overwhelmingly, a residential sector

customer base. Within the residential sector, natural gas serves primarily space heat and water heat end-uses.

This suggests that programs that package DSM technologies and techniques aimed at increasing the energy

efficiency of gas furnaces, reducing air leaks and improving weatherization of the home would be appropriate.

Given the relative rankings of 90+ AFUE gas furnaces and those with slightly lower ratings, two options might

be appropriate. One option would be to propose replacing the 80+ furnaces with a new 80+ variable speed

furnace; for those more inclined to adopt the higher efficiency technology, the AFUE 90+ could be offered as

well. A combination of high efficiency water heater replacements and flow reduction technologies would be

appropriate for either a second program or a single program for existing buildings. For low income customers, a

program could leverage existing community action agency initiatives (federal weatherization and payment

assistance programs) through a coordinated program with added utility money.

Technologies considered in the analysis, but not included in programs at this time, include window replacements

for saving energy costs for homes heated with gas and heat-rated gas fireplaces for displacing other

supplemental heat sources.

Commercial-Small commercial establishments with building characteristics similar to the residential sector

could be targeted through a Small Buildings program, focusing on the end-use technologies noted above. The

Technical Potential Study identifies key commercial end-uses, of which space heat and water heat are the

largest. Based on the site visits with Vectren representatives, a second program targeted at restaurandfood

preparationkitchen end-uses would also be appropriate due to their number and the opportunity for increased

gas energy efficiency.

Renewable Energy-The Technical Potential Study shows curves that include a solar option as a way to increase

gas energy efficiency through two technologies: solar water heating as a way to pre-heat the water in storage


Vectren DSMActwn Plan: Final Report K Program Design Process

tanks prior to the need to turn on the gas heating component and small packaged co-generation systems for large

residential and small commercial application that use natural gas as a backup.9

Industrial-Although individual industrial loads are large, the Technical Potential Study shows that the sum of

the industrial load (including transportation customers) is quite small compared with residential. Industrial

customers are generally less supportive of prescriptive or mandated utility DSM for their sector while

supporting economic applications of DSM technologies on a case-by-case basis. The appropriate approach to

offering energy efficiency services to this sector is therefore a custom approach for each interested participant,

andlor funding of emerging technology research and development and demonstrations. Some industrial

customers expressed potential interest in a technology innovations group or innovative on-site metering options.

For the most part, any technology promotion would be most successful if coupled with a rate incentive.

Vehicles-Some gas utilities include natural gas vehicle (NGV) conversions as a conservation program option.

This program is targeted at municipal governments, schools and businesses that operate fleets of vehicles.

Incentives are provided for covering a portion of conversion costs. In some cases, incentives covering a portion

of the incremental costs of NGVs over standard fleet models are provided for new vehicles. These programs

assume that fueling stations are available for market use or would be installed on the customer's site.

Application of Marketing Approaches In the previous step, the screened technologies are then grouped into programs targeted at specific markets. The

marketing techniques to be used will depend on: 1) the maturity and availability of the technologies in the

marketplace and 2) the specific market barriers that exist that must be overcome in order to achieve desired

levels of market penetration. Technologies that have a high first cost may require incentives and those that are

relatively new or emerging may require demonstration programs. Technologies that would be packaged with

load management systems (load control, innovative rates, or timing devices) may require education regarding

benefits that would accrue to customers. The marketing techniques to be used will also depend upon the

budgets available for program implementation.

Once the set of technologies applicable to Vectren's service territory is developed, then they are grouped into

programs reflecting the customer sectors and whether they would be applied to new or existing applications.

The final step in converting the groups of technologies into programs is by adding budget considerations,

marketing approaches and policy requirements. The primary factor in designing successful DSM programs is to

correctly identify the market barriers to adoption. This is typically done through market research, combined

with appliance or equipment saturation data to understand the existing level of penetration for each technology;

such data as provided by Vectren has been taken into consideration in this analysis. Other data necessary for

program design includes information on technology prices and availability, as well as distribution channels;

This latter program has been promoted in New Jersey under that State's Comprehensive Resources Analysis program. See New Jersey Board of Public Utilities website for more information.


Vectren DSMAction Plan: Final Re~ort V. Propram Desim Process

again, the best information available has been used, and the specific data elements can be improved upon during

the review and revision process. The resulting programs are described in the next section of the report.

DSM Program Designs

According to the original assignment for the project, Vectren North articulated the goals for DSM program

design as follows:

1. Potential programs should give priority to achievable energy savings, customer benefits, cost effectiveness ratios, adoption potential, market transformation capability and ability to replicate in the Vectren North service tenitory.

2. Individual program plans by customer segments will contain the elements listed: a. Detailed description of the program-Based on best practices fiom a variety of sources, including

ACEEEYs 2003 summary report on best natural gas energy efficiency programs. lo

b. Reasons why the program would be successful in Vectren North's service territory-Derived from the Market Assessment section of this report and background research from earlier tasks.

c. Number of customers within the customer classfsegment that are likely to adoptluse the proposed program-Derived fiom the Market Assessment section of the report with a percent participation estimate based on experience fiom other utilities with similar programs; informed by actual results from other utilities offering similar programs."

d. Achievable energy savings-From a variety of sources listed at the end of this report, consistent with the technology assessment and published reports.

e. Cost effectiveness ratiosfrating per individual program-calculated using the Total Resource Cost, Participant, Administrators Cost, and Ratepayer Impact Tests (see Appendix D).

f. Marketing plans which should include incentives, rebates and preferred distribution channels and how each reduces existing barriers to proposed program adoptionlacceptance-Based on best practices from a variety of sources listed at the end of this report; incentive amounts based on examples from natural gas companies identified in footnotes and listed in Appendix C.

g. Detailed budget plans complete with explanations of anticipated increasesfdecreases in financial and human resources during the expected life of the program-Based on best practices from a variety of sources listed at the end of this report and a five-year program life.

h. Recommended methodology or tracking tools for recording actual performance to budget- Based on current standard practice using simple commercially available software (no black boxes or proprietary models recommended).

i. Proposed program evaluations and reports-Based on current standard practice using a logic model approach.

Some discussion is worth noting regarding assumed participation rates in the program designs. First, the

participation levels indicated in the program designs, and used in the analyses, represent target participation

levels that reflect the level of funding suggested and the size of the target markets for each program. They are

relatively conservative and would need to be adjusted after experience in the field to reflect Vectren's actual

market response to the programs. While it is useful to see what other utilities and public benefits entities have

achieved in terms of percentage participation rates for similar programs, such figures are not comparable for a

lo Kushler, Martin, et al.; Responding to the Natural Gas Crisis: America 's Best Natural Gas Energy EfJiciency Progranzs. ACEEE Report No. U035 (Dec. 2003). l1 Primarily People's Natural Gas; see References for citation.

Prepared by Forefront Economics and H. Gil Peach C Associates Page 59

Vectren DSMAcfion Plan: Final Report V. Program Design Process

variety of reasons and offer only scant insights into what might be achievable for Vectren. First, one must

consider differing program budgets and pre-program market conditions such as conservation awareness levels,

market interest and willingness to invest in energy efficiency over other consumer choices. Critical to what one

might expect to achieve in terms of participation levels are factors such as energy costs in general and gas rates

in particular. Given the prediction that gas rates will increase significantly over the time period that these

programs are introduced, the participation levels may indeed be too conservative. The fact is, only time and

experience will tell. If programs become quickly over-subscribed, Vectren will need to consider either reducing

incentive levels to spread annual budgets out over time or accelerating the five-year budgets to be able to serve

more customers than planned for.

The other reason why the experience of other energy efficiency programs elsewhere is of very limited value is

that the numbers available in published reports are themselves not comparable or easily converted into useful

data. Most program status reports and evaluation studies state how many participants each program has

achieved, rather than reporting what percentage of eligible customers have participated. Their chief value is in

viewing participation levels and rates over time, from year to year, rather than compared to market potential.

This makes any comparison to what might be achievable to Vectren a moot exercise.

With those caveats, the program designs by necessity incorporate reasonable levels of market uptake given the

program budgets, incentives and trainingleducation (collective market stimuli) that are based on other similar

program offerings, and evaluation experience of identified "best practice" natural gas programs in the literature.

And while one has to select a set of numbers as a starting point from which to develop a plan, in every case

where a new initiative is undertaken, it is logical to expect the plan to adjust as experience and market

conditions reveal more realistic outcomes.


Vectren DSMAction Plan: Final Report W. DSM Program Plans

VI. DSM PROGRAM PLANS

The table below provides a summary of the recommended DSM Programs and the technologies that would be

promoted to the appropriate market segments. Each program is discussed in the following text.

Table 36. Recommended Programs and Technology Groupings

Program 1. Small Buildings Energy Efficiency Program

The Small Buildings program would provide Vectren's existing residential and small business customers with

DSM Technologies

Energy efficient furnaces, duct sealing, weatherization measures, blower door, EE water heaters, flow restriction measures, tank and pipe wraps, gas ranges, clothes washers (for home with gas water heating), dryers, setback thermostats and natural gas fireplaces Boiler replacement, water heating equipment

All identified gas end-uses

Energy efficient gas ranges, ovens, broilers, warmers and related processes

Energy efficient furnaces, duct sealing, weatherization measures, blower-door, EE water heaters, flow restriction measures, tank wraps

Emerging high efficiency natural gas technologies, use of renewable technologies to off-set or enhance gas technologies

Design incentives to upgrade planned gas equipment to energy efficient options, reduced hook-up fees andlor line extension costs Emerging technology research and demonstration and regulatory liaison activities All technologies

three services:

End-Uses

Heating, water heating, cooking, laundry, f~eplaces

Heating, water heating

Heating, water heating, process uses

Cooking and food preparation

Common area boilers, water heating and laundry; individual unit water heating, weatherization All gas end-uses

Any gas end-uses being considered

All end-uses

All end-uses

Program Type

1. Small Buildings Energy Efficiency Program

2. General Services Energy Efficiency Program

3. Customized Energy Efficiency Program

4. Hospitality Industry Energy Efficiency Program

5. Multi-Family Building Energy Efficiency Program

6 . Innovative Energy Efficiency Technologies Research and Demonstration Program

7. Energy Efficient Builder Program

8. New Program Development and Regulatory Affairs

9. Public Education and

1. A comprehensive energy audit providing a package of low-cost weatherization measures plus specific recommendations of equipment and techniques for using gas more efficiently,

2. Assessment of the home's air leakagefinfiltration levels through the use of blower door technology, and

Target Market

Existing residential (single family up to 4 units) and small commercial buildings (defined by square footage, employees or gas usage)

Existing and new medium to large commercial and industrial facilities Existing and new large commercial and industrial facilities Restaurants, bakeries, institutional housing, hotels, hospitality facilities and other cooking facilities Multi-family buildings with 5 or more units, dormitories, hotels, other large residential facilities

All markets

Residential and non- residential new construction

All sectors

All sectors



3. A menu of technology-specific incentives for recommended measures.

The primary natural gas technologies promoted through this program are presented in Table 37 below.

Table 37. Measures and Incentives - Small Buildings Energy Efficiency program12

1 8.Blower Door Treatment No cost

This program is modeled off of the New York State Energy Research and Development Authority's Home

Performance with ENERGY STAR program, cited in a recent ACEEE report as being an exemplary natural gas

energy efficiency program.'7 The premise of this type of program is to identifl energy saving opportunities

through a highly trained contractor skilled in conducting energy audits plus blower door treatment of the home.

The contractor then makes a customized set of recommendations, showing applicable incentives and free

measures, and proceeds to install the equipment upon approval by the customer.

Rational for Program This program would address the largest segment of Vectren's customer base with cost-saving information and

incentives to help mitigate the increasing costs of natural gas use associated with the primary end-uses of space

and water heating. It would provide customers with information about the condition and safe use of their

existing equipment, improve the efficiency of their homes and facilities through weatherization measures, and

reduce their consumption of gas through upgrading appliances to higher efficiency new models.

Program Participation and Savings Estimated participation and savings over a five-year period is presented in Table 38 below. Approximately

525,000 residential and small commercial buildings are estimated to be potential participants in this program.

After five years, the program is expected to treat nearly 40,000 homes and small businesses (7.5 percent of

potential participants) and deliver 7.8 million therms of annual gas savings. Savings per participant are based on

the weighted average savings of individual measures.

'* Based on various utility programs, see Appendix B. 13 Based on NJNG. l4 Based on NJNG. l5 Based on Aquila. l6 Based on Interstate Power and Light. 17 Kushler, Martin, et al.; Dec. 2003.


Vectren DSMActwn Plan: Final Report n. DSM Program Plans

Table 38. Estimated Participation and Savings - Small Buildings Energy Efficiency Program

Marketing Plans The primary barrier to the adoption of higher efficiency appliances is first-cost, as these products typically sell

for a premium over standard gas appliances. The first-cost barrier is overcome through targeted incentives for

the upgrading of equipment to higher efficiency products such that the incrementally higher cost over standard

equipment is reduced or eliminated.

A second barrier may be the lack of knowledge as to how much customers might save on their gas bills from

weatherization and installing such equipment. This barrier is overcome through the energy audit, where Vectren

contractors conduct a walk-through review of the facility's gas-using equipment and related building conditions

(air leaks, thermostat settings, condition of hot water pipes and water tank, faucets, etc). Literature may also be

provided to show home and business owners what behavioral changes affect energy consumption.

The program would be promoted by an energy audit contractor in combination with point-of-purchase

information at area equipment retailers and distributors of natural gas equipment for the residential and small

commercial markets. Incentives would be paid directly to customers mailing in receipts to the vendor on behalf

of Vectren.

Detailed Budget Plans An estimated five-year budget for this program is provided below in Table 39. The anticipated cost to Vectren

for offering this program to customers involves budgets for:

1. Vectren administrative costs to develop, advertise, oversee and monitor the program 2. A vendor contract to market and deliver energy audits to customers, usually charged on a per completed

audit basis plus a management fee 3. A per audit charge subsidized by vectrenl* 4. Costs for selected low-cost measures that can be installed at the time of the audit (flow restrictors, pipe

wrap, etc.) 5. Incentives for the installation of recommended measures as demonstrated through the provision of

receipts by the customer

ls Evaluation studies have shown that charging the customer a portion of the audit fee, for example $35, helps to attract the more serious buyers who intend to act on the recommendations, and results in a higher perception of value on the part of the customer for the audit service than if it was offered for free. Low income customers may be offered the service for a reduced rate or for fi-ee if the fee is problematic.



Costs to participating customers include:

1. Customer's share of the per audit fee 2. Customer's share of the costs of covered measures and equipment 3. Installation costs

Table 39. Estimated Five-Year Program Budget - Small Buildings Energy Efficiency Program

$25,000 0.2%

$263,732 $263,732 $263,732 $263,732 $263,732 $1,318,660 10.6% General Public

$150.00 $393,750 $787,500 $1,181,250 $1,575,000 $1,968,750 $5,906,250 47.5%

$70.00 $183,750 $367,500 $551,250 $735,000 $918,750 $2,756,250 22.1%

Evaluation $16.48 $43,252 $86,503 $129,755 $173,007 $216,258 $648,775 5.2% Total

Variable $236.48 $620,752 $lJ41,503 $1,862,255 $2,483,007 $3,103,758 $9,311,275 74.8%

Total Budget $1,267,721 $1,863,472 $2,484,224 $3,104,976 $3,725,727 $12,446,120 100.0%

Performance Tracking The audit vendor contract should include monthly reporting of numbers of audits performed, measures installed

and complete justification for invoiced amounts to be compared against program budgets. Program incentives

paid should be supported by documentation from participating customers such as receipts or contractor's

invoices itemizing the measures and appliances installed.

Coordination with Low-Income Program Development of recommendations for low-income issues is a separate area and was not included in tasks for

development of the DSM Action Plan. However, given the need for a balanced understanding for planning,

documentation of greatly increased need for low-income and moderate income customer service planning and

assistance is provided in this report in the section on Alternative Forecast and Policy Parameters (Section VIII),

both in the low-income subsection and, more generally, in the discussion of alternative forecast. General low-

income recommendations are provided at the end of the Section.

The Small Buildings Energy Efficiency Program is the DSM program in this report that overlaps with low-

income residential services. The most efficient way to combine the DSM objectives of the Small Buildings

Energy Efficiency Program with low-income customer service objectives is to coordinate the Small Buildings

Energy Efficiency Program with the federal Low-Income Home Energy Assistance Program (LIHEAP),

administered through Indiana's Energy Assistance Program, which provides financial assistance to low-income


Vectren DSMAction Plan: Final Report vl. DSM Program Plans

households to maintain utility services during the winter heating season, along with low-income home

weatherization services delivered in each county through the Community Action ~ ~ e n c i e s . ' ~ In a coordinated

program, generally, all utility DSM incentive amounts in the Small Buildings Energy Efficiency Program (or a

tailored equivalent) are used to "buy down" costs of equivalent functions or items in the Community Action

Agency Weatherization Assistance Program.

Beginning January 1,2005, eligible customers of Vectren, who have applied for the state's LIHEAP through

local Community Action Agencies, were automatically enrolled in the new Universal Service Program and will

receive bill reductions in addition to LIHEAP. Monthly bill reductions will range from 9 to 32 percent of the

total bill (not including LIHEAP benefits), depending on the consumer's income level. The pilot Universal

Service Program also provides additional funding for weatherization. This type of coordination is the most cost

effective and eff~cient way to combine residential DSM objectives with low-income customer service

objective^.^'

Program 2. General Services Energy Efficiency Program

This program would serve existing large commercial and industrial facilities with prescriptive equipment rebates

for upgrading heating, water heating and gas cooling systems; boiler replacement, water heating equipment, tune

ups (building commissioning) and control systems. An optional Targeted Technical Assessment service would

be available on a cost-shared basis from trained contractors for those customers wishing to have a professional

assessment of energy savings opportunities performed on their facility before making decisions. Larger and

customized retrofits would be covered under the Custom Program (Program 3). The suggested incentive levels

for selected measures are listed in Table 40 below.

l9 During cold winter months, this program helps prevent utility companies £tom shutting off home heating service to low- income families. During hot summer months, the Energy Assistance Program provides limited funds for the purchase of fans, distributed at the local level. The state eligibility criterion is currently set low, at 125% of the current federal poverty level, which equates to one-half of a level of self-sufficiency income for a family. This appears to leave a gap to be addressed for working families who also experience income problems but are above the state eligibility level for the Community Action Agency Weatherization Assistance Program. The Company and the Advisory Board may want to take this into account with modifications so that there will be some participation in the Small Buildings Energy Efficiency Program within this income group. 20 The two key references for development of coordinated utility and Community Action Agency Weatherization Assistance Programs are Hill and Brown (1995) and MacGregor and Oppenheim (2002). Hill and Brown show how to optimize cost-benefit results for coordinated programs with separate, but complementary, objectives and cost-benefit criteria. MacGregor and Oppenheim provide information on the general state of coordinated utility/Community Action Agency programs, including the strengths and problems of coordinated programs using Massachusetts and Texas examples. Hill, Lawrence J. and Marilyn A. Brown, "Cost-Effectiveness of Coordinated Programs," Evaluation Review, Vol. 19, No. 2, Pp. 18 1 - 195. MacGregor, Theo and Jerrold Oppenheim, Coordination between Utility and DOE Low-Inconze Weatherization: What do PubIic Utility Comnzissioners Need to Know? Monograph prepared for Oak Ridge National Laboratory Energy Division and UT-Battelle, LLC, 2002. (Also available online at www.democracyandregulation.com)


Vectren DSMAcfion Plan: Final Report I.?; DSM Program Plans

Table 40. Measures and Incentives - General Services (GIs) Energy Efficiency Program

Rational for Program Helping the commercial and industrial sector reduce operating costs is a standard part of good utility operations,

and providing information and incentives for encouraging equipment upgrades falls into that category. While it

can be assumed that many large customers have in-house or consultant expertise on board to manage the

building's energy consumption, these individuals are often most concerned with keeping things operational

rather than attempting more complex improvements. Improvements and changes must therefore be proposed in

terms of improving the bottom line by reducing operating costs, improving performance or productivity, or

some other value that is in line with business considerations.

This program therefore targets those business customers who know what they want done (e.g., boiler

replacement), but need some incentive to make the best available equipment choice for reducing gas use. Those

unsure of what opportunities exist may elect to have a Targeted Technical Assessment performed through the

program. Those that need more customized information on their building and process systems would apply

under the Custom Program, discussed in the next section.

Building commissioning refers to a process that takes place when a facility is newly constructed, and it involves

checking that all of the installed building systems are working properly and at design efficiencies. Like

automobiles, however, buildings can change soon after they are occupied and operating. Changes in scheduled

occupancy, building use, additions and renovations or deployment of new equipment can all greatly alter the

effective operations of building systems. Re-commissioning is like a building tune-up and it is aimed at

correcting for any building changes that have affected the major building systems. Experience has shown that

periodic re-commissioning can often contribute significant savings in both gas and other fuel savings.

Building control systems are often associated with electric savings, but they can have a significant positive

effect on natural gas consumption as well. Any time electronic controls are installed they can be used to

improve the operation of any energy related building system and thus produce energy saving benefits.21

21 Due to the dual benefits of building controls, a partnership may be sought with the affected electric utilities to share some costs of this program.


Vectren DSMAction Plan: Final Report VI. DSM Program Plans

Program Participation and Savings Estimated participation and savings over a five-year period is presented in Table 41 below. Around 1 1,000

large commercial buildings are estimated to be potential participants in this program. After five years, nearly

800, or seven percent of potential participants, are expected to be treated by the program. Each participant is

expected to save an average of just under 2,000 therms a year for a total resource of 1.5 million therms after five

years of program operation.

Table 41. Estimated Participation and Savings - General Services Energy Efficiency Program

Marketing Plans Many business owners and building operators are reluctant to consider non-critical investments, taking the

approach of "If it isn't broken, why fix it?" They tend to be risk-averse and are generally wary of new and

emerging technologies, particularly when standard options are less expensive and considered to have more

predictable, reliable performance. Even those who keep up with the latest technology options have difficulty

justifying retiring a piece of equipment before it has failed or the incremental cost of replacing it with a higher

efficiency option. This program would overcome the primary market barrier of cost by subsidizing a portion of

the increased cost to bring retrofit equipment more in line with standard, less efficient options.



1. Vectren administrative costs to develop, advertise, oversee and monitor the program 2. A vendor contract to market and deliver Targeted Technical Assessments to customers, usually charged

on a square footage basis plus a management fee 3. A per audit charge subsidized by Vectren 4. Incentives for the installation of recommended measures as demonstrated through the provision of

receipts by the customer or the job scope if a Targeted Technical Assessment is performed


1. Customer's share of the assessment fee 2. Customer's share of the costs of covered measures and equipment 3. Installation costs

- --

Prepared by Forefront Economics and H. Gil Peach & Associates page=


Table 42. Estimated Five-Year Program Budget - General Services Energy Efficiency Program

Cost per 5-Year Pct of Part. Year 1 Year 2 Year3 Year 4 Year 5 Total Total

Fixed Program Costs

Start Up Costs $25,000 $25,000 1.1% Staffing, Admin. and Overhead $51,938 $51,938 $5 1,938 $51,938 $51,938 $259,690 11.3% General Public Education $70,549 $70,549 $70,549 $70,549 $70,549 $352,745 15.4%

Total Fixed $147,487 $122,487 $122487 $122,487 $122,487 $637,435 27.8%

Variable Program Costs

Incentives $2,000.00 $1 10,000 $220,000 $330,000 $440,000 $440,000 $1,540,000 67.1%

Program Expenses $0.00 $0 $0 $0 $0 $0 $0 0.0% Monitoring and Evaluation $151.92 $8,356 $16,711 $25,067 $33,423 $33,423 $1 16,980 5.1%

Total Variable $2,151.92 $118,356 $236,711 $355,067 $473,423 $473,423 $1,656,980 72.2%

Total Budget $265,843 $359,198 $477,554 $595,910 $595,910 $2,294,415 100.0%

Program 3. Customized Energy Efficiency Program

The Customized Energy Efficiency Assessment Program is for large buildings and industrial facilities with

complex features or operations that require assistance to identify energy saving options. This program would

serve existing large commercial and industrial facilities with customized information for making improvements

to their gas end-use operations. Its main feature is a Targeted Technical Assessment conducted by a qualified

engineering fm, under contract to Vectren, with the customer's portion of the costs reimbursed by Vectren if

they proceed with the recommendation^.^^ Participants may also take advantage of the prescriptive equipment

rebates offered under Program 2 for upgrading heating, water heating and gas cooling systems; boiler

replacement, water heating equipment, tune ups and control systems.

Table 43. Incentives - Customized Energy Efficiency Program

Incentives are Calculated as the Lesser of: Buydown to a two-year payback $0.70 per them saved 50% of incremental cost

This program is modeled after the NYSERDA Flexible Technical Assistance Program (Flex-Tech) cited in

ACEEE's report on best natural gas energy efficiency programs.23 Incentive levels are based on People's

Natural Gas CA Custom Rebate

22 Under this program, all participants would receive the Targeted Technical Assessment as a precondition of receiving incentives. 23 ACEEE, U035, Dec. 2003. 24 People's Natural Gas Conservation Improvement Program filing (June 1998).


Vectren DSM Action Plan: Final Report W. DSM Program Plans

Rational for Program While other programs may have some participation from industrial accounts, the Customized Energy Efficiency

program is the only funded program in the DSM Action Plan targeted to the industrial segment. This program

accounts for 5 percent of the total DSM program budget. Large volume customers have unique and individual

needs when it comes to energy efficiency opportunities. Yet, engineering assessments of building systems for

identifying those opportunities can be very costly, and as a result, cost-conscious building and facilities

managers are sometimes reluctant to be comprehensive in considering building improvements. Increasing costs

of gasmight help open this market, but help is often still needed in identifying and prioritizing energy efficiency

improvements. By subsidizing the costs of energy assessments and sharing the cost with customers, Vectren can

help them identify a broader range of improvements that can help customers reduced their gas bills. The

Technical Assessment can also serve as the entry point for the General Services Energy Efficiency Program by

having the contractor complete an application form for any equipment qualifying for incentives. For equipment

or improvements not specifically identified in Program 2, a customized rebate can be calculated based on

anticipated therms saved.

Some programs contract with one firm to conduct Technical Assessment studies, while others have several firms

each with a different specialty. Some programs prevent the firms doing the assessments from also bidding to

install equipment, and others do not.

Program Participation and Savings Estimated participation and savings over a five-year period is presented in Table 44 below. This program is

characterized by relatively few potential participants (350), but with large savings per participant, nearly 1 1,000

therms annually. Although the number of potential participants in this program is based on the number in the

process planning element (large industrial transport customers), other industrial customers with process loads

may be good candidates for this program. Only a few participants are expected a year, so that after five years of

program operation, fewer than three dozen firms have taken advantage of this program. By then the program is

expected to deliver nearly 371,000 therms of resource annually.

Table 44. Estimated Participation and Savings - Customized Energy Efficiency Program


Vecti-en DSM Action Plan: Final Report m. DSM Program Plans

Marketing Plans Larger facilities are often too complex to immediately reveal what opportunities exist to reduce energy

consumption and costs. The first step then is in convincing business energy managers, building operators and

manufacturing managers that upgrading equipment to high energy efficient models is a good business decision.

This involves conducting a technical assessment of the facility, identifying opportunities specific to the

individual customer's operations, and calculating the economics of the investment to be made against savings in

energy costs and perhaps maintenance down the road.

The specific product of the Technical Assessment service is a detailed specification and economic analysis of

recommendations that could then be put out for bid to qualified contractors, or implemented in-house. Once

work is completed, Vectren's engineering fm that did the Technical Assessment would conduct an inspection

to verify what work was completed, collect documentation as to the equipment installed, and calculate any

incentives due to the customer from Vectren (refunded Technical Assistance fee plus rebates).

Other gas that offer customized services use the following types of criteria for determining rebate

amounts:

1. Technical Assessment studies are covered up to $2,500 or not to exceed 50 percent of total cost, with an additional $2,500 reimbursed if measures are installed

2. All projects must pass a TRC benefithost test with a result greater than 1.0 3. The incentive buys down the costs of installation to a two-year payback 4. Rebate amount is 50 percent of the incremental cost, or 25 percent of equipment cost 5. Rebate is calculated on a custom basis with each installation passing a costhenefit test 6. Rebate level is set at some percentage of cost up to a maximum level

All work for this program would be handled through contracts with one or more engineering f m s . Vectren

would hire a qualified engineering f m ( s ) to offer Technical Assessment services to comrnerciaVindustria1

customers and perform inspections of completed projects. The contractor would also be trained in helping

customers apply for the incentives under the prescriptive program, Program 2. The contractor would present the

customer with a standardized report identifying the opportunities, recommended equipment and technical

upgrades, and qualification for rebates, if any. The contractor would be responsible for marketing the program

to Vectren's large commerciaVindustrial customers.

While it is beyond the scope of this project to identify all possible audit firms, there are likely to be several

possibilities available. Not withstanding the use of other audit providers, we recommend that the Indiana Clean

Manufacturing and Safe Materials Institute (CMTI) be recruited as a partner in the marketing and

implementation of this program.26 CMTI has valuable experience assisting Indiana industries with the adoption

25 Center Point Energy Minnegasco, Xcel Energy, Aquila. 26 CMTI is affiliated with Purdue University and is located in West Lafayette, Indiana. Lynn Corson, Director, can be reached by phone at (765) 463-4749.


Vectren DSMAction Plan: Final Report T/I; DSM Program Plans

of manufacturing processes that are less harmful to the environment, some with energy saving benefits. The

advantages of partnering with CMTI for delivery of the Customized Energy Eff~cient Program include their:

familiarity with Indiana industry with over 10 years of experience delivering technical-based solutions to manufactures,

familiarity with existing state and federal programs, including energy programs from the US DOE, for maximum leverage of program funds with other funding sources,

experience with translating technical recommendations to standard business economics such as payback periods and benefit cost ratios, and

ability to target selected industries for adoption of proven methods and practices.

CMTI would likely need to invest some resources before they were ready to market and provide technical

services as part of this program. However, other potential service providers would likely require even greater

investment and would not be in a position to capitalize on the benefits listed above.








Table 45. Estimated Five-Year Program Budget - Customized Energy Efficiency Program


I

Cost per 5-Year Pct of Part. Year 1 Year2 Year3 Year4 Year5 Total Total

F h d Program Costs

Start Up Costs $25,000 $25,000 2.8% Staffing, Admin. and Overhead $12,511 $12,511 $12,511 $12,511 $12,511 $62,555 6.9% General Public Education $16,995 $16,995 $16,995 $16,995 $16,995 $84,975 9.4%

Total Fixed $54,506 $29,506 $29,506 $29,506 $29,506 $172,530 19.2%


Incentives $20,000.00 $80,000 $100,000 $140,000 $180,000 $180,000 $680,000 75.5%

Program Expenses $0.00 $0 $0 $0 $0 $0 $0 0.0% Monitoring and Evaluation $1,419.59 $5,678 $7,098 $9,937 $12,776 '$12,776 $48,266 5.4%

Total Variable $21,419.59 $85,678 $107,098 $149,937 $192,776 $192,776 $728,266 80.8%

Total Budget $140,184 $136,604 $179,443 $222,282 $222,282 $900,796 100.0%

Vectren DSMAction Plan: Final Report M. DSM Program Plans

Program 4. Hospitality Industry Energy Efficiency Program

This program is targeted to restaurants, bakeries, institutional housing (nursing homes, colleges, schools), hotels,

hospitality facilities and other cooking facilities that employ natural gas for cooking and food preparation. The

program promotes installation of energy efficient booster water heaters, dishwashers, fryers, griddles and gas

ranges to replace aging equipment in existing facilities and/or as the efficient option for new facilities.

Incentives cover the incremental added cost of the equipment and installations. As with the General Services

Energy Efficiency Program, an optional Targeted Technical Assessment service would be available on a cost-

shared basis from trained contractors for those customers wishing to have a professional assessment of energy

savings opportunities performed on their facility before making decisions. The suggested incentive levels for

selected measures are listed in the table below.

Table 46. Measures and Incentives - Hospitality Industry Energy Efficiency

11 9. Infrared charbroiler $200

This program is modeled after a similar offering of the People's Natural Gas (CA Food Services ~ r o g r a m ) ~ ~ and

Centerpoint Energy Minnegasco's Foodservice Equipment

Rational for Program This program promotes energy efficiency in the commerciaVindustria1 sector's food services and related

facilities (e.g., nursing homes) by encouraging customers to install the equipment listed above. Other utility

food service programs have shown that the key barriers to energy efficiency for this sector are inadequate

information and lack of subsidies or rebates to help reduce first-costs. This equipment has the potential to

provide significant operating cost savings to customers, both existing and new construction. The rebates

provided will help customers overcome first-cost barriers to implementation.

Program Participation and Savings Estimated participation and savings over a five-year period is presented in Table 47 below. There are an

estimated 3,100 potential participants for this program. This estimate is based on NAICS codes in Vectren

27 Based on PNG. 28 Same as residential/small building equipment. 29 PNG (1998). 30 See Appendix E.


Vectren DSMAction Plan: Final Report F.?; DSM Program Plans

North CIS records. Nearly 300 of these customers are expected to participate in this program after five years of

program operation. Average annual savings ofjust over 1,500 therms per participant are expected from this

program.

Table 47. Estimated Participation and Savings - Hospitality Industry Energy Efficiency Program

Marketing Plans This program would be promoted via direct mail and other contacts to qualifying food service operations,

businesses and institutions. Educational material explaining the benefits of the equipment retrofits and

incentives offered will be provided. Some customers may avail themselves of incentives through the

Customized Energy Efficiency Program. Trade allies (restaurant equipment providers, distributors and

installers) will also be notified about the program to assist in marketing it to their customers. Participation and

energy savings information will be obtained from customer applications and receipts and invoices for installed

equipment that shows manufacturer ratings and consumption data.









Vectren DSMAction Plan: Final Re~ort M. DSM Prowam Plans

Table 48. Estimated Five-Year Program Budget - Hospitality Industry Energy Efficiency Program

Cost/ 5-Year Pct of Part. Year 1 Year2 Year 3 Year 4 Year 5 Total Total

Fiwed Program Costs

Start Up Costs $25,000 $25,000 2.2% Staffing, Admin. and Overhead $1 5,409 $15,409 $15,409 $15,409 $15,409 $77,045 6.7% General Public Education $20,93 1 $20,93 1 $20,93 1 $20,931 $20,93 1 $104,655 9.0%

Total Fixed $61,340 $36,340 $36,340 $36,340 $36,340 $206,700 17.9%


Incentives $3,000.00 $93,000 $141,000 $186,000 $234,000 $234,000 $888,000 76.7%

Program Expenses $0.00 $0 $0 $0 $0 $0 $0 0.0% Monitoring and Evaluation $21 1.92 $6,569 $9,960 $13,139 $16,530 $16,530 $62,728 5.4%

Total Variable $3f 11.92 $99,569 $150,960 $199,139 $250,530 $250,530 $950,728 82.1%

Total Budget $160,909 $187,300 $235,479 $286,870 $286,870 $1,157,428 100.0%

Program 5. Multi-Family Building Energy Efficiency Program

This program would serve existing five or more unit multi-family buildings with prescriptive equipment rebates


ups and control systems. Building occupants and tenants will be given a package of free low-cost

weatherization measures for self-installation. The suggested incentive levels for selected measures are listed in

the table below.

Table 49. Measures and Incentives - Multi-Family Building Energy Efficiency Program

Rational for Program Residents of multi-family buildings are often more vulnerable to energy price increases than the general

population, as heating costs can represent a higher proportion of living expenses, whether included in rent or

paid directly to a utility. This program targets landlords of multi-family buildings with incentives to upgrade

common equipment such as heating systems and water heaters or gas cooling. It also provides assistance to

tenants to give them a direct way to reduce energy costs in their individual units with a package of low-cost

measures; such as, window film, caulk and weather stripping, and hot water pipe wraps.

Measures 1. Water heaters-50 gal or more 2. Boilers-various sizes and types 3. Boiler tune ups and controls 4. Gas cooling systems 5. High efficiency forced air furnace-92% AFUE 6 . Weatherization measures


Incentive Amounts $150

$750 to $7,500 $250 up to $5,000

$6 per ton $100

varies

Vectren DSMAction Plan: Final Report n. DSM Program Plans

Program Participation and Savings Estimated participation and savings over a five-year period is presented in Table 50 below. The unit of

participation in this program is a building with five or more living units. We estimated potential participants by

starting with the 36,000 total multi-family units from the Market Assessment section. From there we arrive at

our estimate of 2,160 potential buildings by applying assumptions that are based loosely on Census data. We

assume that 60 percent of multi-family units are in buildings of five or more units and that these buildings

average ten units per building (2,160 = (36,000 * 0.6 ) / 10)). Annual savings per participant is expected to

average nearly 1,400 therms per building. After five years the program is expected to deliver annual savings of

272,000 therms.

Table 50. Estimated Participation and Savings - Multi-Family Building Energy Efficiency Program

Marketing Plans Landlords are typically slow to adopt above-standard equipment, unless economic pressures caused by such

situations as increased energy prices make their buildings less competitive, with tenants less able to afford rents

or utility bills. Too many business owners and building operators are reluctant to consider non-critical

investments, taking the approach of "If it isn't broken, why fix it?" They tend to be risk-averse and are

generally wary of new and emerging technologies, particularly when standard options are less expensive and

considered to have more predictable, reliable performance. Even those who keep up with the latest technology

options have difficulty justifying retiring a piece of equipment before it has failed or the incremental cost of

replacing it with a higher efficiency option. This program would overcome the primary market barrier of cost

by subsidizing a portion of the increased cost to bring retrofit equipment more in line with standard, less

efficient options.

Detailed Budget Plans An estimated five-year budget for this program is provided below in Table 5 1. The anticipated cost to Vectren


1. Vectren administrative costs to develop, advertise, oversee and monitor the program 2. A vendor contract to market and deliver building assessments to landlords, plus a management fee 3. A per audit charge subsidized by Vectren 4. Incentives for the installation of recommended measures as demonstrated through the provision of

receipts by the customer or the job of an assessment

-- -


Vectren DSMAction Plan: Final Re~ort W. DSM -ram Plans


1. Landlord's share of the assessment fee 2. Landlord's share of the costs of covered measures and equipment 3. Installation costs

Table 51. Estimated Five-Year Program Budget - Multi-Family Building Energy Efficiency Program

Cost per 5-Year Pct of Part. Year 1 Year2 Year3 Year 4 Year 5 Total Total

Fixed Program Costs


Total Fixed %46,659 $21,659 $21,659 $21,659 $21,659 $133,295 20.0%


Incentives $2,500.00 $80,000 $95,000 $107,500 $107,500 $107,500 $497,500 74.7%

Program Expenses $0.00 $0 $0 $0 $0 $0 $0 0.0% Monitoring and Evaluation $177.50 $5,680 $6,745 $7,632 $7,632 $7,632 $35,322 5.3%

Total Variable $2,677.50 $85,680 $101,745 $115,132 $115,132 $115,132 $532,822 80.0%

Total Budget $132339 $123,404 $136,791 $136,791 $136,791 $666,117 100.0%

Program 6. Innovative Energy Efficiency Technologies Research and Demonstration Program

This program would provide funding to the Indiana Clean Manufacturing Technology and Safe Materials

Institute (CMTI) for conducting research into emerging gas technologies that contribute to increased energy

efficiency in industrial applications. There are no direct incentives to customers under this program, rather it

provides a funding stream to support research into technologies that may be added to the portfolio of programs

offered by Vectren in the future. Case studies would be supported at some customer sites on a case-by-case

basis, such that field demonstrations could be performed and studied for potential future market application.

Rational for Program This program is targeted to industrial and other large gas users who already tend to be early adopters of cost

effective technologies for their business operations. These customers would have less of a need for the other

DSM programs targeted at the large building sector because they have already taken steps to maximize their

buildings' energy efficiency. Even so, this target market segment is typically staying on the cutting edge of new

developments that can help them stay competitive. This program is therefore targeted at those interested in

learning about emerging technologies that may be developed through research and demonstration. The Institute

identified to receive funding from this initiative conducts research into new products and processes for

increasing the competitiveness of Indiana's industries. While not limited to technologies that use natural gas,


Vecfren DSMAction Plan: Final Report V1: DSM Program Plans

these research projects and demonstrations invariably have an energy component, and thus strong implications

for industry's bottom line. By providing direct funding to the Institute for specific gas-related projects and those

that have gas efficiency as a secondary benefit, Vectren will ensure that its most innovative large customers will

benefit fiom its DSM program portfolio.

An advisory committee or other forum may be established, consisting of a representative group of these

customers to help provide suggestions for research, participate in demonstrations of new products, and review

the results of program-funded projects. Annual funding request from CMTI would be reviewed annually to

confirm adequacy of subsequent year's funding levels. Research would be monitored by the ~dvisory

Committee to assess whether funding is being adequately supported through research into natural gas

technologies and related projects.

Program Participation and Savings Not applicable.

Cost effectiveness Ratios/Rating per Individual Program Not applicable.

Marketing Plans Not applicable.

Program 7. Energy Efficient Builder Program

This program would promote the incorporation of high efficiency design features in new homes, plus installation

of high efficiency equipment above standard appliances, furnaces and windows. The program would be targeted

at builders of subdivision and track homes. The suggested incentive levels for selected measures are listed in

the table below.

Table 52. Measures and Incentives - Energy Efficient Builder Program

Rational for Program New construction is the best and most cost effective opportunity for incorporating major energy efficiency

measures and construction practices in homes. An incentive is paid to builders for each unit that meets an

energy savings criteria above standard design of 300 therms. The goal of the program is to ultimately move the

new housing market toward a high efficiency standard practice without the need for incentives in the future.


Vectren DSM Action Plan: Final Report W. DSM Program Plans

Program Participation and Savings Estimated participation and savings over a five-year period is presented in Table 53 below. Based on Vectren

North's new residential gas connects, this program would be available to an estimated 13,000 units annually.

Over the first five years of the program, about 550 homes are expected to be built under this program (0.9

percen.t of new construction). After five years of operation the program is expected to deliver 214,000 therms of

savings annually.

Table 53. Estimated Participation and Savings - Energy Efficient Builder Program

Marketing Plans This program would target large-volume builders in the service territory with education, training and incentives

for adopting Energy STAR Home standards in a proportion of the homes they build.



1. Vectren administrative costs to develop, advertise, oversee and monitor the program 2. A vendor contract to market and deliver builder training and plans review 3. Incentives to builders for each home built that meets 300 them energy savings for design features

incorporated

Costs to participants would be at the discretion of the builders, and may include:

1. Customer's share of the costs of covered measures and equipment 2. Installation costs


Vectren DSMAction Plan: Final Report V7. DSM Program Plans

Table 54. Estimated Five-Year Program Budget - Energy Efficient Builder Program

Cost per 5-Year Pct of Part. Year 1 Year2 Year3 Year 4 Year 5 Total Total

Piwd Program Costs


Total Fixed $42,039 $17,039 $17,039 $17,039 $17,039 $110,195 8.9%


Incentives $1,920.00 $124,800 $1 88,160 $249,600 $249,600 $249,600 $1,061,760 85.4%

Program Expenses $0.00 $0 $0 $0 $0 $0 $0 0.0% Monitoring and Evaluation $129.05 $8,388 $12,647 $16,776 $16,776 $16,776 $71,363 5.7%

Total Variable $2,049.05 $133,188 $200,807 $266,376 $266,376 $266,376 $1,133,123 91.1%

Total Budget $175,227 $217,846 $283,415 $283,415 $283,415 $1,243,318 100.0%

Program 8. New Program Development and Regulatory Affairs

This program is a support program; it does not deliver direct energy savings, but rather serves as a budget line

item to allow for new project developments and coordinated interaction with regulatory and legislative bodies

for development of policies supportive of DSM activities (e.g., improved energy efficiency building codes and

standards).

Two additional program areas are the corn stove promotion and the "real" programmable thermostat

demonstration. Corn stoves are pellet stoves and approved corn stoves have federal air quality waivers because

they have virtually no emissions. If there is any thought that supply problems will lead to the kind of price

increase projected by the EIA for the coming winter or to eventual problems of physical supply at any price,

then it makes sense to provide a small discount program for corn stoves. The advantage of the corn stove is that

it burns corn pellets, a "green" fuel that is sustainable from the service territory; also, that the household stoves

can provide a warm room should there be security problems with either the gas or the electric system. This is

not put forward as a fuel replacement program, but as a fuel diversity program so that, should it be necessary to

ration fuel, flexibility will have been established in the home heating end-use so that rationing will have a

mitigated impact in heating and health.

A small demonstration program is proposed, working with CAP agencies, to design a real programmable

thermostat. There is a need for a programmable thermostat that can be easily read in a darkened room or

hallway and that is easy to operate. While there is a range of products available on the market, they do not meet

these two criteria and most programmable thermostats are as friendly to program as a VCR. Development of a

real programmable thermostat would enable the benefits inherent in the product concept to be actually achieved.



Rational for Program It is critical to fund an ongoing campaign to educate legislators and regulatory staffs about policy tools that are

necessary to support DSM programs. A Vectren representative would play the role of corporate liaison to the

regulatory and legislative community to share Vectren plans as well as communicate critical information back to

Vectren's DSM team regarding state and federal programs and positions regarding the natural gas energy

savings activities being pursued by Vectren.

Program 9. Public Education and Outreach Program

This program would provide funding for cross-program public education activities to raise awareness of the

benefits and methods of improving energy efficiency in homes and b~sinesses .~~ DSM portfolios in other

jurisdictions often include significant public education programs to support the direct energy savings programs,

as well as, encourage consumers to take actions on their own to increase energy efficiency. The effects of

education and outreach programs are difficult to measure, even though program evaluations consistently show

that people who go on to participate in programs have increased awareness of energy efficiency options and

often sign up for programs in some measure because of the educational outreach activities. Non-participant

surveys also show increased awareness of energy efficiency options due to public education campaigns, with

some of these customers taking actions on their own.

This program is not subject to cost effectiveness screening, which is deemed inappropriate according to the

California Standard Practice Manual:

"For generalized information programs (e.g., when custonzers are provided generic information on means of reducing utility bills without the benefit of on-site evaluations or customer billing data), cost eflectiveness tests are not expected because of the extreme dzfJiculty in establishing nzeaninsfirl estimates of load impacts. "

Types of activities that would be included in this program are:

General mass media campaign for the public on pending gas price increases and ways to help control utility bills through energy efficiency measures and actions

Development of (update of the) Vectren North website to include the latest energy efficiency information for commercial, residential and school use

Targeted educational campaign for businesses

Targeted training and educational program for trade allies

Distribution of federal ENERGY STAR and other national organization materials in the service territory

A schools curriculum program to educate teachers and direct students to available educational materials on the Web about energy efficiency opportunities

31 Note that it is assumed that the individual program budgets include funds for development of program-specific advertising materials, such as brochures; this program covers public awareness and education campaign materials that would address all programs within a market sector. The funding is therefore not duplicative.

Prepared by Forefront Economics and H. Gil Peach h Associates Page 80

Vectren DSMAction Plan: Final Report M. DSM Program Plans

The schools program component would proceed as follows:

Develop an energy education outreach program targeting Vectren North service territory schools K-12.

Provide energy curricula to schools that teach students the fundamentals of energy and how to change behavior to conserve.

Secure consultant services to provide teacher training and classroom materials.

Provide teacher training.

Rational for Program The energy efficiency market is made up primarily of private sector activities, which can be significantly

influenced by public sector actions. The key to greater energy efficiency is convincing the families and

businesses making housing, appliance and equipment purchases to opt for greater energy efficiency. The first

step in convincing the public and businesses is to raise their energy efficiency awareness. These program

elements are designed to work in tandem to increase the public's understanding of the benefits to them and

society created through greater energy efficiency.

Although it is likely that customers within Vectren North's service territory have been exposed to some public

education material from electric utilities, there is a need for a major outreach initiative on gas energy savings

opportunities. This is particularly so because of media warnings of significant increases in gas prices which are

anticipated for this winter (2005 -2006) due to the effects of hurricanes Katrina and Rita. More directly to the

point of this Plan, since Vectren's programs will all be new to the marketplace, it is imperative that a broad

public education and outreach campaign be launched to not only raise awareness of what consumers can do to

save energy and control their energy bills, but to prime them for participating in the various DSM program

offerings that will be implemented over the next several months following regulatory approval. Without a

significant public outreach campaign, it would be difficult to achieve the levels of participation represented in

this Plan as reasonable targets for the programs.

Program Participation and Savings This program would address markets by sector-general public, businesses and institutions, trade allies and

school children and teachers. There would be no "participants" per se, although for direct contact activities,

feedback forms and other means of identifying those exposed to the educational materials can be developed.

Marketing Plans Not applicable.

Detailed Budget Plans The various educational program elements are adapted from the successful New York program, which is carried

out in partnership with the federal Energy Star Program. The general public education or Awareness-Raising

Program will use the Energy Star ratings as a platform for its "buy energy efficient appliances" message. A

breakdown of budgetary items for the program elements described is shown in Table 55 below. The budget



item and amounts should be used to generate ideas for implementation. We would expect budget allocation

decisions between media channels and specific media buys to be best made by program implementers.

Flexibility is also required on the timing of education expenses. It may be desirable, for example, to front load

spending in the early years of program implementation. Accordingly, the budget figures in Table 55 are for the

full five-year period rather than try to estimate the timing of expenses.

Table 55. Public Education Budget Items and Amounts

I Total W e a r Budget: 1 $2,445,000 1

Performance Tracking General public awareness questions will be added to ongoing corporate satisfaction surveys (typically conducted

by Customer Service staffs at most utilities).


Vectren DSMAction Plan.: Final Report W. Program Cost Effectiveness

VII. PROGRAM COST EFFECTIVENESS

Program cost effectiveness analysis answers the question of would we be better off with the DSM program

compared to not having the program. The answer almost always depends on who is asking the question. In

other words, better off from whose perspective? Standard DSM cost effectiveness analysis includes four

perspectives that will be addressed in this report:

Total Resource Cost (TRC)

Participant

Ratepayer Impact (RIM)

Administrators Cost (formerly named Utility Cost)

A detailed discussion of cost effectiveness methodology, including the four standard tests listed above, is

included in Appendix D. In this section we present the results of the cost effectiveness analysis beginning with

a summary of total budget and them savings across all programs followed by a discussion of avoided gas costs.

Cost effectiveness results are then presented for each perspective and DSM program.

Expected Program Costs

The total program budget and cumulative them savings over the first five years of program activity is shown in

Table 56 below. We recommend a minimum of five years for program implementation and tuning for

maximum effectiveness.

Table 56. Total Program Budget

Year 4 $3,105 $596 $222 $287 $137 $283 $4,630 $8.12 0.6%

Year 5 $3,726 $596 $222 $287 $137 $283 $5,25 1 $9.21 0.7%

Total $12,446 $2,294 $901 $1,157 $666 $1,243 $18,708 $32.82

Program budgets include fixed costs for fully loaded program staff and expenditures for general public

education and awareness. Public education spending is discussed in the Program Plans section. Staffing

assumptions to administer the collective bundle of programs are listed in the table below. Staffing and public

education expenditures have been allocated back to each program based on the distribution of cumulative

savings. Total program cost is shown in the table below.


Vectren DSMAction Plan: Final Report Vn. Program Cost Effectiveness

Table 57. Program Staff~ng Assumptions

The program budget presented above includes all fixed and variable expenses paid by the program

administrator. The last two columns of Table 56 shows total DSM spending per Vectren customer and as a

percent of 2004 revenue, respectively. Spending ramps up gradually throughout the five-year implementation

Staffing Analyst and Support Staff Managerial Staff Total Staffing

period, reaching a maximum of over nine dollars a customer and 0.7 percent of total revenue in year five. About

two-thirds of the spending and 73 percent of the savings is expected to be associated with the Small Building

program, primarily residential and small commercial space and water heat measures.

It is important to understand that actual expenditures will vary from planned expenditures in their timing and

distribution between specific DSM programs. For this reason it is important for the program administrator to

have flexibility in the administration of DSM program funding without having to obtain approval from the

Public Utility Commission. We recommend that flexibility include the following, with each action subject to

FTE 3 .O 1 .O 4.0

review and approval by the Advisory Board:

1. Roll over unspent funds within program budgets at end of year to categories within the same program in the next year

2. Reallocate program funds across line items within a program 3. Shift up to 25 percent of total budget among approved programs at any time within a program year

Fully Loaded Salary $80,000

$120,000

Having some flexibility in the administration of program funding will assist in the management of programs and

enable staff to fine tune efforts for maximum resource effectiveness.

Cost $240,000 $120,000 $360,000

Expected Program Savings

Them savings expected from the program are based on the designs and assumptions presented earlier in this

report. Key assumptions affecting the annual savings and program cost effectiveness are shown in the table

below.

Table 58. Summary of Program Assumptions


Per Participant: Savings (therms) Installed cost Incentive Program costs

Savings Life (years) Net to Gross Ratio

Small Building

198 $799 $150 $86 16.4 0.90

G/S EE

1,998 $10,050 $2,000

$152 15.3 0.90

Customized

10,900 $4 1,000 $20,000 $1,420

14.5 0.70

Hospitality

1,542 $6,225 $3,000

$212 13.5 0.70

MF Buildings

1,3 67 $5,735 $2,500

$177 12.1 0.70

EEHome Builders

3 87 $1,920 $1,920

$129 30.0 0.90

Vectren DSMAction Plan: Final Report r/u. Program Cost Effectiveness

Most of the items listed in Table 58 were addressed in the Program Plans section. The savings life is calculated

from the life of individual measures weighted by program savings and represents the duration of energy savings

flowing from a participant in the program. The net-to-gross ratio captures the effect of free riders, participants

in the program who would have installed the energy efficient measures without the program. Higher ratios

imply a lower rate of free riders in the program.

Annual them savings across all programs are shown in the table below. Cumulative program activity is

expected to result in nearly 11 million therms of annual savings. This represents approximately 1.4 percent of

total therms delivered.

Table 59. Total Program Savings

Avoided Gas Costs

The avoided or marginal cost associated with a reduction in gas loads is of primary importance when evaluating

the cost effectiveness of demand side management programs. These costs represent the value of an avoided

them of gas. Vectren's avoided costs are the reduction in the cost of supplying gas compared to what they

would have been without the reduction in loads and include all incremental commodity, transmission, storage

and distribution costs. Ideally, avoided costs are determined using a mathematical optimization approach that

considers alternative supply side resources over time. The amount and timing of new investments in pipeline

capacity and storage, for example, are identified in an avoided cost study.

Although Vectren North does not have are avoided cost study, they do make quarterly Gas Cost Adjustment

(GCA) filings. These filings detail the anticipated demand and commodity costs expected in the quarter ahead.

Demand costs are the fixed investments in infrastructure capacity, including transportation, storage and

distribution facilities, required to meet anticipated demand. The table below shows capacity and commodity

expenses per them from a recent GCA filing (Cause No. 37394-GCA88).


Predominant Load Shape Space Heat (RS 2 10/220/229) Non-Space Heat (RS 240)

Commodity 1.1568 1.1568

Demand 0.1 190 0.0476

Total 1.2758 1.2044

Vectren DSM Action Plan: Final Report MI. Program Cost Effectiveness

Demand costs per them are dependent on the nature of the load served. System coincident peeking loads, such

as space heating, have greater demand costs per them served than non-seasonal loads, such as water heating.

This relationship is reflected in Vectren's GCA filings, as shown in the table above. The demand costs per

therm of space heat dominated loads are nearly twice as high as less seasonal loads. Although demand costs in

the GCA filings are prepared by rate schedule, we will apply the demand costs according to predominant load

shape affected by the energy efficiency program being assessed. A high efficiency residential water heater

program, for example, would be evaluated using the demand costs associated non-space heat loads (RS 240).

Because cost effectiveness analysis considers the impacts of programs over a planning horizon that can be as

long as 30 or 40 years, depending on the program, a forecast of demand and commodity cost must be developed.

Based on their GCA filings and conversations with staff, Vectren does not appear to be capacity constrained.

Demand costs are a relatively small percentage of the total cost of delivering a therm of gas to its customers.

Given the small percentage of total costs, lack of known capacity constraints, and no detailed avoided cost

study, we assume that demand costs increase over time at the same rate as general price inflation.

The long run outlook for commodity costs is one of the most important assumptions determining the cost

effectiveness of DSM programs. A number of factors, including supply disruption from Gulf hurricanes, have

caused natural gas prices to spike recently to record highs. Prices are likely to fall in the short-term as supply

rebounds and demand moderates in the face of record high utility bills. Our analysis is concerned with both the

short and long run, however, and fundamental supply and demand analysis suggest rising real prices for natural

gas in the years ahead.

Our commodity forecast is based on Vectren's current portfolio of gas supply as reflected in GCA 88 and a

melding of short and long-term price projections. We believe that Vectren's current portfolio of purchased gas

is the best measure of their current commodity costs. In the long run, we adjust the current commodity costs

using the percent change in natural gas prices forecast by the EIA in their 2005 outlook. The 2005 EIA forecast

was published in February 2005, prior to the hurricane induced supply disruptions in the summer of 2005.

Hence, we need a short-term price forecast in order to make the transition from current prices to our long term

forecast. We used the NYMEX futures contract for natural gas for the near-term percent change in price

forecast.

A detailed table showing the calculations of avoided energy costs by year is shown in Table 67 in Appendix D.

A summary of these calculations is shown in Table 60 below.


Vectren DSMAction PZan: Final Report W. Program Cost Effectiveness

Table 60. Real Levelized Avoided Cost per Them

Avoided costs are expressed in real levelized terms for the purposes of calculating the cost effectiveness of

DSM programs. Real levelized costs reflect the annualized value 0f.a them of gas over a specific period. Due

to the influence of very high commodity prices in the first few years of the forecast, real levelized costs are

highest for programs with a five-year life of anticipated savings. With time, commodity prices bottom out and

then rise again in real terms. This relationship is also born out in the table above.

Cost Effectiveness Results

In this section we present the findings of the cost effectiveness analysis which provides a systematic comparison

of the program benefits and costs discussed in previous sections. Results are shown from the four perspectives

mentioned at the beginning of this section. Net present value (NPV) and benefit-costs ratios are shown for all

perspectives. The third measure used to assess cost effectiveness differs by perspectives.

The TRC perspective is the broadest of the tests represented in Table 61. As the name implies, TRC shows the

total cost of the resource relative to supply side resources. The Participant Test shows the economics of

program participation from the participant's perspective and reflects benefits from lower bills and incentive

payments. Elements of program design, such as incentive payments, can greatly impact participant economics.

Since we assumed that fbture marginal cost of gas and gas rates were identical, all programs fail the RIM Test.

In other words, avoided energy costs are equally offset by lost revenue resulting in negative NPV when program

costs are positive. However, the life-cycle rate impact is small, only three-tenths of a cent per them increase in

rates.32 The Administrator's Cost Test reveals that when only costs paid by the program administrator are

considered, the cost of the acquired resource is quite small ranging from less than 20 cents a them for the Small

Buildings and General Services Energy Efficiency Programs to under 50 cents for the Energy Efficient Builder

Program.

32 It should also be pointed out that overall rates and marginal cost may be lowered as a result of a downward shift in demand curve due to DSM programs. For this shift to be apparent in commodity costs, DSM implementation would need to be significant and perhaps regional in scope.


Vectren DSM Action Plan: Final Report Program cost Effectiveness

Table 61. Cost Effectiveness Results by Program

An important finding is that all programs are cost effective from the TRC perspective and that rate impacts from

the programs are negligible. Overall, the programs generate over $37 million in NPV benefits. These results

are obtained using a base case avoided cost of gas forecast which, while based on published sources from

recognized industry experts, we believe is conservatively low.

Global Assumptions Certain global assumptions are required to calculate program cost effectiveness beyond those assumptions

already discussed. These assumptions are shown in Table 62.

All tests except the Participant Test use a nominal discount rate of 8.34 percent, Vectren's weighted cost of

capital. This translates to a real discount rate of 5.18 percent, assuming an inflation rate of 3 percent. The

participant discount rate is set higher reflecting the cost of consumer capital. Externalities are set to zero percent

meaning that no preferential treatment is given DSM resources over supply side options due to avoidance of

environmental impact of gas supply. The Societal Test, a variant of the TRC Test, is not used in this report.

The revenue requirements adder relates to the effects of rate basing utility expenditures. Revenue requirements

of 20 percent mean that for every dollar of utility program expense, $1.20 needs to be collected through rates.

System sales are used in life cycle rate impact calculations for the RIM Test.


Vectren DSMAction Plan: Final Report V17. Program Cost Effectiveness

Table 62. Global Assumptions Used in Cost Effectiveness Calculations

All forecasts contain risks, alternative scenarios that could lead to higher or lower numbers than projected. In

the case of projected gas prices, we believe the risk to this forecast is clearly and significantly on the high side.

If gas commodity costs move higher than projected, the benefits and cost effectiveness of the DSM programs

presented in this report will be even greater than expected. The next section explores this risk to the forecast in

greater detail.


Vectren DSMAction Plan: Final Report VI;L Alternative Forecast and Policy Parameters

VIII. ALTERNATIVE FORECAST AND POLICY PARAMETERS

Low-Income Area Policy Concerns

Members of our team have been doing low-income weatherization and payment assistance program evaluations

since 1988. At first, we did not look much at policy consequences in the low-income area, nor was that

requested as a part of the early evaluations. In the middle 1990's we did several low-income evaluations for gas

and electric utilities and could not help but learn how hard utility and Community Based Organization staffs

were working to make these projects succeed for the customers, yet customers were still having payment

troubles.

This experience and hundreds of interviews with low-income and payment-troubled customers of utilities in

different states and cities caused us to look systematically at the economic environment in which the programs

are operating. Over a number of years we learned that the programs are "swimming against the tide," no matter

how well the programs were implemented and maintained, a tide of economic adversity was gaining. As we

write in the fall of 2005, there is no question about a rampant increase in the need for substantially increased

assistance for low-income customers.

For example, as reported in Table 63, federal assistance, though very helpful, is not reliable and has deteriorated

to about 5 1 percent nationally (in real terms) of the funding provided in 1982, even though utility bills are

considerably higher than they were in 1982 (in real terms). Although the gap compared to 1982 funding has

narrowed in recent years, federal funds are insufficient at a time when we need them most.

Table 63. LI[HEAP, Valuable but Failing


Vectren DSM Action Plan: Final Report ?!iYI. Alternative Forecast and Policy Parameters

A new study from the Center on Budget and Policy Priorities indicates that, (using data from the US Department

of Energy) nationally home heating costs for LIHEAP beneficiaries will increase 47.5 percent between last

winter and the winter of 2006-2007.~~ This will be the largest single year increase since 1974. Figure 30 shows

the effects of this increase on current funding. An increase of 50 percent on a family with a $300 energy bill

would lead to a bill of $450. If that family received a $100 LMEA benefit, their share would increase from

$200 to $350, an increase of 75 percent. Simply providing a 50 percent increase in their LIHEA benefit (to

$150) would still increase their share of the bill to $300 or 50 percent. In fact, it shows that to completely

absorb the 50 percent increase in the household's energy bill, the LIHEA benefit amount would have to be

increased 150 percent. This example illustrates the difficulties low-income families are experiencing in paying

to heat their homes and the problems ahead.

Hypo%b&ticat 'Example: OM Hotlsehsld's. 14lmthly Heating Bill

LMEAPFmdng E.fHspfYmg says sL3JO: tocrease v& No gpmps~r W& lnse-

U H ~ P p;;~rs h ~ e a s e zn UHW B& so DO ~niqjb 61 tB Firml;l C&6 I@$ Covered Tc Falafd

3.t a-Ew Beietiaaties Hannlers

Figure 30. The Winter of 2005-2006 (from CBPP Study)

If we then look at what has happened to real income of low-income families in Indiana since around 1979, for

lower-income (bottom quintile) families with children, there has been about a 22 percent drop (Figure 3 1).

(This is conservative, since the last figure is for 1994-1995 prior to the end of the dot-com bubble, 9-1 1, and

other major economic changes.)

33 "Out in the Cold: How Much LIHEAP Funding Will Be Needed to Protect Beneficiaries fiom Rising Energy Prices?" Available from the Center for Budget and Policy Priorities website: http:Nwww.cbpp.orgllO-6-05bud.htm.


Vectren DSMActwn Plan: Final Report FUI. Alternative Forecast and Policy Parameters

Trends in Real Income: Indiana With Children

co m o .;- N m ~ - b m m r n m m m m

f % % 6 8 % 8 Z 2 % X 8 . - . - , - z z z c z z z z c z z z z z z Year

Figure 31. Trends in Real Income, Indiana Families with Children

If we put the first two bits of information together, we see a shortfall of 49 percent in LIHEAP that would

require funding at 204 percent of the 2004 level plus an energy shock of 47.5 percent in the coming winter

which raises the combined funding requirement to 422 percent of the 2004 level for the winter of 2005-2006 to

bring resources even in real terms to those provided by LIHEAP in 1982.

Then, if we plan low-income assistance to be able to serve Indiana low-income families with children, the

funding requirement to stay even with services in 1982 jumps to 541 percent of the 2004 LIHEAP allocation.

That federal failure has to be made up by the state and utilities.

The remaining adjustment required to fully understand the real situation we face would be the real change in

utility rates (1982 vs. 2004). We do not have those numbers to compare. Still, the need for additional funding

or a new payment assistance program format (such as, Nevada or New Jersey) is overwhelming. In addition, the

price effects of this year's hurricane damage to the US energy infrastructure are supposed to show up next year

(the winter of 2006-2007).

Overall Policy Concerns - An Alternative Forecast

There are more problems. Gas production has peaked in the US with only small increases expected in the years

ahead. Canadian gas imports, currently responsible for 10 to 15 percent of total US supply, are declining. EIA

projections assume a 13 percent annual growth rate in LNG imports. Due mainly to the rapid and sustained

increase in LNG imports, supply keeps pace with the 1.5 percent annual growth in demand projected by the EIA

through 2025. As a result, the base case forecast shows price declines through 201 1 and only modest increases


Vectren DSMAction Plan: Final Report WI. Alternative Forecast and Policy Parameters

after that. We feel that given the uncertainties embedded in the supply forecast, a realistic alternative with

substantially tighter supplies and higher prices should be considered for planning purposes.

With much electricity generated with gas, there is an interaction effect on electricity prices. Also, when natural

gas goes up dramatically, low-income and some moderate income customers have to shift to electric heat, one

way or another, and this shift can radically deplete utility payment assistance budgets, first of the gas utilities,

and then wipe out assistance budgets on the electric side. Compounding this problem is that US houses are built

for central heat; they can be modified to provide for area heating (such as, wood heat, coal grates or corn stoves)

but it will take intense effort and decades to make the necessary changes throughout the housing stock.

In addition, the consensus among scientists that global warming is real is very strong and that we are so far

down this path that, were the government to begin an emergency response now, it would take at least a hundred

or perhaps hundreds of years to restore normalcy.34 Global warming may also be associated with intensity of

hurricanes and the loss of coastal regions (in which much of US energy infrastructure is located) to the sea. This

has the potential to further threaten the supply outlook.

Further, federal policy changes to fast track siting of LNG facilities at the expense of long-standing democratic

decisions to protect natural areas have all of the characteristics of an emergency response to a serious energy

supply situation. Supply constraints, including what may be an overly ambitious EIA LNG forecast, do not

seem consistent with the declining price forecast embedded in the base-case scenario.

All of these factors point towards the series of benefits specified for the recommended DSM programs being far

too low. We put forward a main forecast implicitly accepting industry consensus because that is the standard

and proper practice for a DSM potential study and for development of program. We provide this alternative

forecast to promote critical thinking about fundamental assumptions and to make us all alert to continuing

changes in our energy environment. We need to focus continuing critical intelligences on these issues to ensure

optimal response and workable mid-course corrections.

Policy Recommendations 1. There is a pressing need for a substantial multiple of last year's payment assistance funds. Vectren should

continue to work through the American Gas Association (AGA) for additional federal funding. 2. Vectren should study the Nevada and New Jersey models for state assistance to complement LMEAP, to

develop recommendations to the Commission andlor the legislature for moving in those directions. 3. The solar option in the measures list (opting for legislative action to require solar site orientation and solar

access for new construction) is a very promising least-cost and high effectiveness pathway. This is a quiet program that requires much effort and agreement of many parties, but it is a very powerful solar program and available and least cost. In contrast to the other programs, this is really a policy program and Vectren should adopt it as a policy position and see what progress can be made in this area.

4. If there is concern that gas supply shortages may lead to triage (as is currently happening in the Southeast as industrial customers are curtailed on interruptible rates that were probably never expected to be

34 Welch, Craig, "Global Warming Hitting Northwest Hard, Researchers Warn," Seattle Times, Saturday, February 14, 2004. Also see, Luers, Amy Lind, "A Tale of Two Futures, California Feels the Heat," CataEyst, pp. 8-9, Fall 2004.


Vectten DSMAction Plan: Final Report MII. Alternative Forecast and Policy Parameters

interrupted), then a rationale outside the standard cost tests exists to develop "stand alone" energy systems for specific household end-uses.

This is a different strategy than current "mainline" DSM programs and would involve removing some end-uses

from grid and pipe dependence. For example, solar water heaters can be equipped with small photo-voltaic and

operate entirely off the grid. When the grid connection is down, the household system works independently and

the house has hot water because the system is not grid or gas supply connected.

Similarly, if there is a concern for grid failures or gas supply problems, corn stoves and similar solutions can

remove households from heating dependency and provide ability to maintain "one warm room." This may be

important if triage becomes necessary to keep the economy and essential infrastructure running or if there is

sabotage, weather damage, or an accident with the grid or pipes.

There are several ways to develop this approach. The important difference is that the approach leads in another

direction than current efforts to develop ever-more complex means to extract greater efficiencies. Examples are

the direct current computers, refrigerators, and the like used in forest homes off the grid. Though these are

electric measures, they are examples of moving to "stand alone" end-uses. Another example is gas furnaces that

are combined with gas generation of electricity for the furnace so as to end dependence on the electrical grid.

There is a whole set of established technologies and technical possibilities in this direction. The direction would

be towards simpler, easy to repair and sustain "Plain Technologies," such as have been developed by the

~ m i s h . ~ ~

35 Ed Tenner, "Plain Technology, The Amish have something to teach us," pp. 75, IWT Technology Review, July 2005.


Vectren DSM Action Plan: Final Re~ort IX Propram Evaluation

IX. PROGRAM EVALUATION

The general philosophy of program evaluation recommended is based on five principles:

Use generally accepted methods, rather than innovate new evaluation approaches.

Use an independent (from Vectren and from the program implementer) third party evaluator, where possible. At the same time, some innovation and assessment tasks usually performed by an evaluator can be built in to program implementation reporting in order to conserve dollars. If carefully structured, early program reporting by implementation staff can do double duty of providing Vectren with monitoring information and provide initial (and in some cases baseline data) information for the evaluator.

Use simple evaluation methods where possible, and especially where there is a choice between a more sophisticated, but complex, method and an adequate simple alternative that is easy for everyone to follow and understand. In particular, evaluators may not use any "black box" methods - the evaluations are required to be "transparent" as to method and calculation of results, providing enough information to allow a reader to independently see how results were arrived at, and to raise relevant questions from the source or intermediate information provided in evaluation reports.

In general, any statistical results of evaluations (not all will have statistical results) should use samples capable of yielding at least 90 percent confidence and 10 percent precision.

Evaluation costs should be kept to a moderate level, although a first-cycle evaluation may cost more than subsequent evaluations of ongoing programs, and especially should drop in cost as programs become accepted and mature.

A "high-level" evaluation plan is provided for each of the nine programs. This "high-level" evaluation is not the

actual evaluation plan for each program. Instead, it is a planning document that specifies what is to be included

in each evaluation, requirements for method and sample size, timing, and approximate budget to be allocated for

the fwst-cycle evaluation. The following step would be for the Vectren evaluation group, or evaluation partner

to draft specific requirements for an RFQ or RFP. The final step is for the evaluator to rewrite the plan, assist in

final discussions at the evaluation kick-off meeting for each program, and revise based on input and direction

from Vectren and the Advisory Board for final acceptance by Vectren. The evaluator then implements the plan

in cooperation with Vectren.

In arranging for program evaluation, Vectren and the Advisory Board should consider that there are three basic

approaches. The first approach, which we mention but do not suggest, is to create an internal DSM program

evaluation shop and staff it. This approach was done in the early days of DSM when utilities were first

experimenting with DSM programs as a way to meet customer expectations in the context of the post "oil

shock" energy crisis. The expectation for independent outside evaluations had not yet become established. So

long as early DSM was primarily an internal program, internal evaluation with occasional support from an

outside independent evaluator made sense. Today, this model is not used in the world of DSM and low-income

programs but is found in sectors, such as large foundations and many government bureaus (eg, large city health

departments), that have internal evaluation departments.


Vectren DSMAction Plan: Final Report IX Program Evaluation

The second approach is to use independent evaluators. Since the early 1990's utilities, stockholders, and

commissions have all specified the use of independent evaluators. There may still be an evaluation coordinator

in the utility, and occasionally some evaluation staff, but virtually every evaluation is carried out using an

independent evaluator. Once DSM expanded beyond the early programs, it became an area for coordination of

input and interests among many parties, which all have to be satisfied with the evaluation methods and results.

In a primarily cooperative but sometimes adversarial context, with a number of relevant parties, evaluation has

come to play a special technical role in working out expectations for programs by providing a way to agree on

evaluation methods and to measure results. This leads to open (transparent and independent) evaluations

that can find ways to improve programs to make them more eff~cient from a cost perspective and more effective

in delivery of energy savings results. Well designed evaluations, carried out by independent evaluators, can

show what works and can provide technical information that Vectren and the Advisory Board can use to make

programs better.

The third approach is a concept of the evaluator as a business partner. A number of utilities gearing up for the

new DSM cycle have adopted this concept which was largely developed during the "competitive" era in which

there was quite a bit of outsourcing of utility functions. In this approach, the evaluation team remains

independent, but the utility selects one evaluation partner for at least a five-year period. The evaluation partner

then works closely with the utility, almost, but not quite, as if they were outsourced staff. This means really

getting to know the utility counterparts, helping with program questions that come up, getting to know the

service territory and the expectations of parties, and still carrying out the independent evaluations (usually an

impact evaluation and a process evaluation for each program). This is not a majority pattern - most utilities

continue to treat programs separately and to select different independent evaluators on a program-by-program

basis. However, where commissions have not set a different pattern, the advantages of the evaluation partner

model for the utility are:

1. lower administrative effort; 2. a closer relationship which can get more value from the evaluation team in helping prevent problems

that might otherwise arrive by insuring they are involved with staff, programs, the service territory, and expectations of parties; and

3. the ability to negotiate a lower overall evaluation cost because the program evaluations do not have to stand alone, travel is reduced since there is only one evaluation team, and evaluation overheads are lower.

The second approach is the primary pattern for DSM and low-income program evaluations in the US and

Canada. The third approach has been selected by some utilities, is underway, and will likely prove out as a way

of lowering overall evaluation costs.

First Steps

As a first step, Vectren's evaluation group or the Vectren staff assigned to guide and monitor all evaluation

activities should order the following articles and key books on evaluation:


Vectren DSM Action Plan: Final Rmort IX Pro~ram Evaluation

Campbell, Donald T. "Evolutionary Epistemology." In Methodology and Epistemology for the Social Sciences: Selected Papers. Edited by E. Samuel Overman. Chicago: University of Chicago Press, 1988.

Campbell, Donald T., and David A. Kenny. A Primer on Regression Artifacts. New York and London: The Guilford Press, 1999.

Campbell, Donald T., and J. Stanley. Experimental and Quasi-Experimental Design for Research. Chicago: Rand McNally Publishing Company, 1966.

Chen, Huey-Tsyh, and Peter H. Rossi. Theory-Driven Evaluations. Newbury Park, London and New Delhi: Sage Publications, 1994.

Cook, T. D., and Donald T. Campbell. Quasi-Experimentation: Design and Analysis Issues for Field Settings. Boston: Houghton-Mifflin, 1979.

Finsterbusch, Kurt. "Demonstrating the Value of Mini-Surveys in Social Research." Sociological Methods and Research 5(1): 1 17-136, 1976.

Hill, Lawrence J., and Marilyn A. Brown. "Estimating the Cost effectiveness of Coordinated DSM Programs." Evaluation Review, 19(2): 18 1- 196, 1995.

Kaplan, Robert S., and Robin Cooper. Cost and Effect, Using Integrated Cost Systems to Drive Profitability and Performance. Boston: Harvard Business School Press, 1998.

Mattessich, Paul W., Manager's Guide to Program Evaluation: Planning, Contracting, and Managing for Useful Results. St. Paul, Minnesota: Wilder Publishing, 2003.

Posavac, Emil J., and Raymond G. Carey. Program Evaluation Methods and Case Studies, Sixth Edition. Upper Saddle River, NJ: Prentice-Hall, 2003.

Salant, Priscilla, and Don A. Dillman. How to Conduct Your Own Survey. New York: John Wiley and Sons, 1994.

Shadish, William R., Thomas D. Cook, and Donald T. Campbell. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Boston and New York: Houghton Mifflin Company, 2002.

Second, the Vectren staff member heading the evaluation should become a member of the American Evaluation

Association and provision should be made for that staff to attend the following conferences:

the AEA evaluation conference held annually,

the International Energy Program Evaluation Conference held every other year, and

the American Council for an Energy Efficient Economy Summer Study held every other year.

The two evaluation conferences offer practical short seminars in current evaluation topics. This level of activity,

plus occasional assignment to short courses or seminars is part of the necessary overhead to keep the evaluation

function effective and alert.

Third, Vectren should decide internally whether they will use an Evaluation Partner or proceed with using staff

resources.

Fourth, the plan for assigning specific program evaluations to one or more independent evaluators, and the

procedure for selection of evaluators should be determined and implemented on a timely basis.


Vectren DSMActwn Plan: Final Re~ort LX Prowam Evaluation

Evaluation for Program 1. Small Buildings Energy Efficiency Program

The description of the Small Buildings Energy Efficiency Program is repeated below from Table 36. This

program is suitable for a standard evaluation approach, using a "non-equivalent control group design" and

supplemented by regression analysis (sometimes called ccstatistically adjusted engineering analysis" or

"conditional demand analysis, depending on how the equations are set). These two evaluation approaches are

recommended as a pair for this program area because the regression analysis results are subject to distortions

unless they are "trued-up" to the simple "difference of means test" that is used to implement the calculations that

are integral to the non-equivalent control group design.

Randomized Evaluation Design Not Appropriate Recently, there has been a return among evaluators (supported by foundations and federal agencies) to the use of

randomized assignment of "treatment programs" to "cases." This approach (true randomization) yields the best

and least open to challenge scientific knowledge about program effects. However, true randomization would

DSM Technologies

Energy efficient furnaces, duct sealing, weatherization measures, blower door, EE water heaters, flow restriction measures, tank and pipe wraps, gas ranges, clothes washers (for home with gas water heating), dryers, setback thermostats and natural gas fireplaces

Impact Evaluation


End-Uses

Heating, water heating, cooking, laundry, fireplaces

Program Type

1. Small Buildings Energy Efficiency Program

Target Market

Existing residential (single family up to 4 units) and small commercial buildings (defined by square footage, employees or ccf usage)

Description

Once program size is determined, either a

year data to post-year (b) Similar database for 90% confidence and energy use for program equal size sample for

How does the pilot impact energy use?

near-census sampling

Required Data Type(s) (a) Electronic database of monthly usage, billing cycle, meter read dates, type of read for each program building.

Research Question(s) Analysis Approach Non-Experimental Control Group design with difference of means test.

Compares (two-year) pre-


require forming a list of customers in similar situations who have requested to participate in a program, and

subsequent provision of the program to only those customers selected by a random process (usually a computer

generates pseudo-random numbers, sometimes the random number tables in the back of statistics books are

used).

Those who fail to win participation would be left without a program, given a placebo program, or scheduled for

a future program year. There are situations in which randomized controlled experiments could be appropriate, if

carefully designed and justified. But, because true randomized controlled experiments have this requirement,

they are almost never used for DSM program evaluation even though they represent the best scientific method

and produce the surest knowledge of results.

The Format of the Non-Equivalent Control Group Design A set of quasi-experimental designs was first developed and systematized in the 1960s by Campbell and

~ t a n l e ~ . ~ ~ Since then, the line of progress in this evaluation approach has been developed through a number of

core evaluation methods texts.37 In the quasi-experimental design approach, each of the standard experimental

research designs has been copied, but without provision for random assignment of cases to treatment and control

(no treatment) groups.

Instead, the customers who apply and qualify for a program are accepted. For the evaluation, this group

becomes the Participant Then a very similar group of customers is selected to be the Comparison

group. Strictly speaking, a true experiment incorporates a "control group"; a quasi-experimental design

incorporates a "comparison group." The quasi-experimental designs are weaker designs than true experimental

designs because they are open to certain kinds of interference or bias of results that a control group is strong

enough to prevent in drawing statistical conclusions, but a comparison group is not. Still, on balance they are

more appropriate for DSM where the programs are not truly experimental-although the program is under test,

there is a fundamental expectation that the program designs will work well. The evaluations serve both a

fiduciary purpose -to ensure proper production of benefits and assessments of costs, and also to support the

gradual and ongoing optimization of program cost effectiveness.

36 Campbell, Donald T. and J. Stanley (1966), Experimental and Quasi-Experimental Design for Research. Chicago, Rand McNally Publishing Company. This is still the best introduction to quasi-experimental design and remains in print as a core methods text. 37 The most recent core reference is: Shadish, William R., Thomas D. Cook and Donald T. Campbell (2002), Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Boston and New York: Houghton Mifflin Company, 2002. 38 This can introduce a self-selection bias in which those who are first to enter a program, and especially in contrast to those who never enter, may be more alert, or already searching to fmd was to conserve energy. This is an example of the kind of bias randomized control experiments protect against, but quasi-experimental designs do not. For DSM, however, the effect of this bias is very small and in many cases it can be disregarded or found to be not relevant due to the program logic.


Vectren DSM Action Plan: Final Report M: Program Evaluation

Overall Savings in the Non-Equivalent Control Group Design The non-equivalent control group design is one of the most employed designs in DSM evaluation. Like all

ccnon-equivalenty' designs there is a Participant group and a Comparison group (here called a ccnon-equivalent

control" group in the name of the design).

The non-equivalent control group design is used with simple "difference of means tests." Results are developed

using the t-test or z-test, and based on simple counting and subtractions of group means. As shown in Table 64,

there is a "before" and "after" measurement taken on both the Participant Group and the Comparison Group.

Table 64. Layout for Non-Equivalent Control Group Evaluation Design

Once the program has been run for a cycle and the actual building types and characterizations are known, this

Participant Group Comparison Group

analysis may be subset and run separately for different building characterizations; or a single analysis group may

be used. This will be determined once the actual program data is reviewed.

Measurement Measurement

Analysis Method for the Non-Equivalent Control Group Design: How the Difference of Means Calculation Works

Steps: Collect Baseline and Post-DSM measurements for equal period for both the Participants and the Comparison Group. Subtract the "before" measurement from the "aftery' measurement for each group. This yields the gross energy savings for each group. Subtract the gross energy savings of the Comparison group from the gross energy savings of the Participant group. This yields the net energy savings due to the program.

Calculation of Measure Savings The regression method of the California Measurement and Evaluation Protocols is also a standard approach,

falling within the general classification of Conditional Demand Analysis (CDA) or Statistical Adjusted

Engineering Analysis (SAE) depending on how the equations are set up.39 The regression approach has two

primary virtues:

It produces savings estimates for individual measures andlor groups of measures.

It provides a facility to include all kinds of conditioning variables in the analysis.

X -

39 California Public Utilities Commission, Protocols andProcedures for the Verification of Costs, Ben@$, and Shareholder Earningsfrom Demand-Side Management Prograi7zs, Appendix J, Quality Assurance Guidelines for Statistical, Engineering, and Self-Report Methods for Estimating DSM Program Impacts, PP. 6-14, "Quality Assurance Guidelines for Conditional Demand Analysis (CDA) Models." See also Violette, D., M. Ozog, M. Keneipp & F. Stern, Impact Evaluation of Demand-Side Management Programs, Volume 1: A Guide to Current Practice. Electric Power Research Institute, Palo Alto, California: 1991, Sections 5.3 - 5.5, Pp. 5-10 to 5-32.

Measurement Measurement



It also has three primary weaknesses:

It is more complicated. The regression approach cannot be understood without completing at least an introductory college course in regression, and its problems in DSM analysis (such as multicollinearity) cannot become familiar without completion of two or three courses in regression and a substantial amount of practice with energy use data.

Multicollinearity is inherent in installing a package of measures. Within the "building box," and taken as a set, the measures interact with each other to produce higher or lower effects than they would produce separately or in other combinations. The high overlap of measure effects can make it very difficult for the least squares algorithm to allocate the appropriate "weight" or coefficient to each independent variable. This means that different analysts can get different results from correct application of the method. This is not a problem with the underlying theory of energy conservation or with the mechanics of the least- squares algorithm, as such. It is a problem of adequacy of method in relation to degree of "resolution" that is possible given relations among the independent variables within the data.40 The regression assigns the values and signs of the coefficients so that estimation of change in energy use is as close as possible to the actual case values. The mathematics of this will work in any case. However, if some of the measures overlap then some of the measures that overlap may have the wrong sign.41 They may also have the wrong size in relation to physical knowledge of how buildings and measures work, depending on specification of the regression equation.42

The method assumes a standardization of building conditions that may or may not actually exist.43 In actual practice, the method is easy to "trick" into showing higher overall savings than have actually been generated by the measures installed. It automatically allocates the real savings plus the pseudo savings across the measures.

How the Regression Approach Works Beginning with ex ante estimates for each measure, the regression derived coefficients are realization rates. For each measure, when the ex ante amount for the measure is multiplied by its realization rate the result is the gross savings. The gross savings, multiplied by the measure's net-to-gross ratio yieids the net savings due to the program.

Steps:

Receive Ex Ante values (planning estimates of per measure savings for each measure included in the program). Apply regression analysis using billing data, program installation data, and Ex Ante estimates. Output "realization rates." Multiply Ex Antes by associated realization rates. Output is estimates of gross measure savings. Apply net-to-gross ratios. Output is Ex Post energy savings estimates for each measure.

40 Kahane, Leo H. Regression Basics. Thousand Oaks, California, London, and New Delhi: Sage Publications, 2001, pp. 114. 41 Among several coefficients, some may be given a sign that does not accord with theory, or appears physically impossible. Montgomery, Douglas C., Elizabeth A. Peck and G. Geofiey Vining. Introduction to Linear Regression Analysis, Third Edition. New York: Wiley, 2001, pp. 120-130. 42 Neter, John, Michel H. Kutner, Christopher J. Nachtscheim and William Wasserman. Applied Linear Regression Models, Third Edition. Chicago: Irwin, 1996, pp. 290-29 1. 43 Essentially, the regression assumptions include a "fixed model," that is, the results developed from the regression are specific to the specification of the regression equation fiom which the estimates were developed. Problems arise when measure coefficients are unstable in different regression runs with different specifications, and also in abstracting measure values from the context of a specific regression to use as constants to project savings from a wider population.



Combination of the Two Methods The vulnerability of the regression approach when used alone to overestimate overall program energy savings

and to distribute both real savings and pseudo-savings over the measures can be fxed by using the net savings

developed in the simple difference of means test to "true-up" the overall savings from the regression analysis.

Then the measure savings from the regression analysis can be ratioed back to eliminate any overall pseudo-

savings. The problem of the allocation of (corrected) overall savings among the measures remains and has to be

treated either as a matter of professional judgment or additional protocol steps to specify the regressions.

Process Evaluation The purpose of process evaluation is to describe the program as planned and as delivered. It documents

perceived successes and failures in program definition, administration and actual service delivery. It also

documents changes introduced to resolve problems and improve service. Process evaluation begins with the

assumption that it is possible to improve every program. It collects and organizes technical knowledge

developed through the course of program implementation as a key source for improvements by program staff.

We learn by doing the work, so it is likely that management and staff will change several program factors as the

program matures.

Using the questions below as guides, the process evaluation will describe how the program works and document

the history of the program. It will detail the roles and responsibilities of staff, contractors and other parties,

while tracking promotional and marketing efforts. And, it will relate the story of administration and program

process.

The basic method for the (qualitative) process evaluation is to compare plans for the program with what actually

occurs. Process evaluation includes discussion of barriers to effective implementation. It also includes

discussion of factors that make the program effective, and for developing recommendations for improvement for

future program cycles.

Program Description Questions:

What are the program goals?

How is the program trying to meet these goals?

How is the program organized? What is the program structure, management, and how does the organizational process work?

What are the program energy conservation measures?

What are the educational aspects of the program, and how is the education dimension of the program intended to work?

What are the linkages of the program to other programs/resources? How are these linkages intended to work?

What is the flow of activity; from a customer perspective, the stages of steps that participants pass .through to accomplish the program?

-- -

Prepared by Forefront Economics and H. Gil Peach & Associates page 104

Vectren DSMAction Plan: Final Report lX Program Evaluation

Institutional Questions:

How have program goals changed? Are there implicit or explicit changes in goals?

What factors support achievement of program goals? What factors impede planned program achievements?

How does the program work, both formally and informally? What are the actual organizational processes (contract procedure, operating procedure, contractor perception, staff perception, management perception)? This includes recruitment, handling applications and inquiries, relationships with program partners, and provision of services and quality control.

Assess the adequacy and workability of the program monitoring and tracking systems.

What factors might explain any detected difference between expected and actual energy usage?

Could the program design be improved to improve the capability, efficiency, or effectiveness in achieving program goals? How can management and staff change program implementation to better achieve program goals?

Customer, Customer Relations and Marketing Questions:

Is the program meeting the target market or only sub-segments of the target market? What elements of the program have wide customer appeal? Which program elements do customers perceive as drawbacks?

How did participation change over time? How did participation track with customer communication efforts? What factors might explain the participation or lack of participation achieved by the program?

How is the program perceived by participants?

Is there any detectible difference between participants and non-participants?

How can management and staff improve marketing efforts? In particular, are there any real or perceived barriers that exist for program participation?

Information for the process evaluation should be gathered from program records and data, participant surveys,

and brief interviews with program providers.

Schedule Before the program begins, Vectren should set up a procedure for the evaluator to request baseline energy

consumption records to handle that request when it becomes necessary for the evaluation. The evaluation team

should be designated within three months after the program begins, and the process evaluation (and evaluation

Kick-Off meeting with Vectren and the Advisory Board) should take place six months after the program begins.

An interim process evaluation is scheduled to be delivered to the Advisory Board the first month of the second

program year (month 13). The purpose of this evaluation is similar to the full process evaluation but with

emphasis on providing early program feedback to implementers and the Advisory Board. The program

implementer may do an initial satisfaction survey as it completes installations. This can be used for on-going

project reporting to Vectren, and also by the process evaluator, later.

This evaluation will require two fill years of baseline data. These should be common to all buildings (rather

than different for individual buildings), include both Participant and Comparison group buildings, and most

likely should be calendar years of data. A full year of post-DSM energy consumption for each Participant and


Vectren DSM Action Plan: Final Report IX Program Evaluation

Comparison building will also be required, along with survey information from both Participant and

Comparison group buildings to support the regression analysis.

Schedule Overview Program Implementation begins. Initial Satisfaction Surveys begin and are carried out by implementation contractor throughout the program cycle. [Month 31 Evaluation Team designated. [Month 61 Kick-Off Meeting for Evaluation [Month 61 Process Evaluation begins. [Month 131 Interim process evaluation delivered to Advisory Board. A full year of post-DSM consumption data is required for each building. This means the first evaluation results can be developed one-year following the end of the first program year (to allow a full year of implementation for the program and one year after). The first draft evaluation impact evaluation report will be due two years and five months following program implementation.

Evaluation for Program 2. General Services Energy Efficiency Program

This program will serve existing large commercial and industrial facilities with prescriptive equipment rebates


ups (building commissioning) and control systems.44

Program Type Target Market End-Uses DSM Technologies

2. General Services Existing and new medium to large Heating, water Boiler replacement, water heating Energy Efficiency commercial and industrial heating equipment Program facilities

Engineering Desk Review Since this program uses prescriptive equipment rebates, the recommended impact evaluation method is an

engineering desk review, with no direct measurement in plants or facilities, and no analysis of billing data. This

approach is possible for this program for two reasons:

The list of units approved for this program is discrete and the physical capabilities of the prescribed units are known. Retrofit of this kind is highly predictable based on physical calculations, with virtually no behavioral effects except a firm leaving business.

Another support for this approach is that the facility is paying most of the cost of each measure and Vectren is adding a smaller amount that can be treated as a "buy-down." Facility managers are very careful to manage retrofits in a way that maximizes the return to their company of each dollar invested in an upgrade.

The essential data needed for this evaluation is a characterization for each piece of equipment that is replaced,

and access to the internal management justification for the replacement (or at least the internal summary

Larger and customized retrofits will be covered under the Custom program (Program 3).

Prepared by Forefronf Economics and H. Gil Peach & Associates Page 106


calculation carried out by the customer firm). If the characterization of the existing (to be replaced) equipment

is recorded, a seasoned engineer can verify the approximate energy savings associated with the retrofit.

The evaluation engineer for this program evaluation should have industrial or large commercial experience, so

as to be well accepted by the customer f m s .

Process Evaluation The process evaluation will be focused as low-key, with a short list of research questions, and if surveys are

used they are to be short mini-surveys. Generally, the process evaluation will focus on perceptions of key

individuals at the customer f m s , and perceived strengths and weakness of the program from a customer

perspective. Any barriers to participation should be detailed, along with a short list of recommendations to

make the program more efficient and effective or to better tailor the program to serve customer needs. In

addition to the customer-firm interview, the process evaluator will interview program implementation staff and

the Vectren program manager.

Schedule Before the program begins, Vectren should create a file of program documentation to provide to the evaluator.

This should include the full list of approved prescriptive measures, including date of approval if measures added

after the program begins. Vectren also needs to ensure that the implementation contactor is tasked with

gathering the documentation that justifies each change-out and that these are kept systematically for when the

evaluation begins. The evaluation team should be designated within three months after the program begins and

the process evaluation (and evaluation Kick-Off meeting with Vectren and the Advisory Board) should take

place six months after the program begins.





project reporting to Vectren, and also by the process evaluator later.

This evaluation will not require baseline data, but it will require access to economic justifications for each

change-out and a name on record for the evaluator to call at each customer firm to discuss the change-outs. This

evaluation can stay close to the program implementation activities and a draft final report should be submitted at

the end of the first quarter following the first implementation year.


Vectren DSMAction Plan: Final Reporf 1X Program Evaluation

Schedule Overview Program Implementation begins Initial Satisfaction Surveys begin and are carried out by implementation contractor throughout the program cycle [Month 31 Evaluation Team designated [Month 61 Kick-Off Meeting for Evaluation [Month 61 Process Evaluation begins [Month 131 Interim process evaluation delivered to Advisory Board. The first draft evaluation impact and process evaluation report will be due three months following the end of the first year of program implementation

Evaluation for Program 3. Customized Energy Efficiency Program

This program serves existing large commercial and industrial facilities with customized information for making

improvements to their gas end-use operations. It consists of a Technical Assessment conducted by a qualified

engineering f m , under contract to Vectren, with the customer's portion of the costs reimbursed by Vectren if

they proceed with the recommendation^.^^

Engineering Desk Review This program provides a Technical Assessment by a qualified engineer or engineering firm. Since all of the

analytic work is completed by the implementation engineer, a simple desk review by a seasoned engineer is all

that is required for the impact evaluation, plus verification for those projects for which reimbursement is

claimed. The verification can be built in to the implementation effort and should only require document review

by the engineering evaluator. This program may also include prescriptive equipment rebates, for which the

Program Type

3. Customized Energy Efficiency Assessment Program

recommended impact evaluation method is also an engineering desk review, with no direct measurement in

plants, facilities or billing data analysis.

End-Uses

Heating, water heating, process uses

Target Market

Existing and new large commercial and industrial facilities

Process Evaluation As for the prescriptive program evaluation, the process evaluation for the Custom Program will be low-key, with

a short list of research questions, and if surveys are used they are to be short mini-surveys. Generally, the

process evaluation will focus on perceptions of key individuals at selected customer firms and perceived

strengths and weakness of the program from a customer perspective. Any barriers to participation should be

detailed, along with a short list of recommendations to make the program more efficient and effective or to

better tailor the program to serve customer needs. In addition to the customer-fm interview, the process

evaluator will interview program implementation staff and the Vectren program manager.

DSM Technologies

All identified gas end-uses

45 Participants may also take advantage of the prescriptive equipment rebates offered under Program 2 for upgrading heating, water heating and gas cooling systems, boiler replacement, water heating equipment, tune ups and control systems.


Vectren DSMActwn Plan: Final Report LX Program Evaluation


As the implementation contractor proceeds, the implementation contractor should be tasked to provide and

maintain a file of site-by-site information to be reviewed by the engineering evaluator. This should include both

the result of information assessments, custom recommendations and the full list of prescriptive measures

included in a package.

The evaluation team should be designated within three months after the program begins and the process

evaluation (and evaluation Kick-Off meeting with Vectren and the Advisory Board) should take place six

months after the program begins.





project reporting to Vectren, and also by the process evaluator later.

This evaluation will not require baseline data, but it will require access to economic justifications for each

change-out for cases in which reimbursement is provided and a name on record for the evaluator to call at each

customer firm to discuss the change-outs. This evaluation can stay close to the program implementation

activities and a draft final report should be submitted at the end of the first quarter following the first

implementation year.

Schedule Overview Program Implementation begins Initial Satisfaction Surveys begin and are carried out by implementation contractor throughout the program cycle [Month 31 Evaluation Team designated [Month 61 Kick-Off Meeting for Evaluation [Month 61 Process Evaluation begins [Month 131 Interim process evaluation delivered to Advisory Board. The first draft evaluation impact and process evaluation report will be due three months following the end of the first year of program implementation

Evaluation for Program 4. Hospitality Industry Energy Efficiency Program

This program is targeted to restaurants, bakeries, institutional housing (nursing homes, colleges, and schools),

hotels, hospitality facilities and other cooking facilities that employ natural gas for cooking and food

preparation. The program provides incentives and promotes the installation of energy efficient booster water

heaters, dishwashers, fryers, griddles, and gas ranges to replace aging equipment in existing facilities andlor as

the efficient option for new facilities.


Vectren DSMAction Plan: Final Re~ort E Pro~ram Evaluation

Desk Review This program provides different opportunities depending on how it is implemented each year. If it is largely

implemented through customer communications and rebates, then a small number of spot checks by telephone

or in person by the evaluator, along with the implementation record of equipment receipts and records will be

sufficient. At the same time, this program has an alternative implementation by a charismatic program leader

familiar with the industry. If such a program leader sets up meetings from town to town, the rebate portion of

the program will be the same as in a more remote administration, but there may be significant additional

opportunities to improve work practices to save energy. That part of the puzzle will require a Vectren staff

leader or a highly skilled and experienced implementation contractor to develop turn-out for meetings and then

to actually observe and make recommendations for practices in individual food preparation shops.

Program Type

4. Hospitality Industry Energy Efficiency Program

In the first type of implementation, the evaluator and the evaluation effort can be low-key and primarily focused

on review of written records. In the second type, the evaluator will need to accompany the implementer to a

small number of meetings for presentations and to accompany the implementer in "walk-through" audits of

facilities. In the second type of implementation, the evaluator or the implementer will need ability to informally

spot meter some equipment to estimate effects. In the first type, this will not be necessary. This study will

proceed as with the prescriptive equipment rebates (Program 2), except that some spot metering is expected.

There will be no billing data analysis.

Process Evaluation The process evaluation will be limited to telephone or in-person contact with a small number of facilities.

Generally, the process evaluation will focus on perceptions of key individuals at selected customer firms and

perceived strengths and weakness of the program from a customer perspective. Any barriers to participation

should be detailed, along with a short list of recommendations to make the program more efficient and effective

or to better tailor the program to serve customer needs. In addition to the customer-firm interview, the process

evaluator will interview program implementation staff and the Vectren program manager.

Target Market

Restaurants, bakeries, institutional housing, hotels, hospitality facilities and other cooking facilities


As the implementation contractor proceeds, the implementation contractor should be tasked to provide and

maintain a file of site-by-site information to be reviewed by the evaluator. If the implementer makes use of

town meetings andlor on-site "walk through" audits, the process evaluator should participate in at least three

town meetings and at least seven walk-throughs.

Prepared by Forefront Economics and H. Gil Peach & Associates Page 11 0

End-Uses

Cooking and food preparation

DSM Technologies

Energy efficient gas ranges, ovens, broilers, warmers and related processes

Veclren DSMAction Plan: Final Report M: Program Evaluation






emphasis on providing early program feedback to implementers and the Advisory Board. This evaluation will

not require baseline data, but it will require access to implementation records and a name on record for the

evaluator to call at each customer fm to discuss the change-outs (although only a sample of f m s will be

contacted by the evaluators). This evaluation can stay close to the program implementation activities and a draft

final report should be submitted at the end of the first quarter following the first implementation year.

Schedule Overview Program Implementation begins [Month 31 Evaluation Team designated [Month 61 Kick-Off Meeting for Evaluation [Month 61 Process Evaluation begins [Month 131 Interim process evaluation delivered to Advisory Board. The first draft evaluation impact and process evaluation report will be due three months following the end of the first year of program implementation

Evaluation for Program 5. Multi-Family Building Energy Efficiency Program

This program will serve multi-family buildings with prescriptive equipment rebates for upgrading heating, water

heating and gas cooling systems; boiler replacement, water heating equipment, tune ups and control systems.

Building occupants and tenants will also be given a package of free low-cost weatherization measures for self-

installation.

Evaluation Options If the focus of this program is on whole building applications (exclusive of the low-cost/no-cost packages for the

tenants), then it will be useful to approach the evaluation with a billing analysis of common areas, that is, of the

portion of the building's energy use that is the building owner's responsibility (hot water, furnace, common area

lighting). If this is done, it will be necessary to ensure the correct metered accounts are identified. This

identification can be tasked to the program implementer.

Program Type

5. Multi-Family Building Energy Eficiency Program

However, if the program develops in a way that disburses the program budget in a pattern of one or two

measures per apartment building across a very large number of buildings, an analysis of metered energy use will


Target Market

Multi-family buildings with 5 or more units, dormitories, hotels, other large residential facilities

End-Uses

Common area boilers, water heating and laundry; individual unit water heating, weatherization

DSM Technologies

Energy efficient furnaces, duct sealing, weatherization measures, blower-door, EE water heaters, flow restriction measures, tank wraps

Vectren DSMAction Plan: Final Report DL Program Evaluation

be less valuable. It is likely, in advance of program implementation that the program will take both directions,

with some buildings engaged in fairly holistic retrofit and others participating for one or two measures. In that

case, the buildings will be partitioned into two groups and two different evaluation approaches will be used.

Energy Use Analysis Analysis for buildings with a holistic approach to common area measures will use a standard Non-Equivalent

Control Group design with a double pre-test.46 The double pre-test gives two prior year (calendar year)

measures and adds to the stability of the design. The analysis method will be the simple different of means test,

as developed for Program 1, above.

For buildings with only one or two measures, the evaluator will have to consider whether to include them in the

core analysis with the holistic measure package buildings or whether to estimate them separately without direct

measurement but with review of claimed savings referenced to the building's baseline energy use using a

simulation package, such as E-Z Sim.

Table 65. Layout for Non-Equivalent Control Group Evaluation Design with Double Pre-Test

There is not a high return in analyzing low-cost/no-cost measure impacts because they are typically so small for

apartments. For a single-family home, diligent application of a kit of low-cost/no-cost measures, typically, is

claimed to provide minor savings of perhaps $100 to $150 in the first year as measured across water, gas and

electricity savings. However, much of this is often due to a water heater wrap, which will not apply to an

apartment with central hot water. Also, measured savings for the kit-type programs are more likely to be $50 to

$100, spread over a year. Still, the kits are typically cost effective and customers often like these programs; and

they do provide some physical improvement to the home. It makes sense to do the kits when the apartment

building is also receiving more major measures to improve the building as a whole. However, for evaluation

purposes it is realistic to stipulate a value for the expected savings per apartment and confine the evaluation to

an inspection as to whether and to what extent measures have actually been installed. (When kits are distributed

to apartments, but the items are left to the tenant to install, many of the kits are never installed.) Direct

measurement in this case would have a poor signal-to-noise ratio due to the small amount of the savings

produced by kits and the fact that three streams of data would have to be analyzed (gallons of water, therms, and

kwh). It is possible to do the direct measurement and to detect small effects if the sample sizes are large

enough.

46 Shadish, Cook and Campbell, op. cir., Page 145.


Vectren DSMActwn Plan: Final Report IX Program Evaluation

Process Evaluation Because the major measures are not in the apartments, the process evaluation will be similar to the process

evaluation of the General Services Program (Program 2). The process evaluation will be focused as low-key,

with a short list of research questions, and if surveys are used they are to be short mini-surveys. Generally, the

process evaluation will focus on perceptions of key individuals (building owners andlor landlords, or the

building engineer if there is one). The process evaluation will focus on perceived strengths and weakness of the

program from a customer perspective. Any barriers to participation will be detailed, along with a short list of

recommendations to make the program more efficient and effective or to better tailor the program to serve

customer needs. In addition to the customer-firm interview, the process evaluator will interview program

implementation staff and the Vectren program manager, provide a description of the program and how it works,

and provide the story of any problems encountered by program staff during implementation and how they were

overcome.


This should include the full list of approved prescriptive measures included in the multi-family program,

including date of approval if measures are added after the program begins. Vectren also needs to ensure that the

implementation contactor is tasked with gathering the documentation that justifies each change-out and that

these records are kept systematically for when the evaluation begins, and thereafter. The evaluation team should

be designated within three months after the program begins. and the process evaluation (and evaluation Kick-

Off meeting with Vectren and the Advisory Board) should take place six months after the program begins.




implementer should be tasked to do an initial satisfaction survey as it completes installations. This can be used

for on-going project reporting to Vectren, and also by the process evaluator later.

This evaluation will require two years of baseline data for each building included in the program, careful

identification of appropriate meters (by the implementation contractor) for analysis, and one year of post-retrofit

data. The simplest design is to use two common baseline years (calendar years), implement the program for a

year, then use a common post year for all buildings completed in the first program year; then repeat for each

subsequent program year. The draft impact evaluation will be due three months after the end of the first

program year, and can be updated on a yearly basis thereafter throughout the program cycle.


Vectren DSMAction Plan: Final Re~ort IX Prowam Evaluation

Schedule Overview Program Implementation begins Initial Satisfaction Surveys begin and are carried out by implementation contractor throughout the program cycle [Month 31 Evaluation Team designated [Month 61 Kick-Off Meeting for Evaluation [Month 61 Process Evaluation begins [Month 131 Interim process evaluation delivered to Advisory Board. The first draft evaluation impact and process evaluation report will be due one-year and three months following the end of the first year of program implementation

Evaluation for Program 6. Innovative EE Technologies Research and Demonstration Program

This is an innovative program that will provide funding to the Indiana Clean Manufacturing Technology and

Safe Materials Institute (CMTI) for conducting research into emerging gas technologies that contribute to

increased energy efficiency in both residential and non-residential applications. There are no direct incentives to

customers under this program, rather it provides a funding stream to support research into technologies that may

be added to the portfolio of programs offered by Vectren in the future. In addition, case studies would be

supported at some customer sites on a case-by-case basis, such that field demonstrations could be performed and

studied for potential future market application.

Evaluation Approach (Technology Evaluation) The evaluation approach for this project is "technology evaluation." This is special area of evaluation with a

well developed set of standard methods which are aimed at assessment of technology potentials. The evaluation

will be based on review of funding and project documents, including monitoring reports as well as interviews

with the CMTI and any designated profession and scientific staff. There will be a single evaluation covering

both impact and process elements. The orientation of the evaluation will be forward-looking, with discussion of

potential impacts of technology work that is funded. The evaluation is expected to be low-key and to involve

some research and reporting of related technology implications so as to develop a context within which

technology work can be interpreted. It is not to become as focused as in development of technology road

maps.47

47 The models for technology evaluation are provided by standard evaluations of this type carried out by the federal national laboratories, and in foundation internal program evaluations.

DSM Technologies

Emerging high efficiency natural gas technologies, use of renewable technologies to off-set or enhance gas technologies

Program Type

6 . Innovative Energy Efficiency Technologies Research and Demonstration Program


Target Market

All markets

End-Uses

All gas end-uses



The evaluation team should be designated within six months after the program begins and the evaluation Kick-

Off meeting with Vectren and the Advisory Board should take place nine months after the program begins.

The evaluation will begin with a review of program documents and an informal meeting with Vectren, the

Advisory Board (or interested members) and CMTI. From that point forward, the evaluator will meet with

CMTI by phone or in person every quarter. Two meetings, both in-person and on-site, will be required each

year. Since the evaluation is a technology evaluation, the approach will be different from all of the other

evaluations, similar to foundation program evaluations which typically involve considerable consultation and

reporting and integration of perceptions and potentials. Since the evaluation will track with the program

development, the draft impact evaluation will be due three months following the end of the first program year,

then on a yearly schedule with dates to be determined throughout the program cycle.

Schedule Overview Program Implementation begins [Month 61 Evaluation Team designated [Month 91 Kick-Off Meeting for Evaluation [Month 91 First meeting with CMTI The first draft evaluation will be due one-year and three months following the initiation of program implementation

Evaluation for Program 7. Energy Efficient Builder Program

This program will promote the incorporation of high efficiency design features in new homes, plus installation

of high efficiency equipment above standard appliances, furnaces and windows. It will be targeted to builders

of subdivision and tract homes.

Evaluation Approach (Engineering Desk Review) Since this program will be keyed to the national Energy Star program and also incorporate the prescriptive

measures list for Program 2 above, an engineering desk review is the recommended impact evaluation method

(as in Program 2). Unless there is some reason to investigate a particular measure, we will rely on measure

values as established by Energy Star and for the prescriptive measures list for Program 2. In addition to a

secondary engineering review, the evaluation will rely on installation and/or verification records developed by

the program implementation contractor. Vectren will need to task the program implementation contractor with

development of reliable installation and/or verification records.

Program Type

7. Energy Efficient Builder Program


Target Market

Residential and non- residential new construction

End-Uses

Any gas end-uses being considered

DSM Technologies

Design incentives to upgrade planned gas equipment to energy efficient options, reduced hook-up fees and/or line extension costs

Vectren DSMActwn Plan: Final Report IX Program Evaluation

Process Evaluation The process evaluation will be focused as low-key, with a short list of research questions, and if surveys are

used they are to be short mini-surveys. Generally, the process evaluation will focus on perceptions of key

individuals at the builders and perceived strengths and weakness of the program from a builder perspective.

Any barriers to participation should be detailed, along with a short list of recommendations to make the program

more eficient and effective or to better tailor the program to serve customer needs.

In addition to builder interviews, the process evaluator will interview program implementation staff and the

Vectren program manager. Also, the process evaluator with be tasked with describing the builder community,

the extent to which local builders can enter into programs, and the extent to which subdivision standards are

dependent on national programs of multi-state builders. The process evaluation should fully relate the nature

and direction of new building markets in relation to the program.

Schedule Before the program begins, Vectren should create a file of program documentation to provide to the evaluator

and identify Vectren staff with key responsibilities for builder relationships. The evaluator should proceed in

careful coordination with Vectren staff to ensure that relationship expectations are fully observed in the work.






emphasis on providing early program feedback to implementers and the Advisory Board. This evaluation can

stay close to the program implementation activities and a draft final report should be submitted at the end of the

first quarter following the first implementation year.

Schedule Overview Program Implementation begins [Month 31 Evaluation Team designated [Month 61 Kick-Off Meeting for Evaluation [Month 61 Evaluation begins [Month 131 Interim process evaluation delivered to Advisory Board. The first draft evaluation impact and process evaluation report will be due three months following the end of the first year of program implementation

Evaluation for Program 8. New Program Development and Regulatory Affairs

This program is a support program; it does not deliver direct energy savings. Instead, this program serves as a

budget line item to allow for new project developments and coordinated interaction with regulatory and



legislative bodies for development of policies supportive of DSM activities (e.g., improved energy efficiency

building codes and standards).

Two additional program areas are the corn stove promotion and the "real" programmable thermostat

demonstration. Both of these initial program areas are practical, but neither is particularly "high-tech." Corn

stoves, of course, are special in that they have an EPA waiver because they are so clean. A "senior-friendly"

programmable thermostat is a simple device using ordinary technology that is needed.

DSM Technologies

Emerging technology research and demonstration and regulatory liaison activities

Program Type

8. New Program Development and Regulatory Affairs

With regard to regulatory affairs, there is a continuing need for Vectren to be pro-active in gaining state

regulatory and legislative policies that provide inexpensive energy conservation (such as, slow improvement of

energy efficiency housing codes, slow improvement of code enforcement, solar orientation of new construction,

and recognition of passive solar savings as DSM).

Evaluation Approach (Policy Evaluation) The evaluation for this program will be a single evaluation, emphasizing policy and process evaluation. There

are standard methods, for policy evaluation, which will be adapted for this policy and program development

area. As specific DSM test or support research is initiated, each program area will require its own specific

evaluation approach.

Target Market

All sectors

Schedule Before the program begins, Vectren should create a file of program documentation to provide to the evaluator,

and discuss any new program or policy areas underway with the evaluator when the evaluator begins work. The

evaluation team should be designated within three months after the program begins and the evaluation (and

evaluation Kick-Off meeting with Vectren and the Advisory Board) should begin six months after the program

begins. This evaluation can stay close to the program implementation activities and a draft final report should

be submitted at the end of the first quarter following the first implementation year.

End-Uses

All end-uses

Schedule Overview Program Implementation begins [Month 31 Evaluation Team designated [Month 61 Kick-Off Meeting for Evaluation [Month 61 Evaluation begins [Month 131 Interim process evaluation delivered to Advisory Board The first draft policy and new program development evaluation will be due three months following the end of the first year of program implementation


Vectren DSM Action Plan: Final Report LX Program Evaluation

Evaluation for Program 9. Public Education and Outreach Program

This program will provide funding for cross-program public education activities to raise awareness of the

benefits and methods of improving energy efficiency in homes and businesses.

This program is not subject to cost effectiveness screening, which is deemed inappropriate according to the

California Standard Practice Manual:

Program Type

9. Public Education and Outreach Program

"For generalized information programs (e.g., when customers are provided generic information on means of reducing utility bills without the benefit of on-site evaluations or customer billing data), cost effectiveness tests are not expected because of the extreme difficulty in establishing meaningful estimates of load impacts."

Types of activities that would be included in this program are:

Target Market

All sectors

General mass media campaign for the public on pending gas price increases and ways to help control utility bills through energy efficiency measures and actions

Development of (update of the) Vectren North website to include the latest energy efficiency information for commercial, residential and school use

Targeted educational campaign for businesses

Targeted training and educational program for trade allies

Distribution of federal ENERGY STAR and other national organization materials in the service territory

A schools curriculum program to educate teachers and direct students to available educational materials on the Web about energy efficiency opportunities

The schools program component will require developing an energy education outreach program targeting

Vectren North service territory schools K-12; providing energy curricula to schools that teach students the

fundamentals of energy and how to change behavior to conserve; and securing consultant services to provide

teacher training and classroom materials.

End-Uses

All end-uses

Evaluation Approach (Process Evaluation) The process evaluation will tell the story of the educational and promotional effort, and will take its specific

DSM Technologies

All technologies

content from the directions taken in the program effort. Some of the areas covered in the process evaluation:

Review of promotional and marketing materials.

Evaluation of promotional and market plans, and implementation efforts.

Reporting on showinglairings of materials, communications events, and awareness efforts.

For any demonstrations or seminar events, the evaluation will rely upon mini-survey questions on beginning and completing the specific activity.


Vectren DSMAction Plan: Final Report LX Program Evaluation

Activities will be grouped into types and evaluated using a "case study" approach.48 At certain points, methods of "reason analysis" may also be used.49

When a schools program component begins, evaluation will be based on interviews with selected teachers and students plus: number of schools participating, number of teachers trained in the curricula, and number of students receiving energy awareness education through the program. If the schools program includes any before versus after assessments, they will be incorporated into the process evaluation.

The evaluation report for this program will be a process evaluation, focused on promotion, marketing and

communication. It will be primarily descriptive, telling the story of the effort.

Schedule Before the program begins, Vectren should create a file of program documentation, to provide to the evaluator,

that tracks promotional, marketing, communication and education effort initiated under this program. Vectren

staff andlor implementation contractor for this area (if any) should be tasked to meet with the evaluator to

discuss activities and plans for the year as the evaluation begins. The evaluation team should be designated

within three months after the program begins and the evaluation (and evaluation Kick-Off meeting with Vectren

and the Advisory Board) should begin six months after the program begins. This evaluation can stay close to

the program implementation activities and a draft final report should be submitted at the end of the first quarter

following the first implementation year.

Schedule Overview Program Implementation begins [Month 31 Evaluation Team designated [Month 61 Kick-Off Meeting for Evaluation [Month 61 Evaluation begins [Month 131 Interim process evaluation delivered to Advisory Board The first draft (process) evaluation will be due three months following the end of the first year of program implementation

48 Yin, Robert K. and Donald T. Campbell. Case Study Research: Design & Methods, Vol. 5, Third Edition. Newbury Park, California, London and New Delhi: Sage Publications, December 2002. 49 Zeisel, Hans. Say It With Figures, Fifth Edition, Revised. New York: Harper and Row, 1968.


Vectren DSMActwn Plan: Final Report Appendix A. Data Sources and References

APPENDIX A. DATA SOURCES AND REFERENCES

Primarv Nectren)

Revenue Ledger Reports

Gas Cost Adjustment Filings

Customer Information and Billing Data Extracts from Banner System

Residential Customer Survey

Secondary

Census - Population - Housing Attributes - Housing Permitted for Construction

Woods and Poole - Employment

References

Amann, Jennifer Thorne, and Eric Mendelsohn. Comprehensive Commercial RetroJit Programs: Review of Activity and Opportunities. Report No. A052. Washington, DC: American Council for an Energy-Efficient Economy, April 2005.

Buying and Maintaining an Energy-Eficient Home: A Resource Guide for Housing Educators and Counselors. Publication H1593. Washington, DC: Fannie Mae, 2004.

Marks, Michael, Joseph Lopes, and Luisa Freeman. Comprehensive Resources Assesslnent Plan for New Jersey Natural Gas Utilities. Hauppauge, NY: Applied Energy Group, April 2000.

Natural Gas Utility Conservation Rebates (Gas Rebate Facts). St. Paul: Minnesota Department of Commerce, February 2003.

Nadel, Steve, et al. Emerging Energy Saving Technologies and Practices for the Buildings Sector. Report No. A985. Washington, DC: American Council for an Energy-Efficient Economy, December 1998.

Elliott, Neal, R., and Anna Monis Shipley. Impacts of Energy Eficiency and Renewable Energy on Natural Gas Markets: Updated and Expanded Analysis. Report No. E052. Washington, DC: American Council for an Energy-Efficient Economy, April 2005.

Energy Conservation Improvement Program. St. Paul, Office of the Legislative Auditor, State of Minnesota, 2005.


Vectren DSMAction Plan: Final Report Appendix B. Methodology

APPENDIX B. METHODOLOGY

Choice of Methodology-The simplest approach to DSM analysis, oflen used in larger multi-utility DSM

planning, uses synthesizes estimates from demographics applied to engineering prototypes. This approach is

easy to apply to individual measures and to small groups of measures where the result of all the measures is

small relative to the total energy sales. But the simple synthesis approach becomes unstable where a large or

comprehensive technical potential is contemplated because the sum of the savings can sometimes exceed the

total energy sales. In this case, where a technical potential will be derived from a maximum application of a

wide variety of measures, it is particularly important to be able to establish a reasonable upper bound to the

space heat technical potential and to the base load technical potential.

A second problem with the simple synthesis approach is the interaction of measures. Whenever there is a load

reduction measure, the net realized energy savings will also be dependent on an assumed thermal conversion

efficiency. Where a conversion efficiency is changed at the same time as a load reduction, the result is

interactive, and it is important to consider the effect of both measures simultaneously. In this case, where a wide

range of efficiency and load reduction measures will be applied, it is also particularly important to be able to

deal with measure interactions in an orderly way.

Following the need for a reasonable bound for technical potential, and following the need to deal accurately with

measure interactions, we have chosen to use a calibrated engineering model. This approach is calibrated to

existing use which provides a realistic starting point for calculating savings or technical potential. The model is

structured to include variables for conversion efficiency and load reduction measures so that these types of

measures may be modeled simultaneously. However it is important to note that a calibrated model can only be

used if there is a coherent body of information to calibrate to. In this case, the DSM planning is for a single

compact utility with a coherent body of available billing information which can support a calibrated modeling

approach.

Use of General Ledger Records-It is fortunate that Vectren has readily available monthly ledger record

summaries that have the total quantity of gas sales for each rate class. While these records are timely and

available, they were created for accounting purposes, not engineering ones. Therefore, a few adjustments need

to be made to the ledger data in order to use it for these purposes.

The principal adjustment lies in associating the correct average temperature to the gas sales noted in the ledger.

It is assumed that the ledger sales are accumulated as each of the 21 meter read cycles is completed. Under

these circumstances, the temperatures during the actual time that the energy is used will be for the prior month

as well as the current month. In this analysis the temperature associated with ledger gas sales for a particular


Vectren DSM Action Plan: Final Report Appendix B. Methodology

month will be the average of the particular month and the month preceding it. These temperatures are referred

to in this analysis as the "lagged monthly temperatures."

A second potential adjustment lies in the fact that the ledger records are month by month with a different

number of business or meter read days in each month. It is assumed that each month 2 1 meter read cycles are

completed regardless of the number of days in the month, and that the ledger usage and customer numbers for

the month actually do reflect a single month of usage.

Usage Normalization - For planning purposes, usage data is normalized to the average 30-year temperatures for

the region, in this case Indianapolis. Figure 32 shows the actual temperatures in the test year and the long-term

average temperatures.

80

70 - 60 !L g 50 *

40 at 2 30 Q)

I- 20

10

0

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Month

A i r 2004-05 -Water -30-Yr Air

Figure 32. Air and Water Temperatures

In Figure 32, it is evident that the test year, green, is close to the 30-year average, red. The water temperature in

Figure 32 refers to the ground water temperature which is used in the end-use models for hot water heating

energy. In this case, the 30-year estimate of the groundwater temperature is assumed the same for the test year.

- --


Vectren DSMActwn Plan: Final Report Appendix C. Technology Characteristics and Assumptions

APPENDIX C. TECHNOLOGY CHARAaERISTICS AND ASSUMPTIONS

The ECM and program assumptions used in this report derive from a variety of sources, including our own

expert opinion and experience. Secondary sources include government and utility studies in the public domain

or available from our direct involvement with research.

The natural gas utilities that were consulted (through published filings with state regulatory agencies) are:

Atlanta Gas Light (ATL)

Elizabethtown Natural Gas (ELIZ)

Long Island Lighting Company (LILCO, now KeySpan)

Minnegasco

New Jersey Natural Gas (NJN)

South Jersey Natural Gas (SJG)

UtiliCorp

Additional information used in this study came from an overview of the Conservation Improvement Programs

offered by the Minnesota natural gas utilities as described in the 2005 report of the State Department of

Commerce. The utilities represented in that study include:

CenterPointEnergy Minnegasco

Great Plains Natural Gas

Interstate Power and Light

Northern Minnesota Utilities

Peoples Natural Gas

Xcel Energy

A few web sites also offer valuable information at the technology level, including:

The California Database for Energy Eficient Resources (DEER), http://eega.cpuc.ca.gov/deer/.

Work files of the Fifth Northwest Electric Power and Conservation Plan, http://www.nwcouncil.org/energy/powerplaplefault.htm. Although the plan does not include gas DSM technologies, many of the technologies can be expected to have similar measure lives and cost.

The research team also consulted the "200 1 DEER (Database for Energy Efficiency Resources) Update Study Final Report," prepared for California Energy Commission.


Vectren DSMAction Plan: Final Report Appendix D. Cost Effectiveness Methodology

APPENDIX D. COST EFFECTIVENESS METHODOLOGY

Cost effectiveness analysis refers to the systematic comparison of program benefits and costs using standardized

measures of economic performance. In this report, cost effectiveness is discussed at both the technology level

and the program level. The assumptions and approach used to calculate technology and program cost

effectiveness are presented in this appendix. Much of the material in this section is taken from the California

Standard Practice Manual: Economic Analysis of Demand Side Management Programs and Projects, October

2001 (SPM 200 1);' which has broad industry acceptance.

Technology Cost Effectiveness

It is desirable to consider some measure of a technology's cost effectiveness in the preliminary stages of

program design. This allows program planners to subjectively tradeoff cost and other attributes of energy

conservation measures (ECM) when considering possible packages and program designs. Cost effectiveness

analysis is less precise at the technology screening stage because estimates of energy savings and costs at the

measure level are subject to a great deal of variance due to interaction with other measures and actual program

implementation. Still, measure cost effectiveness provides a useful metric for consideration along with the

many other factors outlined in the Program Design section of this report.

What is needed at the technology or measure level is a simple measure of cost effectiveness that does not require

assumptions of avoided resource cost, rebates, program delivery cost and other program level details. Levelized

Cost (LC) provides just such a measure by expressing the cost of a measure in annual terms per unit of energy

saved. This allows an easy way to compare and rank order the cost effectiveness of measures. The formula

used for the LC calculations in this report is presented below:

LC=DCosts / DSavings

" IC, DCost =

t=l ( 1 + dIt-l

where:

LC = Levelized cost per unit of the total cost of the resource (cents per them) IC = Incremental cost of the measure or technology DCost = Total discounted costs DSavings = Total discounted load impacts AENit = Reduction in net energy use in year t

50 Prepared by the California Public Utilities Commission (CPUC) and the California Energy Commission (CEC). All formulas and discussion are based on the SPM 200 1. Formulas have been modified to remove peak savings, multiple costing periods, and otherwise adapted to be relevant for use with this project.


Vectren DSMAction Plan: Final Revort Avvendix D. Cost Effectiveness Methodolopv

Although not suited for fuel substitution and load building programs, LC provides an easily calculated way of

comparing measures. Measure cost, savings, useful life, and discount rate are the only assumptions required for

calculating LC.

Program Cost Effectiveness

Many additional assumptions over and above those required for calculating ECM cost effectiveness must be

made when calculating program cost effectiveness. Cost effectiveness of energy efficiency programs involves

describing the economic impact of the program from the perspective of various groups. This analysis required

detailed program budgets and design elements such as rebate levels and other program features. Perspectives,

also called tests, presented in this report are listed in the table below along with the primary benefits and costs

used to compute cost effectiveness.

Table 66. Benefits and Costs by Cost Effectiveness Test

Reference to "nety' indicates that the load used to measure the benefit or cost is net of fiee riders. ECM

installation includes all incremental costs to acquire and install an ECM. Program expenses include all costs

related to delivery of the program and include staffing and overhead, advertising, incentive payments,

administration fees, and monitoring and evaluation expenses.

Various measures of the economic impact are available for each perspective. The two primary measures we will

use in this report are listed below:

Costs ECM installation Increased O&M costs

Lost gas revenue (net) Program expenses ECM installation Program expenses Increased O&M costs Program expenses paid by program administrator

Cost Effectiveness Test Participant

Ratepayer Impact

Total Resource Cost

Program Administrator Cost (formerly named Utility Cost)

Net Present Value (NPV)

Benefit-Cost Ratio (BCR)

Benefits Reduced gas bill Incentive payments Tax credits Decreased O&M costs Avoided gas costs (net)

Avoided gas costs (net) Tax credits Decreased O&M costs Avoided gas costs (net)

In addition to the economic criteria listed above, other criteria may be unique to a given perspective. For

example, simple payback of investment is often cited as an important criterion from the participant perspective.

Each of the perspectives is discussed in detail below including the assumptions and formulas required to

calculate the measures of economic impact. Each of the cost effectiveness tests are discussed below.


Vectren DSMAction Plan: Final Report Appendk D. Cost Effectiveness Methodology

Participant Test This test compares the reduction in energy bills resulting from the program with any costs that might have been

incurred by participants. Other benefits included in this test include incentive payments and tax credits. When

calculating benefits, gross energy savings are used rather than reducing savings for free-riders.

The main value of the Participant Test is that it provides insight into how the program might be received by

energy consumers. The incentive level required to achieve some minimum level of cost effectiveness, for

example, can be useful in program design efforts. It should be noted, however, that consumer decision making

is far more complex than reflected by the Participant Test. For this reason, the test should be used as one

consideration of likely program acceptance and not an absolute indicator.

Ratepayer Impact Measure Test The Ratepayer Impact Measure (RIM) Test measures the impacts to customer bills and rates due to changes in

utility revenues and operating costs caused by the program. Rates will go down if the change in revenues from

the program is greater than the change in utility costs. Conversely, rates will go up if revenues collected after

program implementation is less than the total costs incurred by the utility for implementing the program. This

test indicates the direction and relative magnitude of the expected change in customer rate levels.

The benefits calculated in the RIM Test are the savings from avoided supply costs. These avoided costs include

the reduction in commodity and distribution costs over the life of the program.

The costs for this test are the lost revenues from gas sales and all program costs incurred by the utility, including

incentives paid to the participant. The program costs include initial and annual costs, such as the cost of

equipment (either total cost for a new installation or net cost if done as a replacement), operation and

maintenance, installation, program administration, and customer dropout and removal of equipment (less

salvage value). The decreases in supply costs and lost revenues should be calculated using net savings.

Total Resource Cost Test The Total Resource Cost Test measures the net costs of a demand-side management program as a resource

option based on the total costs of the program, including both the participants' and the utility's costs. Of all the

tests, the TRC is the broadest measure of program cost effectiveness. This makes the TRC Test useful for

comparing supply and demand side resources.

The primary benefit in the TRC Test is the avoided cost of gas. Loads used in the avoided cost calculation are

net of free riders. Tax credits and reductions in annual O&M costs, if applicable, are also treated as a program

benefit (or a reduction in costs). Costs used in the TRC calculations include all ECM installation costs, program

related costs and any increased O&M costs no matter who pays them. Incentive payments are viewed as

transfers between participants and ratepayers and are excluded fiom the TRC Test,


Vectren DSM Action Plan: Final Re~ort A ~ ~ e n d i x D. Cost Effectiveness Methodolorn

Program Administrator Cost Test The Program Administrator Cost Test measures the cost of acquired energy savings considering only the costs

paid by the program administrator. Benefits are similar to the TRC Test but costs are more narrowly defined.

Its primary purpose is for assessing resource acquisition from the perspective of the program administrator. In

this sense, it is similar to the Participant Test in that the test provides a measure of cost effectiveness from a

single perspective that does not include all costs.

Benefits included in the calculation are the avoided cost of gas. Net loads are used for the purpose of

calculating avoided cost of gas benefits. The costs include all administrator program expenses including

incentive payments for ECM installation.

Avoided Cost of Gas Details All tests, except the Participant Test, rely on an estimate of the avoided cost of gas. The details behind these

calculations are presented in the table below. See the Avoided Gas Costs discussion in the Program Cost

Effectiveness section for a description of how these results are used to estimate avoided gas costs.

Table 67. Avoided Cost of Gas Details

Notes: a) NYMEX Mures. February wnlraci (1011 3/05) a) 2003 dollars per million BTU, East Notlh Central Region

Description Amounl Heat 0.119 0.119

Space Heat YFhenn NPV Level'd $fThenn NPV Level'd 0.0476 $12758 $ 1.2758 $1.2044 $ 1.2044 0.0476 S 1.0560 $ 2.17 $ 1.1687 $0.9846 $ 2.03 $1.0973 0.0476 $ 0.9314 O 2.97 $ 1.0935 S 0.8600 $ 2.77 $1.0221

Weighted average wsi of capital 8.34% 1 13.88 9.30 Inflation 3.00% 2 11.242 -19.0% 8.54

Inflation adjusted discount rate 5.18% 3 9.747 -13.3% 7.96 4 8.847 -9.2% 7.45 5 8.212 -7.2% 7.20 6 6.96 7 6.87 8 6.95 9 7.07

Year


Short Term Pet ElAZW5

Outlook a] Change Outlook b]

Vectren DSMAction Plan: Final Report Error! Reference source not found.

APPENDIX E. SUMMARY OF NATURAL GAS PROGRAM REBATES, MINNESOTA

Natural Gas Utility Conservation Rebates

The following is an excerpt from Gas Rebate Facts, Minnesota Department of Commerce, February 2003, pages

1-5.

INTRODUCTION

The following provides general information about the variety and type of rebates being offered to customers of Minnesota's investor owned natural gas utilities. The information has been compiled from programs submitted to the Commissioner in the Conservation Improvement Program filings. The intent is not to provide an exhaustive listing of every rebate being offered, but rather to summarize and describe a range of program offerings. The amount of rebate offered varies depending upon the unique circumstances of the utility; the rebate amount shown is simply what is being offered and is not a recommendation. We hope this information will help persons involved in developing programs become more familiar with rebate possibilities. For more information, please contact Christina Brusven (65 1-282-5008).

RESIDENTIAL PROJECTS

The energy saving focus of residential projects is space heating, which is the largest use of natural gas in the residential sector. Rebates are offered for furnaces, boilers and setback thermostats. Some utilities also provide rebates for domestic water heaters and integrated space and water heating appliances. Rebates are listed for each utility as they were provided to the department; unless otherwise stated, the listed efficiency criterion is the minimum requirement.

Aquila (Northern Minnesota Utilities and People's Natural Gas)

Forced air furnaces (92% AFUE)-$200.


Integrated systems (90% Combined Annual Efficiency)-$250.

ENERGY STAR setback thermostats-$40.

Water Heaters (.62 Energy Factor )-$45.

Centerpoint Energy Minnegasco

Forced air furnaces (92% AFUE)-$100 participant; $15 dealer.

Boilers (85% AFUE)-$100 participant; $15 dealer.

Integrated systems (88% Combined Annual Efficiency)-$15 0 participant; $15 dealer.

Great Plains

Forced air furnaces (92% AFUE)-$15 0.

Integrated systems (88% Combined Annual Efficiency)-$150.

Setback thermostats-$20.

Water heaters (.62 Energy Factor)-$50.


Vectren DSMAction Plan: Final Report Error! Reference source not found

Interstate Power and Light

Forced air furnaces (92% AFUE )-$200; 15 percent dealer incentive.

Boilers (85% AFUE)-$200. - Eligible boiler and furnaces must be listed in the GAMA directory and cannot exceed 300,000

B T U h input.

Setback thermostats-$25.

Water heaters (.62 energy factor)-$50.

Horizontal axis ENERGY STAR clothes washers-$100.

Windows (U value of -35 or less)-$20 per window.

Xcel Energy


Forced air furnaces (94% AFUE )-$loo.

Boilers (85% AFUE)-$100.

Water heaters (.62 Energy Factor)-$50.

COMMERCIAL AND INDUSTRIAL PROJECTS

Projects for this customer class fall into two general categories: prescriptive and custom. Prescriptive projects are similar to residential rebate projects in that they offer a set rebate amount for a specific technology. Custom projects are tailored to the specific customer; rebates are determined by evaluating the project with a benefitlcost model.

Prescriptive Projects


Water heaters 50 gallons or more (.62 Energy Factor)-$150 per unit.

Forced air furnaces less than 225,000 BTU/hr (92% AFUE)-$200.

Forced air furnaces less than 225,000 B T U h (94% AFUE)-$250.

Also provides incentives for heating system retrofit measures.


Foodsewice Equipment

Targeted to customers with large cooking loads such as restaurants, hospitals, schools, etc.


Vectren DSM Action Plan: Final Report Error! Reference source not found.

Boiler System Tune-up

Targets customers using up to 75,000 thenns annually.

Rebate of up to 25% of the tune-up cost with a cap of $200 per boiler and $1,000 per facility.

Heating Systems

Heating System RetroJ;ts

Equipment Type and Size High efficiency forced-air furnace <225,000 B T U h Unit HeatersIDuct Furnaces (all sizes) High efficiency boiler system 4 0 MMBTUs per system Continuous airlfuel modulating Boiler Burners < 10 MMBTUs per system High efficiency boiler system > or equal to 10 MMl3TUs per system

Xcel Energy

Boiler System Rebates

Follows ASHRAE 90.1 and Federal Energy Management Program Standards.

Efficiency Requirements 92% AFUE

83% AFUE

85% combustion efficiency or greater Minimum 6-step modulation system

Requirements determined on a case-by-case basis. Rebates must pass the BENCOST financial modeling criteria.

Equipment Type or Service Steam trap replacement

Continuous Air/Fuel Modulating Boiler Burners <10 MMBTUs per system Single pipe steam balancing

Vent dampers

Boiler reset control

Boiler cut-out control

Customized heating system rebate


Rebate Amount $100/Furnace

10% of equipment cost; $1 ,000Iunit cap $1 ,OOO/MMBTU; $1 0,000lsystem cap $6OO/MMBTU; $6,000/system cap

Rebate will use the Custom Rebate criteria and will vary case-by-case; $50,00O/system cap

Efficiency Requirements Steam trap survey (infrared or ultrasonic evaluation of existing steam trap operation) is required. Minimum 6-step modulation system

Requirements determined on a case by-case basis. Rebates must pass BENCOST financial modeling criteria.

Rebate Amount 3 5% of equipment cost; $10,00O/building cap

$6OO/MMBTU; $6,00O/system cap

25% of equipment cost; $1,000 cap 25% of equipment cost; $250/boiler cap Up to $150/control system; not to exceed equipment cost Up to $150/control system; not to exceed equipment cost Rebate will use the Custom Rebate criteria and will vary case-by-case; $50,00O/system cap

Vectren DSMAction Plan: Final Report Error! Reference source notfound.

Boiler RetroJts, Controls, and Improveinents

Tune-ups 25% up to $250.

Modular burner controls (5 to 1 turndown ratio min.) 25% up to $2,500.

Modular burner controls (10 to 1 Turndown ratio or greater) 25% up to $5,000.

~urbulators - 25% up to $400.

Blowdown heat recovery, stack economizers and 0 2 trim controls - 25% up to $5,000.

Outdoor air reset controls - 25% up to $500.

Stack dampers - 25% up to $250.

Natural Gas Fired Engine Driven Cooling Systems

COP must be greater or equal to 1.95.

Rebates calculated on a custom basis with each installation passing a cost/benefit test. Rebate level is $6/ton plus $8 MMBTU saved.

Other Rebates

Xcel also provides prescriptive rebates for infrared heaters, setback thermostats and hot water heaters.

Rebates are 25% of equipment costs or $1,500, whichever is the least amount.

Custom Projects


All projects must pass societal benefithost test (result must be greater than 1.0).

Buydown to a two-year payback.

Rebate amount is 50% of incremental cost.


Custoln Process Rebates

Provides custom rebates on a case by case basis to industrial dual fuel customers.

Examples of technologies include: process boilers, heat recovery systems, tower melters, and heat treat systems.

Customer receives the lesser of: $.70 per them saved; buy down to a 2-year payback;

50% of incremental equipment cost; 25% of equipment cost.


Vectren DSMAction Plan: Final Report Error! Reference source not found.

Engineering Assistance

Reimburses C/I customers for a portion of the engineering fees for the design and installation of qualifying energy efficient process technologies.

Provides up to $2,500 (not to exceed 50%). If a qualifying process is installed, an additional maximum $2,500 can be rebated.

Industrial Audits

Largest industrial customers may qualify for $5,000 up front.

May also qualify for an additional $5,000 with the installation of qualifying efficient natural gas process technologies.

Great Plains

All projects are pre-screened and must pass societal benefitlcost test.

Maximum rebate is $2,000 or 50% of the cost, whichever is less.

Interstate Light and Power

All C and I customers are served through the custom program.

Includes a detailed energy analysis to identify energy management and efficiency recommendations.

Projects must pass societal benefit/cost screening test.

Xcel Energy

Custom Eflciency Projects

Provides incentives of up to $2 per MCF saved.

Also includes food service equipment.

Energy Design Assistance

Focuses on gas savings for new and major building renovations.

Incentive is $2 per MMBTU saved.


Vecfren DSMAction Plan: Final Report Appendix F. Energy Conservation Measure Assumptions

APPENDIX F. ENERGY CONSERVATION MEASURE ASSUMPTIONS

The table below contains the ECM assumptions used in developing the program budgets and cost effectiveness

analysis. Readers can use this information to see the ECM details that make up each of the proposed programs.

Install Rate refers to the percentage of program participants who actually install each of the ECMs. Cost is the

incremental cost of the ECM over a standard efficiency option. The Therms Saved represent annual savings per

installation and the expected life is the number of years the ECM is expected to deliver savings. Our approach

in developing these assumptions was to consider a broad range of relevant research (see Appendix C) as well as

our experience. Since each assumption is derived from many sources, including our own experience and expert

opinion, it is not possible to map each of the assumptions listed below to any one source.

Table 68. Energy Conservation Measure Assumptions Used in Program Planning and Design


Vectren DSMAction Plan: Final Report Appendix F. Energy Conservation Measure Assumptions

Table 68. Energy Conservation Measure Assumptions Used in Program Planning and Design, Continued

Note: HE = High Efficiency; WH = Water Heat; ES = Energy Star


lndlana utiliw vectren dsm action plan: commission final

Documents