20130930Hermanson Exhibit 3.pdf

Line
1
2
3
Avista Utilities
Summary of ldaho Demand€ide Management Energy Savings and Levelized Costs
January 1,2010 through December 31,2012
Regular income portfolio Limited income portfolio
Electric savings derived from gas DSM programs include the impact of electric to natural gas conversions as well as
savings resulting from natural gas DSM projects. Therm savings derived from electric DSM projects recognize
impacts of electric DSM measures.
DSM Prooram Portfolio Levelized Cost Calculations
Electric DSM Proqram Portfolio Natural Gas DSM Proqram Portfolio
4
5
6
7I
I
10
11
12
13
14
15
'16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Total Resource Cost (TRC)
Weighted average measure life
Discount rate
M\M energy savings
TRC levelized
Program Administrator Cost (PAC)
Weighted average measure life
Discount rate
32,976,939
12.56
6.80%
1 09,1 00
18,016,365
12.s6
6.80%
Total Resource Cost (TRC)
Weighted average measure life
Discount rate
Program Administrator Cost (PAC)
Weighted average measure life
Discount rate
Therms energy savings
PACT levelized
Therms energy savings
TRC levelized u 1 .1iz5
11,824,492
21.17
6.80%
950,822
5,551,544
21.17
6.80%
950,822
$ u.526
Exhibit No. 3
Nos. AVU-E-13 AVU-G-1 3
L. Hermanson, Avista
Schedule 1, Page 1 of 1
M\M energy savings '109,100
PACT levelized
Total portfolio
Case
Line
1
2
Avista Utilities
Summary of ldaho Electric Demand-Side Management Cost-Effectiveness
January 1,2010 through December 31,2012
TOTAL RESOURCE COST TEST Regular income portfolio Limited income portfolio Overall portfolio
Electric program electric avoided cost
Electric program natural gas avoided cost
Electric program non-energy benefits
TOTAL TRC BENEFITS $
Electric program non-incentive utility cost
Electric program customer cost
TOTAL TRC COSTS $
NET TRC BENEFITS $
TRC BENEFIT/ COST RATIO
TOGRAM ADMINISTRATOR COST TEST
$
$
$
$
$
$
$
$
$
61,139,419
(1,316,541)
2,306,200
62,129,078
5,509,020
26,363,799
31,872,819
30,256,259
1.95
Kegurar rncome
portfolio
759,715
124,7U
979,336
$ 61,737,619
$ (1,363,449)
$ 2,514,623
$ 62,888,793
$ 5,633,8M
$ 27,s43,1s5
1,104,120 $ 32,976,939
(344,405)ll$ 2e,e11,8540.6911 1.s1
Lrmlleo rncome
portfolio Overall portfolio
598,200
(46,
3
4
5
6
7
8
10
11
12
13
14
15
16
17
18
19
2A
21
2Z
23
24
25
2A
27
28
29
30
31
32
33
34
35
56
37
38
39
40
41
42
43
44
45
464t
48
49
50
51cz
53
54
Electric program electric avoided cost
Electric program natural gas avoided cost
TOTAL PAC BENEFITS
Electric program non-incentive utility cost
Electric program incentive cost
TOTAL PAC COSTS
NET PAC BENEFITS $
PAC BENEFIT / COST RATIO
PARTICIPANT TEST
42,870,046 $
3.53
Kegurar tncome
portfolio
598,200 61,737,619
46
60.374.170
124,7U
749
(512,24D11$ 42,3s7,8050.52 s.3s
Ltmtteo tncome
portfolio Overall portfolio
$
$
$
$
$
$
61,139,4'19
5,509,020 $5,633,804
11.443,812 12.382.561
Electric program electric bill reduction
Electric program gas bill reduction
Non-energy benefits
TOTAL PARTICIPANT BENEFITS
Electric program incentive cost
TOTAL PARTICIPANT COSTS
NET PARTICIPANT BENEFITS $
PARTICIPANT BENEFIT / COST RATIO
NON-PARTICIPANT TEST
Customer project cost $26,36s,799 $
37,077,169
170,470
$
24,633,852 $
2.65Kegurar rncome
portfolio
$ 25,276,906
2.69
portfolio overallportfolio
543,949
(68,731
37 ,621,118
101 ,739
423 2.514.623
979,336 27,343,135
11 7 561
Electric program electric avoided cost
TOTAL NON-PARTICIPANT BENEFITS
Electric program lost electric revenue PV
Electric program non-incentive utility cost
Electric program incentive cost
TOTAL NON-PARTICIPANT COSTS
NET NON-PARTICIPANT BENEFITS
I.PARTICIPANT BENEFIT / COST RATIO
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 2, Page 1 ot 2
61.139.419 61 .737.619
6'1 ,7s7,619
37,247,639
5,509,020
11.443.812
475,218
124,784
749
37,722,857
5,633,804
561
Line
1
2
3
Avista Utilities
Summary of ldaho Natural Gas Demand€lde Management Cost-Effectiveness
January 1 , 20'l 0 through December 31 , 201 2
TOTAL RESOURCE COST TEST Regular income portfolio Limited income portfolio Overall portfolio!
6
7I
I
10
11
12
13'14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
Gas program natural gas avoided cost $
Gas program electric avoided cost $
Gas program non-energy benefits $
TOTALTRC BENEFITS $
Gas program non-incentive utility cost
Gas program customer cost
TOTAL TRC COSTS
NETTRC BENEFITS $
TRC BENEFIT/ COST RATIO
16,792,478 $
1 ,381,089 $
1 19,766 $
18,293,3s3 $
1 ,561,763
I
10,924,172
7,369,1 61
1.67
307,917
810
187,445
$ 17,100,395
$ 1,381,899
$ 307,211
496,172 $ 18,789,505
Gas program gas avoided cost
Gas program electric avoided cost
TOTAL PAC BENEFITS
Gas program non-incentive utility cost
Gas program incentive cost
TOTAL PAC COSTS
NET PAC BENEFITS
PAC BENEFIT / COST RATIO
13,457,479
3.85
(s27,12s)ll$ 12,e30,7500.37 3.33
PARTICIPANT TEST Regular income portfolio Limited income portfolio Overall portfolio
$
$
$
$
149,609 1,711,372
750.7',t1 113.120
(404,148)ll$ 6,e65,013
0.5511 1.5e
pROGRAM ADMTNTSTMTOR COST TEST Regular income portfolio Limited income portfolio Overall portfolio
$
$
16,792,478 $
1.381.089 $
18,'173,567 $
307,917 ll $
810 lt $
1 7,1 00,395
1 ,381,899
$
$
1 ,561,763 $14e,60s ll $1,711,372
153,92s $686.247 ll $172
30
3'l
32
33
34
35
36
37
38
39
40
41
42
43
Gas program gas bill reduction
Gas program electric bill reduction
Non-energy benefits
TOTAL PARTICIPANT BENEFITS
Customer project cost
Gas program incentive cost
TOTAL PARTICIPANT COSTS
NET PARTICIPANT BENEFITS $
PARTICIPANT BENEFIT / COST RATIO
$
$
$
8,386,091
1,426,754
1 19,766
9,932,61 1 $
3,724,127 $
1.60
.,'
' ::"ll '
$ 8,681,090
$ 1 ,426,910
$ 307,211
$ 1 0,4'15,21 '1
10,'.t't3,120
6,272,948
4,142,263
1.66
$
$
$
294,999
156
187,445
482,600
750,711
64,464
NON-pART;C;pANT TEST Regular income portfolio Limited income portfolio Overall portfolio
$
$
9,362,409 $
'153.925) $
$
$
44
45
46
47
16
49
50
51
52
53
54
Gas program natural gas avoided cost
TOTAL NON.PARTICIPANT BENEFITS
Gas program lost gas revenue PV
Gas program non-incentive utility cost
Gas program incentive cost
TOTAL NON.PARTICI PANT COSTS
NET NON-PARTICIPANT BENEFITS $
NON-PARTICIPANT BENEFIT / COST RATIO
$
$
$
9,812,84s $
1,561,763 $
3,153,92s $
14,528,533 $
2,263,945 $
't.16
295,1 55
149,609
686,247
$ 10,108,000
$ 1 ,711,372
$ 3,840,172
$ 15,659,544
1,440,851
1.09
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 2, Page 2 ot 2
16.792.478 $307,e17 ll $1 7,1 00,395
16.792.478 $
-*'::;lll'
FINAL REPORT
AVISTA 2OL2 IDAHO
ELECTRIC IMPACT
EVALUATION REPORT
August 30, 2013
Avista Corporation
1411 E Mission Ave
Spokane, WA 99220
The Cadmus Group, lnc.
Exhibit No. 3
Case Nos. AW-E-1 3 AVUG-I 3
L.Hermanson, Avista
Schedule 3, Page 1 of75
An Employee-Owned Company . www.cadmusgroup.com
Prepared by:
Danielle C6t6-Schiff Kolp, MESM
Andrew Wood
Jeff Cropp, P.E.
Scott Reeves
M. Sami Khawaja, Ph.D.
Cadmus
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3,Page2ot75
TABLE OF CONTENTS
PORTFOLIO EXECUTIVE SUMMARY.. .........1
Evaluation Activities ........................... 1
Key Findings and Conclusions............. ....................2
Recommendations and Further Analysis....... .........3
1. Residential Electric lmpact Report......... ...................5
1.5 Recommendations ..............41
2. NonresidentialElectric lmpact Report......... ...........43
2.L lntroduction .........................43
2.2 Methodo1ogy.................. .........................45
2.3 Results and Findings ............50
2.4 Conclusions ......55
2.5 Recommendations ..............56
3. Low lncome Electric lmpact Report ........................57
3.1 lntroduction .........................57
3.2 Data Collection and Methodology ..........58
3.3 Results and Findings ............59
3.4 Conclusions ......61
3.5 Recommendations ..............62
4. CFL Contingency Program.............. .........................54
4.L lntroduction .........................64
4.2 Methodo1ogy,................. .........................64
4.3 Overall Program Savings....,... ..................57
5. Portfolio Gross and Net Savings. .........58
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G13
L.Hermanson, Avista
Schedule 3, Page 3 of 75
1.1
1.2
1.3
L,4
5.1
5.2
5.3
5.4
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 4 of 75
DEFINITIONS
Reported Savings - Electricity savings that are reported in Avista's tracking database.
Gross Evaluated Savings - Electricity savings that have been verified through evaluation activities such
as records review, verification surveys or site visits, and engineering analysis.
Realization Rate - The ratio of gross evaluated savings over the reported savings.
Net Evaluated Savings - Net savings signify the portion of savings directly attributable to the program;
savings that would have othenryise not occurred without program influence.
Net-to-Gross - The ratio of net evaluated savings to gross evaluated savings.
Savings Goal- The lntegrated Resource Planning (lRP) savings goal.
Achievement Rate - The ratio of the net evaluated savings over the savings goal.
ilt Exhibit No. 3
Case Nos. AW-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 5 of75
PORTFOLIO EXECUTIVE SUM MARY
For several decades, Avista Corporation has been administering DSM programs to reduce electricity and
natural gas energy use for its portfolio of customers. Most of these programs have been implemented
in-house, but a few utilize external implementers. Avista performed a potential study for lD in 2011 to
determine the savings goals for 2012 and 2013. Avista contracted with Cadmus to complete process and
impact evaluations of the company's20L2 electric demand-side management (DSM) programs. Cadmus
completed a combined 2010-2011 electric report for both Washington and ldaho. This report presents
our impact findings for the PY 2OL2 electric portfolio for ldaho.
Evaluation Activities
We conducted the evaluation using a variety of methods and activities, as shown in Table 1.
Residential
Nonresidential
Low lncome
Residential/
Nonresidential
Simple Steps, Smart
Savings'"
Second Refrigerator
and Freezer Recycling
ENERGY STARO
Products
Heating and Cooling
Efficiency
Weatherization/Shell
Water Heater
Efficiency
ENERGY STAR Homes
Space and Water
Conversions
Prescriptive Programs
Site-Specific
EnergySmart Grocer
Low lncome
Programs
CFL Contingency
Savings Results
Overall, the ldaho portfolio achieved a98.7% realization rate, and acquired 37,483,952 kWh in annual
gross savings (Table 2).
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 6 of75
Table 1. 2012 Electric Programs Evaluation Activities
Table 2. 2012 Reported and Gross Evaluated Savings for ldaho
Residential*
Nonresidential*
Low lncome
L3,627,696
24,093,322
274,9L3
14,098,435
23,L04,034
28L,483
37,483,952
L03.5%
95.9%
L02.4%
94.7%Total 37,995,93L ,
*lncluding compact fluorescent lamp (CFL) Contingency savings.
Table 3 shows evaluated gross and resulting net savings for ldaho's 2012 DSM programs.
Residential*
Nonresidential*
Low lncome
Total
* lncl uding CFL Contingency savings
Residential
Nonresidential
Low lncome
Total
14,098,435 93%
23,L04,O34 ,,79%28L,483 \O0%37,83,952 U%
L3,LO7,862
18,250,606
281,483
31,539,951
Table 4 shows net evaluated savings, as compared to the lntegrated Resource Plan (lRP) goal of
17,115,000 kWh. The IRP states its goal as a portfolio-level target; so, for purposes of sector-level
comparison, Cadmus adopted the Avista Business Plan goals by sector, and applied those proportions to
the IRP target. The 2012 program year achieved L84.9% of the IRP target in ldaho with 31,639,951 kwh.
Even excluding the CFL Contingency savings, ldaho still surpassed the IRP goal, at 111.9% with
19,151,861 kwh.
7,495,L08
8,423,0N
L,L96,892
17,115,000
L3,L07,862
18,250,606
28L,483
31,639,951
L74.9%
2L6.7%
23.5%
t8/..9%
Key Findings and Conclusions
Residential
o For PYlOL2, residential electric programs produced 13,L07 ,862 kWh in net savings, yielding a
LO3.5o/o overall realization rate. Residential electric savings achieved L749% of !RP goals.
o Overall, residential electric customers responded well to the programs, often installing several
measures within the same year.
o Tracking databases proved adequate for evaluation purposes, providing sufficient contact
information, and measure and savings information. The database review confirmed the
information was reliable and accurate.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 7 of75
Table 3. 2012 ldaho Net Savings
Table 4. 2012 Reported and Gross Evaluated Savings for Idaho
o All rebated measures had been installed and continued to operate. With one exception, all
measures reviewed met the program qualification standards.
Nonresidential
. ln general, Cadmus determined that Avista implemented the programs well. The overall
nonresidential electric portfolio achieved a95.9% realization rate, upon comparing gross
evaluated savings to gross reported savings, and achieved 2L6.7% of the IRP goal.
o Power metering on one industrial process measure indicated lower-than-expected post-
installation power consumption, which increased energy savings.
o Light logging on three projects identified a slight decrease in operating hours from the
reported values.
o Cadmus applied algorithms different from those used by Portland Energy Conservation, lnc.
(PECllto determine energy savings for electrically commutated motors (ECMs). This resulted in a
slight decrease in energy savings.
o One project installed PC Network Controls in 2fi)9, but did not provide the final data that
demonstrated a reduction in consumption until 2012. Avista paid the incentive in20t2, but the
participant reported deactivating the system soon after.
Low lncome
o Avista's low income electric programs produced 28L,483 kWh in savings, yielding an overall
LO2.4% realization rate. Low income electric savings achieved 23.50/6 of IRP goals.
Recommendations and Further Analysis
Residentia!
Based on the evaluation results, Cadmus offers the following recommendations for Avista:
o List energy factors (or, at least, model numbers) for appliances. lncluding more information
about the actual efficiency of equipment installed would allow greater accuracy in estimating
gross energy savings.
o lf possible, include existing equipment information.
o lf the ENERGY STAR Clothes Washer measure is reinstated, consider moving all rebates to the
electric program.
The following research recommendations draw upon this impact evaluation's results and from known
future changes to program requirements:
o Perform a targeted bi!!ing analysis on weatherization participants using both electricity and gas
to heat their homes.
Exhibit No.3
Case Nos. AW-E-13 AVU-GI 3
L.Hermanson, Avieta
Schedule 3, Page 8 of 75
. Perform a billing analysis on ENERGY STAR homes using a nonparticipant comparison group
once enough homes have participated under the new requirements to justifo conducting
the work.
Nonresidential
Cadmus recommends that Avista continue to offer incentives for measure installation through the
evaluated programs. Based on the results from the ldaho projects, the following recommendation
focuses on improving program energy savings impacts and evaluation effectiveness:
o Work with participants to accelerate the process of claiming energy savings and paying the
project incentive. This preferably should occur within one year of measure installation,
depending on Avista's requirements for post-installation data on the particular project.
Low lncome
The impact evaluation revealed several areas where program performance and savings accuracy could
be improved. Consequently, Cadmus recommends Avista consider the following:
o lnclude a comparison group in future billing analyses.
o Work with ldaho agencies to provide refrigerator replacements.
e Consider targeting high-use customers.
o Track and compile additional data from agency audits.
o Consider analyzing easy-to-quantifi7, non-energy benefits, which could be added to program
cost-effectiveness re porti ng.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 9 of75
1. RESIDENTIAT ETECTRIC IMPACT REPORT
t.l lntroduction
During the 2012 program year, Avista's residential electric demand-side management (DSM) programs
in ldaho reported unverified savings of 5,073,009 kWh for 435,837 measures. The 2012 DSM residential
electric programs included :
o Simple Steps, Smart Savings'"
o Second Refrigerator and Freezer Rerycling
o ENERGYSTARO Products
o ENERGY STAR Homes
o Heating and Cooling Efficiency
o Water Heating
o WeatherizationMeasures
o Space and Water Conversions
This report explains the methods used to qualify and verify these savings.
1.1.1 Evaluation Methodology
Using the following methods, Cadmus designed the impact evaluation to verify tracked program
participation and energy savings:
e Data collected in the tracking database;
o Online application forms;
o Phone surveys; and
. Applicable deemed values developed for Avista's technical reference manual (TRM).1
As shown in Table 5, Cadmus employed up to two evaluation methods and activities for each program.
ln 2011's first quarter, Cadmus created a TRM for use in performing deemed measure savings calculations,
and updated it where necessary for the 2012 program year. The TRM first looks to the RTF.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 10 of75
Table 5. Evaluation Methodology
Simple Steps, Smart Savings'"
Second Refrigerator and Freezer Recycling
ENERGY STAR Products
Heating and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
Space and Water Conversions
L.L.z Energy Savings
Table 6 shows aggregated evaluated gross savings and resulting realization rates by program.
Table 6. Reported and Evaluated Gross Savings
Simple Steps, Smart Savings'"
Second Refrigerator and Freezer Rerycling
ENERGY STAR Products
Heating and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
Space and Water Conversions
Total
3,330,478
268,752
380,897
676,843
37,575
8,933
24,698
344,734
5,o73,oog
3,914,480
350,968
193,953
67L,428
37,373
8,861
24,698
34L,977
5,543,749
LL75%
L30.6%
50.9%
99.2%
99.2%
99.2%
LOo.O%
99.2%
1o9.3%
Cadmus evaluated gross savings of 5,543,748 kwh through the installation of 435,837 measures during
PY 2012. Table 7 shows reported measure counts. Overall, residential electric programs achieved an
adjusted LOg.3o/o gross realization rate.
Simple Steps, Smart Savings'"
Second Refrigerator and Freezer Recycling
ENERGY STAR Products
Heating and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
Space and Water Conversions
Total
433,777
327
L,79L
769
49
75
11
38
436,837
Table 7. Avista 2012 DSM Programs Reported Measure Counts in ldaho
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 11 ol75
1.2 Methodology
1.2.t Sampling
Cadmus randomly sampled program participants to complete verification surveys, and another,
separate random sample of participant applications for documentation review. Where possible,
sampling was designed to utilize similarities between programs and states to decrease necessary sample
sizes, while maintaining sufficient confidence and precision. The following subsections describe methods
used to select the required samples.
Record Review Sampling
To determine the percentage of measures incented that qualified for the Avista's programs, Cadmus
designed sample sizes to achieve 90% confidence and t10% precision levels for each application type,
across both states and fuels served by Avista's programs. Cadmus randomly selected individual
participant measures for a record qualification review from the 2OL2 gas and electric program
populations. However, if a customer applied for multiple rebates on the same application form during
the program year, the record review checked all measures included in the application for qualification,
whether for electric or gas.
Table 8 shows the number of record reviews completed for unique accounts and unique measures.
Total Participants Reviewed
fotat fvteisures neviewea
Suruey Sampling
For program-level survey results, Cadmus designed participant survey sample sizes to achieve 90%
confidence and t10% precision levels for each program. The participant survey sampling plan drew upon
the following multiple factors:
o The feasibility of reaching customers;
r The program participant population; and
o Research topics of interest.
Fueltypes did not factor into survey sampling.
Cadmus did not survey home buyers for the ENERGY STAR New Homes program as home builders
received the rebates. Surveys for the Simple Steps program could not be conducted as it is an upstream
program without participant records. The evaluation completed 274 surveys with ldaho participants.
Table 9 shows the number of surveys achieved and the resulting absolute precision for each program.
The absolute precision achieved did not always meet the t10% goal (due to low program participation),
but falls safely within the portfolio precision goal of 90/10.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L.Hermanson, AMsta
Schedule 3, Page 12 ol 75
Table 8. Measure Level Record Review Completes
Table 9. Participant Survey Sample Sizes and Savings-Weighted Precision Estimates by Program
Watef He_?!iLg
ENEiGY_STAR Products
He.ating and Cooling Efficiency, ;trf*. ; ;
L 2*_E_etlig gElqr qF re e41!9qp!!1e
I Weatherization and Shell Measures
-?84-__ ---?9-- !4tr 1L. - ts%
__!89q L _ _J9 _ 3&) ___ _1\ r r1o% l-__Il1q[- -6\ Earl -- 6i i - - - ls% I
_LzL' 6!' 27.L%_ _ 31 41%
Cadmus randomly called program participants included in the survey sample frames. As shown in
Figure 1, geographic distributions of survey respondents clustered around urban centers within Avista's
service territory (specifically, the cities of Spokane, Pullman, Moscow, and Lewiston).
Figure 1. Geographic Distribution of Participant Sutvey Completes
8 Exhibit No.3
Case Nos. AVU-E-l3 AVU-G13
L,Hermanson, Avista
Scfiedule 3, Page 1 3 of 75
L.2,2 Data Collection and Analysis
Record Review
Cadmus reviewed all records for the selected account sample, using the data they contained to check for
completion and program compliance. Measures qualified if all data in the application complied with
program specifications. As the evaluation randomly sampled customers by application type (and several
measures can be found on different application forms), Cadmus tracked qualification rates at the
application type level.
The review revealed one improperly issued insulation rebate on a Home lmprovement application (it
had an existing R-value above the participation requirements). Applied qualification rates include
this result.
Surueys
Cadmus contracted with Discovery Research Group (DRG)to conduct surveys with sampled participants.
To minimize response bias, DRG called customers during various hours of days and evenings (including
weekends), and made multiple attempts to contact individual participants. Cadmus monitored survey
phone calls to ensure accuracy, professionalism, and objectivity. Analysis addressed survey data at the
program leve! rather than the measure level, and weighted survey results at the portfolio level by
program participation to ensure proper representation.
Datobose Analysis
Cadmus reviewed the participant database Avista provided to check for inconsistencies in tracked
savings and measure duplications. This review did not identify inconsistencies in data tracking. All
tracked savings were based on the 2012 Avista TRM.
Unit Energy Savings Analysis
When necessary, Cadmus updated the unit energy savings (UES) achieved by residential measures based
on new survey data of Avista participants, improved analysis methodologies, recent decisions by the
Regional Technical Forum (RTF), and the results of the Residential Building Stock Assessment (RBSA), all
of which are incorporated into our TRM. Each section below describes the changes made.
1.2.3 Verification Rates
Cadmus determined verification rates for each program (but analysis was performed at the measure
level). Where applicable, the review covered the following topics:
o Checking the database tracked the correct measures;
o Accounting for correct quantities; and
o Determining whether units remained in place and were operable.
All measures researched remained in place and were operable, resulting in a 100% verification rate
across all programs.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 14 ol 75
L.2,4 Measure Qualification Rates
Cadmus considered a measure qualified if it met the requirements particular to its category, such as
receiving an ENERGY STAR certification or achieving program minimum efficiency standards. When
necessary, the evaluation included online database searches for model numbers and noted
characteristics necessary to verify achievement of all qualifications.
Of the entire verification sample, Cadmus identified one nonqualified measure:
o An attic insulation project had a base case condition that should have prevented it
from qualifying.
1.3 Program Results and Findings
1.3.1 Overuiew
Cadmus analyzed data records, maintained by either Avista or an implementation contractor, to
determine appropriate UES and measure counts for each supported measure within each program. The
end result is the total adjusted gross savings for each measure and program as well as overall realized
savings for each program.
Cadmus followed the same steps for calculating adjusted gross measure savings for all programs, except:
Simple Steps, Smart Savings*, Second Refrigerator and Freezer Recycling and Residential
Weatherization (which necessitated individual methodologies). The calculations required the following:
1. Reviewing the program database to determine if adjusted measure counts correctly represented
the number of installations.
2. Conducting a phone survey or site visit to verify the installation occurred within Avista's
service territory.
3. Calculating verification and qualification rates.
4. Calculating deemed measure savings for products rebated during the program period.
5. Applying verification and qualification rates and deemed savings to the measure counts to
determine the adjusted gross savings for each measure.
Details regarding the calculation methods used for Simple Steps, Smart Savings*, Second Refrigerator
and Freezer Recycling, and Residential Weatherization follow in their specific sections, below.
L.3.2 Simple Steps, Smart Savings"
Progrom Description
An upstream incentive program, Avista's Simple Steps, Smart Savings * serves as an effective alternative
to tradational mail-in incentives, given its ease of participation, widespread accessibility, and low
administrative costs. Such programs allow the utility's incentives to pass directly from manufacturers to
retailers, which then reduce prices to their customers. The program motivates retailer participation by
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule3, Page 15of75
10
reducing bulb prices without causing a loss in profits. For the customer, participation may occur so
seamlessly they remain unaware that they have purchased an incentivized bulb or participated in a
utility program.
Upstream programs, however, pose particular evaluation challenges because calculating metrics-such
as in-service rates (lSR) and attributions-traditionally rely on finding purchasers of incentivized
products. ln determining program savings, Cadmus referred to:
o The Northwest RegionalTechnical Forum (RTF) UES assumptions;
o The Residential Building Stock Assessment (RBSA) results;
o Avista's program records; and
o The CFL Contingency Program (discussed in Chapter 4).
The program incents various compact fluorescent lamp (CFL) products, from standard twist bulbs to
specialty bulbs (including three-way, reflector, dimmable, globe, and others). As standard twist bulbs
and speciahy bulbs require unique assumptions, Cadmus analyzed each separately.
Anolysis
This program utilizes six different parameters to inform the calculation of gross savings for the lighting
component: CFL wattage (CFL Watts); delta watt multiplier (DWM); hours-of-use (HOU); days-per-year;
waste heat factors (WHF); and lSR. The following algorithm shows annua! energy lighting savings:
@r@r@r@r(Dr@=C
Where:
Wattage of the CFL
The difference in wattage between the baseline bulb and the CFL
divided by the CFL's wattage
Daily lighting operating hours
Days per year (355)
An adjustment representing the interactive effects of lighting measures
on heating and cooling equipment operations
The percentage of units installed
CFL Watts
DWM
HOU
DAYS
WHF
rsR
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-GI3
L.Hermanson, Avista
Schedule 3, Page 16 of75
71
The annual savings algorithm is derived from industry-standard engineering practices, consistent with
the methodology used by the RTF for calculating energy use and savings for residential lighting. The
following sections discuss each component in detail.
CFL Watts
According to Avista's reported sales, the program incented over 456,746 CFLs. Cadmus reviewed Avista's
sales database and verified approximately 433,777 CFLs. This discrepancy likely resulted from monthly
adjustments made in the database, which could have caused over or undercounting.
I 327,350_.1 12,9,396 _1 433lll__:
Avista sales data included: CFL wattage, units sold, and bulb type. Cadmus analyzed savings for each
bulb type separately. Analysis for three-way bulbs used the middle wattage. ln PY 20L2, the standard
twist and specialty lamps sold had average weighted CFL wattage of t6.2 watts and 15.6 watts,
respectively.
DWM
Cadmus relied on the RTF methodology for both standard twist and specialty bulbs for each wattage and
type of bulb sold. The standard twist bulb DWM used by the RTF assumed the Energy lndependence and
Security Act (EISA) of 20072 impacted the baseline incandescent wattage, per the schedule shown in
Table 11. EISA did not impact the baseline wattage for specialty CFLs. The RTF uses this table to reduce
the assumed average, standard twist baseline bulb wattage for lOt2 by replacing all 85 W to 150 W
incandescent bulbs with 72 W bulbs in the calculations. The RTF analyses produce average baseline
wattages and average installed CFL wattages for each bulb type; these can then be used to calculate the
DWM for all bulbs of that type.
January L,20t4 I 29 W (3LO-749lumens)9-11W CFL (44G-600 lumens)
13-15 W CFL (75f900lumens)
T-
1A-ZO W CFL (1,100-1,300 lumens)
23-26 W CFL
' EISA 2007. Public Law 110-140. December Lg,2OO7. Section 121 Stat. 1577
Table 10. Total Reported and Evaluated CFLs Sold byYear
Table 11. Assumed EISA Effectiveness Schedule, Standard Twist CFLs
Exhibit No. 3
Case No3. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 17 ol75
12
i (approx. 1,690 lumens)
This evaluation calculated energy savings for each wattage and bulb type purchased during the program
year. Cadmus determined the baseline wattage for each bulb, based on the type of bulb purchased and
the lumens produced by that bulb. Looking up stock keeping unit (SKU) numbers in the ENERGY STAR
lighting database provided bulb !umens,3 a procedure matching 91% of the bulbs sold. For the remaining
9% of bulbs, Errorl Reference source not found. estimated the bulb's lumen output (Cadmus developed
he regression equation using the ENERGY STAR lighting database):
Equation 1. Estimating CFL Lumens
C FL _ Lumens = 68.7 39 x C FL Wattage - 56.549
Figure 2 and Figure 3 compare lumens determined by the lookup method and lumens determined using
the regression equation, along with the percentage of program sales for the wattage and type. The
charts indicate that the regression method provided a better match for looking up standard twist CFLS
than specialty bulbs. Cadmus assumed the lumen output estimated by the regression as adequate for
both types of bulbs, due to the Iow percentages of sales volumes for which the regression was required.
Figure 2. Results of Lumens Determination, Standard Twist CFLs
5,000
4s00
4000
3,500
p 3,000
I z,soo
=J 2,000
1,500
1,000
500
0
40%
35%
32% 3o28% 2
24% g,u
20% E
t6% z
L2% i,c8%t
4%
o%
CFL Wattage
-%
of Sales, Lumens by Lookup
-ls6gps
by Lookup -%
of Sales, Lumens by Regression
-Lumens
by Regression
a
/
I I
,
/
-/
zar-/
_--I
rl lrrll
9 1011L2t3 141518t9202326 30324042 5568
ENERGY STAR website:
htto://www.enerwstar.eovlialoroducts/orod lists/comoact fluorescent liqht bulbs orod list.xls
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G.13
L.Hermanson, Avista
Sdredule 3, Page 18 of 75
73
Figure 3. Results of Lumens Determination, Specialty CF[s
3,000
2,500
2,000
6cI r,soo
=
1,000
s00
0 5 7 9 11!2 13L4 15161819202L22232526 40
CFLWattage
a% of Sales, Lumens by Lookup J% of Sales, Lumens by Regression
-lslngps
by Lookup
-lumsns
by Regression
L8%
L5%
L2%
9%
6%
3%
o%
I
I A /
^ I\-J
^l
I
z<-a"'
V-:l I I I I
Cadmus then determined the baseline wattage for each bulb, based on the CFL's lumen output and if
the bulb included a reflector, as EISA did not affect reflector bulbs. Table 12 and Table 13 show the
schedules used to determine baseline wattages for bulbs included in PY 2012.
Exhibit No.3
Case Nos. AW-E-I 3 AW-G13
L.Hermanson, Avista
Schedule 3, Page 19 of75
Table 12. Baseline Wattage based on CFL Lumens, Non-Reflector Bulbs
_50
75
72
150
200
68,t7L I 15.7
14
Table 13. Baseline Wattage based on CFL Lumens, Reflector Bulbs
H19
420-560
561-837
838-1,203
L,204-L,68L
L,682-2,339
2,34V3,075
Specialty
30
45
65
75
90
t20
L75
11.00
14.74
L4.93
L9.70
23.55
26.00
N/A
519
L,27L
64,779
4,2t3
LO,8t2
319
0
0.1
0.3
L4.9
1.0
2.5
0.1
0.0
The evaluation then calculated the DWM for each bulb using the baseline wattage and the purchased
CFL wattage.
Table 14 compares the current DWM assumed by the RTF and the DWM determined through the
evaluation. Differences occurred due to the distribution of sales expected by the RTF and those achieved
by the program. The program records indicated 57% ol the standard twist bulbs rebated were 13 W or
14 W CFLs, which presumably replaced a 50 W incandescent.
Ail
Three-Way
Dimmable
Cold Cathode Candelabra-decorative
Cold Cathode Candelabra-primary
CFL Candelabra
Dimmable Reflector
Globe
Outdoor
Reflector
Any Specialty CFL
2.38
2.O5
2.68
4.00
4.00
3.79
3.23
2.98
2.34
3.23
3.L2
2.96
2.74
N/A
2.74
N/A
2.84
2.86
3.19
3.13
HOU
Cadmus estimated standard twist CFL HOU for residential installations using Avista's survey of room
types and a multistate modeling approach, built on light logger data collected from five states: Missouri,
Michigan, Ohio, Maine, and Maryland.4 The Maine HOU study, completed in the past year, was added to
the model used for the previous evaluation. A regression statistical model calculated the average HOU,
using combined multistate, multiyear data. Cadmus used the multistate model's estimate of HOU by
o The Cadmus Group, \nc.2070 Evoluotion, Meosurement, ond Verificotion Report. Dayton Power and Light.
March 15,2011
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 20 ol 75
Table 14. Comparison of RTF 2012 DWM to Evaluation DWM
15
room type, weighted based on Avista's survey results to determine an overall average HOU of 2.38, a 3o/o
reduction from the 2.45 estimated previously.
Similar to our DWM analysis, HOU for specialty CFLs were derived from the current approved RTF
assumptions.
Though the Simple Steps, Smart Savings'" program could introduce bulbs into residential and
commercial applications, an all-residential application presented the more conservative assumption. As
compelling evidence did not exist to assume a proportion of commercial sales, Cadmus exclusively used
residential assumptions in this analysis.
Waste Heat Factor
The WHF is used to account for the change in annual HVAC energy, either lost or gained, due to a
reduction in facility lighting energy. Cadmus based the WHF on SEEM building models, developed by the
Northwest Power and Conservation Council. These SEEM building models estimate the change in HVAC
equipment energy use resulting from a change in lighting technology (e.g., from incandescent lamps to
CFLs). ln general, the models account for the interaction using load shape profiles of the HVAC and
lighting equipment, based on dwelling occupancy.
The Counci! uses an inherently conservative method, as it assumes a closed shell (i.e., all interior lamps),
including ceiling recessed cans contained in a closed system. Thus, heat produced by the bulbs enters
the building. ln reality, waste heat could transfer out of the conditioned space.
Cadmus based the calculation on Avista's share of electric heating equipment,s along with its associated
efficiencies and surveys of interior and exterior distributions, producing a WHF of 89.8%.5
Cadmus used the commercial WHF of 85.5Yo provided in the 6th Power Plan.
tsR
The program's ISR was derived from the results of the 2012 Residential Building Stock Assessment
(RBSA), which determined the CFL storage rate for each home visited. The RTF recently accepted and
approved this storage rate.7 All PY 2Ot2 bulb purchases had a 76% assumed first-year lSR.
Cadmus considers the utilized Council method inherently conservative as it assumes the remaining24%
of bulbs in storage never provide energy savings. Research indicates almost all bulbs will be installed
within three years of purchase. Despite its conservative nature, the evaluation assumed the RTF
methodology presented the appropriate method for determining energy savings in ldaho.
Avista equipment-type saturations derived from a 2011 participant survey for the CFL Contingency Program.
Given an RTF WHF ot 86.4% and an adjusted Avista WHF of 89.8%.
htto://rtf . nwcou ncil.orelm easures/measure.aso?id=142
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 21 ot 75
5
6
7
16
Results ond Findings
Table 15 compares the current approved RTF assumptions for CFLs to the assumptions used in this
evaluation, and the resuhing UES.
CFL Watts
i DWM (Weighted Average)_*L_ ?.38i r.goI HOU (Weighted Average) _
36s i
Table 15. Comparison of Current RTF Assumptions to PY 20t2Assumptions
I Daysiwnr :
-
i tsR
**
Overall Program Savlngs
For PY 2012, Avista's reported ldaho savings of 3,330,478 kwh and evaluated savings of 3,914,480 kWh,
as shown in Table 15. Determining the regional distribution of purchased CFLs drew upon Avista's
service territory of residential customers, with two-thirds in Washington and one-third in ldaho.
Table 16.Simple Steps, Smart Savings * PY 2012: Reported and Evaluated Total Savings
Avista-All 9,265,946
7LL,578 3,088,549 825,831
An 118% realization rate resulted for PY 20L2for al! bulbs.
1.3.3 Second Refrigerator and Freezer Recycling
Summory of Progrom Participotion
Cadmus reviewed the participant database, maintained by JACO, the program implementer, to test the
reliability of program data. As shown in Table 17, the program recycled 327 units during PY 2012. Some
participants recycled more than one appliance through the program.
Exhibit No. 3
Case Nos. AW-E-I 3 AVU-C-I 3
L.Hermanson, Avista
Schedule 3, Page 22ot75
17
Table 17. Program Participation by Measure
As shown in Figure 4, refrigerator configurations did not change substantially during the last two
program years.
Figure 4. Refrigerator Configurations by Program Year
LOO%
90%
80%
70%
60%
so%
40%
30%
20%
to%
W"
I Bottom Freezer
' Side-by-side
I Single Door
lTop Freezer
As shown in Figure 5, the program recycled more upright freezer units than chest units in 2012.
Exhibit No.3
Case Nos. AW-E-13 AVU-GI3
L.Hermamon, Avista
Scfiedule 3, Page 23o175
18
Figure 5. Freezer Configurations by Program Year
L00%
90%
80%
70%
50%
so%
40%
30%
20%
t0%
0%
I Upright
I Chest
ln 20L2, recycled refrigerators averaged 28 years old, with 18 cubic feet of internal capacity. Recycled
freezers averaged 36 years old, with 18 cubic feet of internal capacity.
Determination of Averoge Annual Gross Sovings
Cadmus developed a multivariate regression modelto estimate gross UEC for retired refrigerators and
freezers. Mode! coefficients were estimated using an aggregated in situ metering dataset, composed of
over 600 appliances (metered as part of five California, Wisconsin, and Michigan evaluations, conducted
between 2009 and 2OL2l. These evaluations offered a wide distribution of appliance ages, sizes,
configurations, usage scenarios (primary or secondary), and climate conditions. The diversity of the
Uniform Methods Project and RTF Protocols
Recent guidelines developed by the U.S. Department of Energy (DOE) informed Cadmus' impact
evaluation methodology for the 2Ot2 prograrn year. ln 2011, DOE launched the Uniform Methods
Project (UMP), intending to "strengthen the uedibility of energy sovings determinotions by improving
EM&V, increosing the consistency ond tronsporency of how energy sovings ore determined."s
The UMP identified seven common residential and commercial DSM measures, and enlisted a set of
subject matter experts to draft evaluation protocols for each measure category, with refrigerator
rerycling one of the seven identified measures. The DOE recruited Cadmus to manage the UMP process
and to serye as the lead author for the refrigerator recycling protoco!.
t U.S. Department of Energy. "About the Uniform Methods Project." Last modified January 2L,20L3. Accessed
June d 2013. htto://wwwl.eere.enercv.qov/office eere/de umo about.htm I
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 24 of75
19
Through a collaborative process that included reviews by a technical advisory group and a steering
committee as well as a public review and response period, the UMP resulted in a set of protocols
(including one for refrigerator recycling) capturing the collective consensus of the evaluation
community. Each protocol established broadly accepted best practices for evaluating key measures in
the category, including identifying and explaining key parameters, data sources, and gross- and net-
related algorithms.
This evaluation followed the methodology outlined in the refrigerator recycling protocol, which largely
mirrored the method Cadmus used in the 201G-2011 program evaluation, except for changes
recommended by the UMP. A discussion follows of the two most notable changes, with each discussed
in greater detail in the Errorl Reference source not found. and Net-to-Gross (NTG) sections.
L. Prospective Pan-Use. The UMP recommends assessing part-use based on how the recycled
appliance likely would have been used if not recycled (not on how it was previously used). For
example, if a primary refrigerator would have become a secondary refrigerator independent of
the program, Cadmus based its 2012 part-use on the average usage of secondary refrigerators
rather than primary refrigerators.
2. Secondary Morket lmpocts. The UMP recommends evaluations utilize a grid-level approach to
estimating net program savings. Therefore, in20t2, Cadmus considered the program's impact
on the used appliance market. The secondary market impact adjustment accounted for changes
in the availability of used appliances resulting from the program.
DOE's Websitee provides more information about the UMP.
Refrigerator Regression Model
Table 18 shows the variables used to estimate refrigerators' annual energy consumption and its
estimated parameteB.
U.S. Department of Energy. "Uniform Methods Project for Determining Energy Efficiency Program Savings."
Last modified April 9, 2013. Accessed June 4, 2013. http://wwwl.eere.energy.gov/office_eere/de_ump.html
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 25ot75
20
Table 18. Refrigerator UEC Regression Model Estimates
(Dependent Variable = Average Daily kWh, R-square = 0.30)
lntercept.,,
Age (years)
Dummy: Manufactured Pre-1990
Size (ft.3)
Dummy: Single Door
Dummy: Side-by-Side
Dummy: Primary
lnteraction: Unconditioned Space x HDDs
lnteraction: Unconditioned Space x CDDs
0.805
0.021
1.036
0.059
-L,75L
L.L2O
0.s50
-0.040
0.026
0.166
0.1s2
<.0001
o.o44
<.0001
<.0001
0.008
0.001
0.188
Results indicated:
o Older refrigerators experienced higher consumption due to year-on-year degradation.
o Refrigerators manufactured before the 1990 NAECA standard consumed more energy.
o Larger refrigerators consumed more energy.
o Single-door units consumed less energy, as these units typically did not have full freezers.
o Side-by-side refrigerators experienced higher consumption due to greater exposure to outside
air when opened and due to the through-door features common in these units.
o Primary appliances experienced higher consumption due to increased usage.
o At higher temperatures, refrigerators in unconditioned spaces consumed more energy.
o At colder temperatures, refrigerators in unconditioned spaces consumed less energy.
Freezer Regression Mode!
Table 19 shows the freezer model's details.
Table 19. Freezer UEC Regression Mode! Estimates
(Dependent Variable = Averate Daily kWh, R-square = 0.38)
lntercept
Age (years)
Dummy: Manufactured Pre-1990
Size (ft.3)
Dummy: Chest Freezer
lnteraction: Unconditioned Space x HDDs
lnteraction: Unconditioned Space x CDDs
-0.95s
0.045
0.543
0.120
0.298
-0.031
0.082
0.237
0.001
0.108
0.002
0.292
<.0001
0.028
Results indicated:
o Older freezers experienced higher consumption due to year-on-year degradation.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 26 ot 75
27
. Freezers manufactured before the 1990 NAECA standard consumed more energy.
o Larger freezers consumed more energy.
o Chest freezers experienced higher consumption.
o At higher temperatures, freezers in unconditioned spaces consumed more energy.
o At colder temperatures, freezers in unconditioned spaces consumed less energy.
Extrapolation
After estimating the final regression models, Cadmus analyzed the corresponding characteristics (the
independent variables) for participating appliances (as captured in the JACO database). Table 20
summarizes program averages or proportions for each independent variable.
* Cooling Degree Days (CDDs) and Heating Degree Days (HDDs) derive from the weighted average from Typical
Meteorological Year (TMY3) data for weather stations that Cadmus mapped to participating appliance ZIP codes.
TMY3 uses median daily values for a variety of weather data, collected from 1991-2005.
For example, using values from Table 19 and Table 20, Cadmus calculated the estimated annual UEC for
2012 freezers as:
20L2 Freezer UEC = 365.25 dalts * (-0.955 + 0.045 ," [35.79 years old] + 0.543 *
1860/o units manuf actured pre - 1990] + 0.L20 * ll8.l4 f t.3 I + 0.298 *
l24o/o units that are chest freezers] + 0.082 ,r, 10.52 Unconditioned. CDDsI - 0.031 *
ILL.B4 Unconditioned HDDsI) = L,LL7 kWhlyear
Figure 6 compares distributions of estimated UEC values for refrigerators and freezers.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 27 ot75
Table 20. 2012 Participant Mean Explanatory Variables*
Refrigerator
Age (years)28.40
Dummy: Manufactured Pre-1990 0.74
Size (ft.')18.16
Dummy: Single Door 0.02
Dummy: Side-by-Side 0.17
Dummy: Primary 0.38
lnteraction: Unconditioned Space x HDDs*8.51
lnteraction: Unconditioned Space x CDDs*0.40
Freezer
Age (years)35.79
Dummy: Manufactured Pre-1990 0.85
Size (ft.-)18.14
Dummy: Chest Freezer 0.24
lnteraction: Unconditioned Space x HDDs*11.84
lnteraction: Unconditioned Space x CDDs*0.52
22
Figure 6.2OL2 Distribution of Estimated Annual UECs by Appliance Type
L8%
L6%
p L4%
0)
E tzxooc
P LO%
E
E8%v
E.noc4%
2%
@6 oooooooooooooooooo rt ut |,.t ln tn u1 !n lJ1 l,) ln l,) l,l ln l,) U! r^(ir (n rf ur (o N € crr I = S P =
p 9
Unit Energy Consumption (kWh/year)
Table 21 presents estimated, per-unit, average annual energy consumption for refrigerators and
freezers recycled by Avista in 2OL2. The next sections describe how Cadmus adjusted these estimates to
arrive at gross per-unit saving estimates for participant refrigerators and freezers.
Table 22 presents the 2012 UEC results for Avista and compares it with utilities located in Canada and
the U.S. For 2OL2, Cadmus found Avista to have a slightly higher UEC for refrigerators and freezers than
other utilities.
to Relative Precision for Freezers was substantially higher than refrigerators due to a small sample size of 13
Exhibit No.3
Case Nos. AW-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 28 ot75
Table 21. Estimate of Per-Unit Annual Energy Consumption
23
Table 22. Benchmarking: Average UEC Values
Avista 2010-2011 Evaluation Report
Part-Use
"Part-use" serves as an adjustment factor, specific to appliance rerycling, used to convert the UEC into
average per-unit gross savings value. The UEC itself does not equal gross savings value, due to the
following:
o The UEC model yields an estimate of annual consumption.
o Not all recycled refrigerators would have operated year-round, had they not been
decommissioned through the program.
As Cadmus applied UMP's methodology, the determination of 2012 part-use differs slightly from that
used in previous evaluations. Specifically, the previous evaluation assumed that the way customers
operated participating appliances prior to the program served as a reasonable prory for how the same
appliances would likely be operated in the future, had they not been recycled through the program
(either by the participant or, if the appliance was transferred, by the would-be recipient).
While the UMP part-use methodology uses information from surveyed customers regarding pre-
program usage patterns, the final part-use estimate reflects the way appliances would likely be
operated, had they not been recycled (not how they were previously operated). For example, a primary
refrigerator operated year-round could become a secondary appliance and be operated part-time.
The updated methodology accounts for such potential shifts in usage types. Specifically, it calculates
part-use using a weighted average of the following, prospective part-use categories and factors:
o Appliances that would have run full-time (part-use = 1.0).
o Appliances that would not have run at all (part-use = 0.0).
o Appliances that would have operated for a portion of the year (part-use between 0.0 and 1.0).
Using information gathered through the participant survey, Cadmus utilized the following multistep
process to determine part-use, as outlined in the UMP:
L. The surveys determined if recycled refrigerators were primary or secondary units (with all stand-
alone freezers considered secondary units).
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-GI3
L.Hermanson, Avista
Schedule 3, Page 29ot75
24
2. For participants indicating they recycled a secondary refrigerator, the survey asked if the
refrigerator was unplugged, operated year-round, or operated for a portion of the preceding
year (and assuming all primary units operated year-round). Allfreezer participants were asked
the same question.
The survey asked participants indicating their secondary refrigerator or freezer operated for only a
portion of the preceding year to estimate how many months during that time their appliance was
plugged in. This subset of participants estimated 5.0 and 3.9 months for secondary refrigerators and
freezers, respectively. Dividing both values by 12 provided the annual part-use factor for all secondary
refrigerators and freezers operated for only a portion of the year. For 20L2, the average secondary
refrigerator and freezer operating part-time had part-use factors of 0.42 and 0.33, respectively. These
two steps determined how refrigerators and freezers operated prior to recycling, with results shown in
Table 23.
Table 23. Historical Part-Use Factors by Category
Secondary Units Only
Not in Use
Used Part Time
Used FullTime
Weighted Average
All Units (Primary
and Secondary)
Not in Use
Used Part Time
Used FullTime
Weighted Average
n=280% 0.00
LL% O.42
89o/o 1.00t00% 0.937s
n=49
0% 0.00
6% 0.42
94% 1.00
LOo% 0.96
5o;
1,199
t,124
500
1,199
L,L57
0%
38%
62%
LOO%
n=13
0.00
0.33
1.00
o.74
372
t,tl7
831
Table 24. Benchmarking: Part-Use Factors by Appliance Type
Avista 2010-2011 Evaluation Report
Avista 2012 Evaluation Report
25 Exhibit No.3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 30 of75
Cadmus then asked participants how the appliances likely would have been operated, had they not been
recycled through the program. For example, if surveyed participants indicated they would have kept a
primary refrigerator (independent of the program), the survey asked if they would have continued to
use the appliance as their primary refrigerator or would have relocated it and used as a secondary
refrigerator.
Participants indicating they would have discarded their appliance independent of the program were not
similar questions (as the future usage of their appliance would be determined by another customer).
Combining the historically based, part-use factors shown in Table 23 with participants' self-reported
action, had the program not been available, resulted in the distribution of likely future usage scenarios
and corresponding part-use estimates.
Table 25 shows the weighted average of these future scenarios, determining the program's part-use
factor for refrigerators (0.95 and freezers (0.74).11
Table 25. Part-Use Factors by Appliance Type
Net-to-Gross
Cadmus used the following formula to estimate net savings for recycled refrigerators:
Net savings = 6ross Savings - Freerid.ership and Secondary Market Impacts
- Induced Replacement
Where:
Gross Sovings = The evaluated in situ UEC for the recycled unit, adjusted for
part-use
Freeridership ond Secondory Market lmpods
As the future usage type of discarded refrigerators cannot be known, Cadmus applied the weighted part-use
average of all units (0.88) for all refrigerators that would have been discarded independent of the program.
This approach acknowledged that discarded appliances could be used as primary or secondary units in a
would-be recipient's home.
Exhibit No. 3
Case Nos. AW-E-13 AW-G13
L.Hermanson, Avista
Schedule 3, Page 31 of75
26
lnduced Replocement
= Program savings that would have occurred in the program's
absence
= Average, additional energy consumed by replacement units
purchased due to the program
Applying the UMP protocol introduced an additional parameter related to net savings-secondary
market impacts-and required use of a decision-tree approach to calculate and present net program
savings. Cadmus did not include this adjustment for the 201G-2011 impact evaluation; therefore,
changes in net savings could partially be attributed to changes in evaluation methodology,
The decision tree-populated by responses of surveyed participants-presented savings under all
possible scenarios concerning the participants actions in regard to the discarded equipment. Cadmus
used a weighted average of these scenarios to calculate net savings attributable to the program. This
chapter includes specific portions of the decision tree to highlight specific aspects of the net savings
analysis.
Freeridership
Cadmus' freeridership analysis first asked participants if they considered discarding the participating
appliance prior to learning about the program. lf the participant did not indicate a previous
consideration to dispose of the appliance, Cadmus categorized them as a non-freerider and excluded
them from the subsequent freeridership analysis.
Next, Cadmus asked all remaining participants (i.e., those who had considered discarding their existing
appliance before learning about the program) a series of questions to determine the distribution of
participating units likely to have been kept versus those discarded absent the program. Three scenarios
independent of program intervention could occur:
o The unit would be discarded and transferred to someone else.
o The unit would be discarded and destroyed.
o The unit would be kept in the home.
To determine the percentage of participants following each three scenario, Cadmus asked surveyed
participants about the likely fate of their recycled appliance, had it not been decommissioned through
the program. Cadmus categorized their responses into the following options:
o Kept the appliance.
o Sold the appliance to a private party (either an acquaintance or through a
posted advertisement).
o Sold or gave the appliance to a used appliance dealer.
o Gave the appliance to a private party, such as a friend or neighbor.
o Gave the appliance to a charity organization, such as Goodwill lndustries or a church.
o Left the appliance on the curb with a 'Tree" sign.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-13
L.Hermanson, Avieta
Schedule 3, Page 32ot75
27
. Had the appliance removed by the dealer who provided the new or replacement unit.
o Hauled the appliance to a landfill or recycling center.
o Had the appliance picked up by local waste management company.
Once Cadmus determined the final assessments of participants' actions independent of appliance
recycling program, the percentage of refrigerators and freezers that would have been kept or discarded
could be calculated, with the results shown in Table 25.
Table 26. Final Distributaon of Kept and Discarded Appliance
Kept No
Discarded |il.ilb["r*
Total
2s%
75%
LOO%
L7%
83%
I:O|J'% Total
Yes
No
78%
22%
too%
85%
L5%
100%
Table 27, Benchmarking Kept Appliances
Avista 201 0-201 1 Evaluation Report
*http:/Arvunr.powerauthority.on.calsites/defaulUfiles/new-files/2009/20090/o20ResidentiaP/AlGrealo/o20Refrigeratof/o20RoundupTl0Program
%20Evaluation.pdf (Ihe more recent 2010 evaluation cited previously relied on the NTG analysis from this 2009 evaluation).
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 33 of 75
28
Secondary Market lmpacts
lf a participant would have directly or indirectly (through a market actor) transferred the program-
recycled unit to another Avista customer, absent the program, Cadmus determined what actions the
would-be acquirer might have taken, with the unit unavailable due to the program.
Some would-be acquirers would find another unit; others would not. This possibility reflects some
acquirers being in the market for a refrigerator (and would acquire another unit), while others were not
(and would have taken the unit opportunistically). lt is difficult to quantify this absent program-specific
information, regarding changes in the total number of refrigerators and freezers (overall and for used
appliances) in use before and after implementing the program. Without this information, the UMP
recommends evaluators assume one-half of the would-be acquirers would obtain an alternate unit.
Without information to the contrary, Cadmus applied the UMP recommendation to this evaluation.
Next, Cadmus determined whether the alternate unit would likely be another used appliance (similar to
those recycled through the program) or a new, standard-efficiency unit (presuming fewer used
appliances remained available due to program activity).12
As discussed, estimating this distribution definitively proves difficult. The UMP again recommends taking
a midpoint approach when primary research is unavailable: evaluators should assume one-half of the
would-be acquirers would obtain a similar used appliance, and one-half would acquire a new, standard-
efficiency unit.
Cadmus used the ENERGY STAR Website'3 to determine energy consumption for new, standard-
efficiency appliances. Specifica!!y, Cadmus averaged the reported energy consumption of new, standard-
efficiency appliances of comparable sizes and configurations as the program units.
Figure 7 details Cadmus' methodology for assessing the program's impact on the secondary refrigerator
market and for applying the recommended midpoint assumptions when primary data were unavailable.
As shown, accounting for market effects resulted in three savings scenarios:
o Full per-unit gross savings;
o No savings; and
o Partial savings (i.e., the difference between energy consumption of the program unit and the
new, standard-efficiency appliance acqu ired instead).
The would-be acquirer also could select a new ENERGY STAR unit. However, Cadmus assumed most customers
in the market for a used appliance would upgrade to the next lowest price point (a standard-efficiency unit).
htto://www.energvsta r.qov/index.cfm?fuseaction=refris.calculator
Exhibit No. 3
Case Nos. AW-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 34 ot75
29
Figure 7. Secondary Market lmpacts-Refrigerators
i' "r. --.i-i-- "r -j-1._
t-ffi-y5'Exsy,nc-ulgJ trMMsnxo-uEJ - a-D
C;s"**"] -[,^*.,*=fr-^';=fC
[ *o-,oXi,*o-*" I - [---;------) = @
Integration of Freeridership and Secondary Market lmpacts
After estimating the parameters of the freeridership and secondary market impacts, Cadmus used the
UMP decision tree to calculate the average, per-unit program savings, net of their combined effect.
Figure 8 shows how Cadmus integrated these values into an estimate of savings, net of freeridership and
secondary market impacts. Again, Cadmus applied secondary market impacts to maintain consistency
with UMP: previous Cadmus Avista appliance recycling evaluations did not account for this.
To ensure survey participants provided the most reliable responses possible, and to mitigate socially
desirable response bias as much as possible, Cadmus averaged the participant and nonparticipant
transfer and disposal ratios.
Figure 8. Savings Net of Freeridership and Secondary Market lmpacts-Refrigerators
BDaa rllrlrmEqtlrmurl1ilaram
tts*ru lnlcilrtE,rmtt ,tt5t?rr*!n:rclrrLrr'llmr
m mmrlncm ffimIM MFg6mrxffirm
moiltrrnlnm msl.[*-*:#"J-G;til;=@
'E=CO
(-----r----__]-f . -')=CI
[ ,. t-r--:_i_]=COI ffisBffiG I L
xEt-Fn-{-rilhrsehg.n tdltt.d.tclrtipxt.ond..ymr*drrrg.d @
!nduced Replacement
The UMP states that evaluators must account for the energy consumption of replacement units only
when the program induced that replacement (i.e., when the participant would not have purchased the
replacement refrigerator in the recycling program's absence).
ln the case of non-induced replacements, the energy consumption of the replacement appliance does
not prove germane to the savings analysis as the appliance would have been purchased or acquired
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 35 of 75
30
regardless of the program. The acquisition of another appliance in conjunction with participation in the
program does not necessarily indicate induced replacement. Again, this is consistent with the methods
outlined in the UMP.
Cadmus used the results of the participant surveys to determine which replacement refrigerators and
freezers program participants acquired due to the program. Survey results indicated the program:
reduced the total number of used appliances operating within Avista's ldaho service territory; and raised
the average efficiency of the active appliance stock.
Cadmus then used participant survey results to estimate the proportion of replacements induced by the
custome/s participation in the program. Specifically, Cadmus asked each participant that indicated they
replaced the participating appliance: "Would you hove purchosed the new refrigerator/freezer without
the incentive you received for recycling the old one?"
As a S30 incentive likely will not provide sufficient motivation for most participants to purchase an
othenrise unplanned for replacement unit (which can cost SSOO to $2,000), Cadmus asked a follow-up
question of participants who responded "No." lntended to confirm the participant's assertion that only
the program caused them to replace their appliance, the question was: iust to confirm: you would not
hove reploced your old refrigerotor/freezer without the Avisto incentive for recycling, is thot correct?"
To further increase the reliability of these self-reported actions, the induced replacement analysis
also considered:
t. Whether the refrigerator was a primary unit.
2. The participant's stated intentions in the program's absence.
For example, if participants would have discarded their primary refrigerators independent of the
program, the replacement could not be program induced (since it is extremely unlikely a participant
would live without a primary refrigerator). For all other usage types and stated intention combinations,
however, induced replacement could stand as a viable response.
As expected, results indicated the program only induced a portion of the total replacements: the
program induced O% of all refrigerator participants and 0% of freezer participants to acquire a
replacement unit, as shown in Table 28. As shown in Table 29, Avista's induced replacement was lower
than the comparison utilities.
Table 28. 2OLL-2OL2 lnduced Replacement Rates
Exhibit No.3
Case Nos. AW-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 36 of75
37
Table 29. Benchmarking: lnduced Replacement
Figure 9. lnduced Replacement Refrigerators
mclontrcffiuxl mtgrclott tlcDr ffi
CM trpYSOa{rrHmruttAr
i.otffi usE sr^No to uEc
tl.tot oilar-rExrm,, rloatlr mG[n-r---r----t=@
f;;.=x*-;l-(;=r*"*"l=@
lItruCED_ItIUh: lnduccd Consumflim
FinalNTG
As summarized in Table 30, Cadmus determined final net savings as gross savings less freeridership,
secondary market impacts, and induced replacement kWh.
As noted, the application of the UMP protocol introduced two parameters related to net savings-
secondary market impacts and induced replacements-not included in the previous evaluation. The
application of these factors, through adherence with the UMP, contributed to a downward shift in the
program's NTG from previous years.
Table 30.2012 NTG Ratios
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 37 ol 75
32
Summary of tmpac Findings
Using the above per-unit values, Cadmus calculated total program savings for the Second Refrigerator
and Freezer Recycling program in ldaho at 154,811 kWh per year, after adjustments (as shown in
Table 31).
Table 31. tdaho 2012 Annual Second Refrigerator and Freezer Recycling Program Savings
Refrigerator Recycling
Frgezel fgcyg!!ng
Totals
257 2t3,O50 292,8L8,L35,79L',
As shown in Table 32, Avista's NTG gross ratio is less than other utilities. The NTG gross results from
2012 were driven downward primarily by the ratio of appliances that would have been discarded in
absence of the program as well as the mature nature of the program relative to other programs.
L,3,4 ENERGY STAR Products
Program Description
The ENERGY STAR Products program includes the following measures:
o Clothes Washer (Electric and Gas)
o Dishwasher (with Electric or Gas Water Heater)
o Freezer (Electric)
o Refrigerator(Electric)
The program offers direct financial incentives to motivate customers to use more energy-efficient
appliances. The program indirectly encourages market transformation by increasing demand for ENERGY
STAR products. The program includes electric and gas measures, but this report only considers electric
savings.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 38 of 75
Table 32 Benchmarking NTG Ratio's
Avista 2010-201 1 Evaluation Report
33
Analysis
Energy savings credited to the ENERGY STAR Products program had to meet the following criteria:
o Measures had to remain in place and operate properly at the time of verification;
o Numbers of installed equipment pieces and their corresponding model numbers in the
applications had to match the database; and
o Units must have been ENERGY STAR-qualified at the time of the program offering.
Clothes Washers
Energy-saving calculations drew upon a 2009 Cadmus study,la which metered more than 100 clothes
washers in California homes for three weeks-the largest rn sftu metering study on residential clothes
washers and dryers conducted in the last decade. Cadmus updated the analysis for this evaluation to
improve the accuracy of the savings estimated.
Dryers produced the majority of energy consumption and savings, as high-efficiency washing machines
removed more moisture from clothes, allowing shorter drying times.
Determining adjusted gross savings required using the following, additional input assumptions:
. Recent independent evaluation surveys from the RBSAI5 and 2Ot2 Avista Participant surveys
estimated 262 washing cycles per year. UES values have been adjusted accordingly, as reflected
in this measure's realization rate.
o Cadmus utilized the data from the California metering study to estimate consumption per wash
and dry cycle for the base and efficient equipment.
Dishwasherc
Cadmus estimated dishwasher savings based on methods currently used in the ENERGY STAR
Calculatorl6 (the onty calculator available providing consistent energy-savings estimates in the presence
of a gas or electric domestic hot water heater). The utilized the following input assumptions:
o Cadmus calculated the average base case and efficient case Energy Factor (EF), with both based
on data utilized by the RTF. The baseline EF equaled the average market efficiency of units not
qualifying for the program, and the efficient EF equaled the average market efficiency of units
qualifying for the program at the time of their rebate.
The Cadmus Group, lnc. 2010. "Do the Savings Come Out in the Wash? A Large Scale Study of ln-Situ
Residential Laundry Systems."
htto://www.cadmusgrouo.com/odfs/Do the Savinss Come Out in the Wash.pdf
Ecotope lnc. 2077 Residentiol Building StockAssessment: Single-Fomily Choroaeristics ond Energy Use. Seattle,
WA: Northwest Energy Efficiency Alliance. 2012.
http://www.energystar.gov/ialbusiness/bulk_purchasing/bpsavings_calc/
Calcu latorConsu merDishwasher.xls?7 182-1c92
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 39 of 75
34
Recent evaluation surveys conducted in the region estimated 245 washing cycles per year.tz tt
Water heating consumed 55% of the electricity required to run a dishwasher connected to an
electric domestic hot water heater.le
Refrigerators
Cadmus used the methodology shown in the RTF's FY11v2_1 refrigerator analysis to estimate gross per-
UES. The RTF's analysis assumed 32% of baseline units would be ENERGY STAR-qualified. This
assumption embedded NTG in the calculated savings. Cadmus modified the analysis to assume 0% of
baseline units would be ENERGY STAR-qualified. The resulting savings equaled the gross savings
achieved by the installation of an ENERGY STAR refrigerator. Chapter 5 addresses net savings.
Freezerc
Cadmus used the methodology shown in the RTF's FY10v2_0 freezer analysis to estimate gross per-UES.
The RTF's analysis assumed LO% of baseline units would be ENERGY STAR-qualified. This assumption
embedded NTG in the savings calculated. Cadmus modified the analysis to assume 0% of baseline units
would be ENERGY STAR-qualified. The resulting savings equaled the gross savings achieved by the
installation of an ENERGY STAR freezer. Chapter 5 addresses net savings.
Results and Findings
Table 33 shows: total reported and qualified counts, savings, and realization rates of electric ENERGY
STAR Products measures in ldaho.
100.0% L20,384 39.2%- '- r l_!,911, _ _199.0% l_ 4,g3L _ _1oO.O% 1oO.O% _4,_9_31 _ _ __l!qq%
!6,8!6 I _ __1. !q4E_ 100.q% - s5,805 100.0%
a
a
LL,842
193,963
100.0% 100.0% LL,842 100.0%,lI 4910tr i !09'9?6, r9!9qL !q%
L7
t8
19
Pacific Power. Woshington 2009-2070 Residentiol Home Energy Sovings Evoluation January 2012.
Rocky Mountain Power. 2N9-2070 ldoho Residentiol Home Energy Sovings Evoluation. February 2012.
htto://www.enerevstar.sov/ialbusiness/bulk ourchasins/bpsavinss calc/CalculatorConsumerDishwasher
.xls?7L82-tc92
Exhibit No. 3
Case Nos. AW-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 40 ot75
Table 33. ENERGY STAR Products Program Results
35
1.3.5 Heating and Cooling Efficiency
Program Description
The electric Heating and Cooling Efficiency program included the following equipment:
o Ductless Heat Pumps
o Air Source Heat Pumps
o Electric Forced Air Furnace to Air Source Heat Pumps
o Variable Speed Furnace Fans
o Air Conditioner Replacements
Analysis
The PY 201f2011 electric impact evaluation report documented analysis Cadmus performed to
determine the change in energy consumption resulting from installation of electric heating and cooling
measures. As the analysis continued to provide the best information on this measure, results were
retained for the 2OL2 program year.'o
Besults ond Findings
Table 34 shows: totaltracked and qualified counts, savings, and realization rates for electric Heating and
Cooling Efficiency measunes in ldaho.
The program achieved a99.2% realized adjusted gross savings rate, reduced slightly due to qualification.
20 Cadmus. Avista 2070-2077 Mutti-Sector Electric lmpoct Evaluotion Repoft. May 2012.
Table !14. Heating and Cooling Efficiency Program Results
E Air Source
Heat Pump 2U 58,550 68,550 99.2%LOo%58,101 99.20%
E Ductless
Heat Pump 34 6,277 6,277 99.2%LOo%6,227 99.20%
E Electric To
Air Source
Heat Pump
60 395,359 395,359 99.2%100%392,196 99.20%
E Variable
Speed
Motor
47t 206,557 206,557 99.2%100%204,905 99.20%
Program
Total 769 6798/,'3 6768,,3 99.2%1(xr%671,428 99.2M
Exhibit No.3
Case Nos. AW-E-13 AW-GI3
L.Hermanson, Avista
Scfpdule 3, Page 41 of 75
36
1.3.5 Space and Water Conversions
Program Description
The Space and Water Conversions program incents two measures available to residential electric
customers currently using electricity to heat their homes and water, but may be able to use natural gas
instead:
o Electric Forced Air Furnace to Natural Gas Forced Air Furnace
o Electric Water Heater to Gas Water Heater
Avista customers receive a rebate to reduce the cost of purchasing new equipment when making a
conversion. These measures may be claimed in addition to the heating and cooling efficiency measures
previously described. The installed, efficient equipment case therefore is assumed to be the standard
efficiency equipment assumed for the base case equipment in the measures discussed.
Anolysis
AllMeasures
The PY }OLO-IOLL electric impact evaluation report documented analysis Cadmus performed to
determine the change in energy consumption resulting from conversion of electric air or water heating
to gas air or water heating. As the analysis continued to provide the best information on this measure,
results were retained for the 2012 program year." For Q1 2014, a billing analysis is slated to address
2012 participants.
Results and Findings
Table 35 shows total tracked and qualified counts, savings, and realization rates for electric Space and
Water Conversion measures in ldaho.
E Electric To
Natural Gas
Furnace
24 288,298
I e Etectricro
Natural Gas
Water 56,435 99.2%,,100.0%
2! Avista. 2O\U2O77 Multi-Sector Etectric tmpdct Evoluation Report. May 2OL2.
Table 35. Space and Water Convercion Measures and Reported and Adjusted Savings
Exhibit No.3
Case Nos. AW-E-13 AVU-G13
L.Hermanson, Avista
Schedule 3, Page 42 of75
37
The program achieved a99.2% realized adjusted gross savings rate, reduced slightly due to
qualifications.
L.3.7 Residential Weatherization
Program Description
The Residential Weatherization program incented four categories of measures available to residential
electric and gas customers heating their homes with fuel provided by Avista:
o Fireplace Dampers (Electric and/or Gas Savings)
o lnsulation-Ceiling/Attic (Electric andlorGas Savings)
o lnsulation-Floor (Electric and/or Gas Savings)
o lnsulation-Wall (Electric andlor Gas Savings)
Avista customers primarily heating with electric or natural gas and having a wood burning fireplace
could receive up to S1OO for installing a rooftop damper. This measure was removed for the 2012
program year. The one participant is a legacy from the previous program year.
The program incented qualifying ceiling and attic insulation (both fitted/batt and blown-in), which
increased the R-value by 10 or more, at S0.2S per square foot of new insulation, and up to 50% of
installation costs. Homes qualified if they had existing attic insulation less than R-19.
The program incented floor and wall insulation (both fitted/batt and blown-in), which increased the
R-value by 10 or more, at $0.50 per square foot of new insulation, up to 50% of the installation cost.
Homes qualified if they had existing floor and/or wall insulation less than R-5.
Analysis
The PY 2OL0-I20LL electric impact evaluation report documented a census billing analysis Cadmus
performed to determine the change in energy consumption resulting from installation of weatherization
measures. As the billing analysis continued to provide the best information on this measure, results
were maintained for the 2012 program year."
The billing analysis did not include Fireplace Dampers, retaining the deemed savings value developed for
the 2011 Avista TRM.
Table 35 shows total reported and qualified counts, savings, and realization rates of gas weatherization
program measures.
22 Avista. 2070-2077 Multi-Sector Electric tmpact Evoluation Report. May 2012.
Exhibit No.3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 43 ot75
38
Table 36. Weatherization Program Results
X E Fireplace
Damper With
Electric Heat
E lnsulation
Program Total
E Electric
Water Heater
Program Total
48
49
163
37,5L2
37,675
163
37,512
37,675
99.2%
99.2%
99.2%
L00.0%
L00.0o/o
100.0%
L62
37,272
37,373
99.2%
99.2%
99.2%
1.3.8 Water Heater Efficiency
Progrom Description
The Water Heater Efficiency program represented one measure:
o High-Efficiency Water Heater (Electric)
Through this program, Avista offered a S50 incentive to residential electric customers installing an
eligible high-efficiency water heater. Electric water heaters with a tank had to have a 0.93 EF or greater
to qualify for the program.
Anolysis
The PY 201f2011 electric impact evaluation report documented analysis Cadmus performed to
determine the change in energy consumption resulting from installation of this measure. As the analysis
continued to provide the best information on this measure, results were retained for the 2Ot2 program
year.'3
Results and Findings
Table 37 shows total tracked and qualified counts, savings, and realization rates for the electric Water
Heater Efficiency progra m measu re.
8,933
8,933
8,933
8,933
99.2o/o
99.2%
LOO.O%
100.0%
8,851
8,851
99.2o/o
99.2%
23 Cadmus. Avisto 207f2077 Mutti-sector Electric lmpoct Evoluotion Report. May 2012.
Table 37. Water Heater Efficiency Measure and Reported and Adjusted Savings
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 44 of 75
39
1.3.9 ENERGY STAR Homes
Program Description
This program offered incentives to builders constructing single-family or multifamily homes complying
with ENERGY STAR criteria and certified as ENERGY STAR Homes. Avista provided a $900 incentive for
homes using its electric or electric and natural gas service for space and water heating.
Anolysis
The PY 2011electric impact evaluation report documented the simulation modeling Cadmus performed
to determine energy savings achieved by these measures. As the simulation resuhs continued to provide
accurate estimates of savings, results were maintained for the 2OL2 program year.2n
Results ond Findings
Table 38 shows totaltracked and adjusted counts, savings, and realization rates for measures within
ENERGY STAR Homes. The electric and gas programs funded participating homes using both Avista
electric and gas.
1.4 Conclusions
For PY 2012, Avista's ldaho residential electric programs produced 5,543,748 kWh in gross savings,
yielding an overall realization rate of 109.3%. Table 39 shows reported and evaluated gross savings as
well as realization rates per program.
24 Cadmus. Avistd 2077 Multi-Sector Gas tmpoct Evaluation Report. May 2OL2,
Table 38. ENERGY STAR Home Program Results
Itm.o% | 100.0%
2,108 I 100.0%
Exhibit No.3
Case Nos. AW-E-13 AW-G13
L.Hermanson, Avista
S$edule 3, Page 45 of 75
40
Table 39. Total Program Reported and Evaluated Gross Savings and Realization Rates
, 3,330,478' 3,330,478lSavings- j:-::__'-- "-: 't__ + : : l -
1
i :t::li ff iserator and tt ,ur,rrr. i eso,gsa NA 1oo.o% i sso,gss L30.6% )\
ilr"elsr-Bgel.lllg +--'-""" | ' ''^-r -- -'-:-:^ I "-::--- '. :':-":*'1
frue.leiiieRTiducts i - sso,asz ; lsa,s_qa - roo.ox , roobx ;_ rslssa sopr r
i Heating and cooling I i i ,oo.ox i 67L,428 gg.zN ii#f,reil -vvvrr,D t, 675,843 | 676,843 ssz% l Lp*tl9l4.!9ryslgl :' _ Et,stsi 37,67s _- ,r2%_ 1oo.o% t, 37,?79 ss.z%-i,---=_- ..-_---Lwe!er!e3!el!!e!@
i enrRcvstaR xomes , _u,698_;_2!!69L) _ _ _- 100.0%
ispaceandWater I -..--., ^..,^, ^^^^,
l-99.2% 1oO.O% 8,851 99.2% ,:l-::- T-
-:--=:-:::--
... ---100.0% 24,698 100.0% :
^^.^,;i JPoLEorru YYoLtr ' 344,734 | 9M,734 gg.2% 1oo.o% i 34L,gll gg.2%
ii Conversions r ii uonversrons I ,
L_reIrL_ _ _ _l_f,lZl,gqslf,geatz'L_!t3%f - r@.gf-i_1q!9,i19 i rog€f:
1.5 Recommendations
Cadmus recommends the following changes to Avista's residential electric programs:
o Consider updating per-unit assumptions of recycled equipment to reflect this evaluation,
ensuring planning estimates of program savings more closely match evaluated savings.
o Move all clothes washer rebates to the electric program unless gas dryers achieve substantial
penetration. Forthcoming RBSA data can support future analysis.
o lnclude a SEER requirement to increase savings for high-efficiency heat pump participation.
Consider continuing the Variable Speed Motor measure in conjunction with any change to
equipment efficiency requirements. Often, the highest efficiency heat pump systems use an
electrically commutated motor (ECM) as standard equipment.
o Consider restricting dual-fuel customers who acquire multiple rebates and have interactive
effects. lf program changes reduce the participation of dual-fuel customers in certain measure
categories, future evaluation activities should reassess the participant penetration of the dual-
fuelhome.
r lncrease measure-level detailcaptures on applications and include in the database. Specific
additional information should include: energy factors or mode! numbers for appliances; baseline
information for insulation; and home square footage, particularly for the ENERGY STAR Homes
program.
o Consider estimating savings and incenting systems separately for all-electric heating systems.
o Consider tiered incentives by SEER rating, as higher SEER systems generally require ECM fan
motors to achieve certain SEER ratings.
Exhibit No.3
Case Nos. AW-E-13 AVU-G-13
L.Hermanson, Avista
Schedule3, Page46of75
41
1.5.1 Future Research Areas
Cadmus recommends the following future research areas:
e Review all available secondary research andlor collect primary data on the penetration of gas-
heated clothes dryers within Avista's gas territory. This information can be used to refine the
estimated gas and electric savings associated with the purchase of an ENERGY STAR clothes
washer in a home with a gas domestic hot water tank.
o Perform a targeted billing analysis on weatherization participants using both electricity and gas
to heat their homes.
o Perform a billing analysis on ENERGY STAR homes using a nonparticipant comparison group,
once sufficient homes have participated under the new requirements.
r ldentify new, cost-effective measures to be added to the portfolio.zs
E At the time of this report, Cadmus was aiding Avista in identifying new programs and measures.
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3,Page47 ot75
42
2. NONRESIDENTIAI ETECTRIC IMPACT REPORT
2.1 Introduction
Avista's nonresidential portfolio of programs promotes commercial utility customers' purchases of high-
efficiency equipment. Avista provides rebates to partially offset the difference in cost between high-
efficienry and standard equipment.
The nonresidential electric portfolio offers 16 programs in three major categories: Prescriptive, Energy
Smart Grocer, and Site-Specific (Custom). Descriptions of the programs follow.
2,L,1 Prescriptive
Prescri ptive Commercial Clothes Wosher (PCW)
To encourage customers to select high-efficiency clothes washers, this program targets nonresidential
electric and naturalgas customers in multifamily or commercial Laundromat facilities. The program's
streamlined prescriptive approach seeks to reach customers quickly and effectively to promote ENERGY
STAR or Consortium for Energy Efficiency (CEE) listed units.
Prescriptive Commerciol Windows and lnsulation (PCS)
Beginning in January 2011, the installation of commercial insulation has been processed through a
prescriptive program, in addition to the site-specific program. Projects eligible for the prescriptive
commercial shell program have preexisting:
o Wall insulation levels of less than R4, improved to Rl1 or better.
o Attic insulation of less than R11, improved to R30 or better.
o Roof insulation of less than R11, improved to R30 or better.
Prescriptive Food Seruice (PFS)
Applicable to nonresidential electric and gas customers with commercial kitchens, this program provides
direct incentives to customers choosing high-efficiency kitchen equipment. The equipment must meet
ENERGY STAR or CEE tier levels (depending on the unit) to qualify for an incentive.
Prescriptive Green Motors lnitiative (PGM)
Operated in partnership with the Green Motors Practices Broup, this program provides education to
foster organization and promotion of member motor service centers' commitments to energy-saving
shop rewind practices for motors ranging from 15 to 500 HP.
Prescriptive Lighting (PL)
Since a significant opportunity exists for lighting improvements in commercialfacilities, this program
offers direct financial incentives to customers that increase the efficiency of their lighting equipment.
Existing commercial and industrial electric customers qualify if their facilities have rate schedules 11 or
above. This program provides pre-determined incentive amounts for 38 measures, including:
Exhibit No.3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 48 of 75
43
. Tt2 fluorescents to T8 fluorescents.
o High bay, high-intensity discharge lighting to T5 fluorescents or T8 fluorescents.
o High bay, high-intensity discharge lighting to induction fluorescents.
o lncandescents to compact fluorescents or cold cathode fluorescents.
o lncandescents to LEDs.
o lncandescent exit signs to LED exit signs.
Prescriptive HVAC Vorioble Frequency Drive (PHV)
The use of single-speed motors to drive fans or pumps often allows energy savings through use of a
variable frequency drive (VFD). The VFD can convert a single-speed motor to variable speed without
modification to the motor itself. This can be an efficient way to convert, for example, constant volume
air systems into variable volumes. VFDs are readily available for motors from t hp to 300 hp, and can be
easily installed directly into the power !ine leading to the motor, replacing the existing motor starter.
VFDs can earn installation incentives through Avista.
Prescriptive PC Network Controls (PNC)
Computers remaining in a full-power state when idle can waste significant energy for customers with
numerous PCs. This program, available to nonresidential electric customers, provides an incentive to
install a network-based power management software solution to controlthe power of networked PG.
Prescriptive Stondby Generator Block Heater (PSG)
Most block heating technology employs natural convection within the engine block's system to drive
circulation-more commonly known as thermosiphon. This program promotes the replacement of
thermosiphon style engine block heaters with pump driven circulation units, which reduces overall block
temperature. Because it also decreases the heat transfer rate from the block to the environment, it can
reduce overall block heater energy consumption, which is tied to the circulation method.
Renewobles (REN)
This program provides prescriptive incentives for residential and nonresidential projects that install
photovoltaic (solar electric) systems and/or wind turbines.
2.1.2 Energy Smart Grocer (ESG)
Though refrigeration offers a high potential for energy savings, it is often overlooked due to the
technical aspects of the equipment. The Energy Smart Grocer program assists customers with technical
aspects of their refrigeration systems while providing a clear view of the savings they can achieve. A field
energy analyst offers customers technical assistance, produces a detailed report of the potential energy
savings at their facility, and guides customers through the ESG process, from inception to payment of
incentives for qualifying equipment.
2.L.3 Site Specific (SS)
The site-specific program provides nonresidential measures that do not fit under the prescriptive
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 49 ol 75
applications and thus must be considered based on their project-specific information. For a measure to
be considered, it must have demonstrable kWh and/or therm savings. All commercial, industrial, or
pumping customers that receive electric or natural gas service from Avista may qualify for these
measures. The program includes the following electric and gas saving measures:
o Site Specific HVAC (SSHVAC)
r HVAC Combined
! HVAC Cooling
r HVAC Heating
I Motor Controls HVAC
. Multifamily
o Site-specific Lighting (SSL)
' Lighting Exterior
r Lighting lnterior
o Site-specific Other (SSO)
. Appliances
! Compressed Air
r lndustrial Process
t Motors
. Motor Controls lndustrial
o Site-specific Shell (SSS)
Avista implements the site-specific and prescriptive programs, and PECI implements the ESG program.
Both Avista and PECI design and manage program details, and have developed algorithms for calculating
measure savings and determining measure and customer eligibility.
Avista staff fields inquiries from potential participants and contractors, and maintains a tracking
database for projects. Throughout the program, Avista manages projects by: reviewing and approving
applications at all stages of the process; calculating project savings; and populating the database with
relevant information.
2.2 Methodology
Cadmus designed the impact evaluation to verify tracked program participation and to estimate energy
savings. The evaluation determined gross savings using the following: engineering calculations, desk
reviews, verification site visits, and some project-level billing analysis.
For a sample of sites, Cadmus reviewed Avista's tracked gross energy savings and available
documentation, such as audit reports and savings calculation work papers, particularly focusing on
calculation procedures and documentation for savings estimates. The review also verified the
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 50 of 75
45
appropriateness of Avista's analyses for calculating savings, and the analyses' operating and structural
parameters. Through site visits or desk reviews for a sample of projects, Cadmus also collected data and
evaluated gross energy savings through engineering calculations.
Cadmus collected baseline, tracking and program implementation data through on-site interviews with
facility staff. The visits included verifying measure installations and determining changes to the
operating parameters following measure installation. Facility staff interviews included questions
regarding the installed systems' operating conditions, additional benefits, or shortcomings. Using the
savings realization rates from the sample sites, savings could be estimated and recommendations
developed for future studies.
2,2.t Sampling
Table 40 presents the rigor levels of precision targets for ldaho and Washington, combined. Cadmus
developed a sampling calculation tool to estimate the number of on-site visits required to achieve these
levels, using preliminary program population data provided by Avista. Meeting the levels required
metering 52 projects and verifying 66 projects aross the combined PY 2OL2 and 2013 program
populations. By meeting targets for each stratum, the evaluation will achieve 90/10 precision at the
overall nonresidential program level. Calculated following the PY 2013 evaluation, the final precision will
be based on the combined program populations for both years.
Prescriptive
ESG
SSHVAC
SSLsso ,
sss
Total
90l20
eo/2o
23
5
8
10
5
0
52
L4
7
18
10
10i
7
65
90120
90/20
e0120
e9[v:9
9,0/20 .,
Cadmus selected both a census and random sample for each stratum. The census projects represented a
small number of participants with large savings impacts for the stratum. Table 41 presents the cutoff for
the census savings for each stratum. We visited all sites with reported savings above this census level.
From the remaining population of projects, the study also randomly selected additional participants in
each stratum. Subsequent sections of this report willexplain the differences between the initially
proposed and the actual sampling plans for evaluation activities. Table 42 and Table 43 show final
samples for 2Ot2 projects. Cadmus will evaluate the remaining portions of the proposed sample shown
in Table 40 for 2013 projects. Sample sizes will be modified, as appropriate based on final population
sizes, to meet the expected confidence and precision levels.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 51 of75
Table tt0. Proposed PY 2Ot2-2O13 Nonresidential Evaluation Activities
45
Table 41. Census Leve! Cutoff by Stratum
Prescriptive
ESG
SSHVAC
SSL
sso
sss
1,000,000
no census level cutoff
500,000
1,000,000
1,000,000
no census level cutoff
Table 42. Final PY 2012 Electric Evaluation Activity Sample-Washington and ldaho Combined
Prescriptive
ESG
SSHVAC
ssL
sso
SSS
Total
9
0
0
5
4
0
18
1
4
5
1
1
2
t4
3
0
0
0
1
0
4
5
7
L2
5
5
5
41
Prescriptive
ESG
SSHVAC
SSL
sso
SSS
Total
The database extract provided information at the program-level but not at the measure level (e.g.,
chillers, anti-sweat heater controls, LED lighting fixtures). Therefore, the study sought to verify savings
for every incented measure at each site, regardless of whether it achieved gas or electric savings.
Cadmus could not, however, determine whether the study evaluated an accurate distribution of specific
measure types within each program. Establishing this distribution would have required an exhaustive
review of project files, which fell outside of the evaluation's scope.
2.2.2 Data Collection
Cadmus collected data from four metered projects and 14 on-site verifications in ldaho for PY 2OL2
(though the full sample with both states was used for extrapolation). The process began with a
document review to determine measure types, quantities, operational parameters, and the calculation
methodology.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 52 ot 75
Table 43. Fina! PY 2012 Electric Evaluation Activity Sample-ldaho Only
47
Document Review
Avista provided Cadmus with documentation on the sample sites' energy-efficiency projects, including:
o Program forms;
o The tracking database;
o Audit reports; and
o Savings calculation work papers for each rebated measure.
The review of calculation spreadsheets and energy simulation models emphasized calculation
procedures and documentation for savings estimates.
Cadmus reviewed each application for the following information:
o Equipment replaced: descriptions, schematics, performance data, and other
supporting information.
o New equipment installed: descriptions, schematics, performance data, and other
su pporting information.
o Savings calculation methodology: the methodology type used, specifications of assumptions,
sources for these specifications, and correctness of calculations.
Short-Term Metering
Cadmus performed short-term metering, lasting two to four weeks, for lighting projects in the
Prescriptive Lighting and Site-Specific Lighting programs. This involved installing light loggers to estimate
annual operating hours for each lighting measure. Cadmus developed a light logger plan to capture
representative lighting operations for each site, basing the number and location of loggers for the site
on the number of space types and the magnitude of savings by space and fixture-type.
The effort also installed power meters on a chiller retrofit that was part of an industrial process energy
savings project. Meters recorded power data over a period of one month to characterize retrofit
performance and power consumption.
Site Visits
On-site visits completed the following primary tasks:
t. Verifying the implementation status of all measures for which customers received incentives.
This required verifying the energy-efficiency measures had been installed correctly and
functioned properly. lt also included verifying the operational characteristics of installed
equipment, such as temperature set points and operating hours.
2. Collecting physical data, such as boiler capacities or operationaltemperatures, and analyzing the
energy savings realized from installed improvements and measures.
3. Conducting interviews with facility personnel to obtain additional information regarding the
installed system, thus supplementing data from other sources.
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 53 of 75
48
2.2.3 Engineering Analysis
Prescriptive and site-specific programs required significantly different methods of analysis.
Overuiew
Procedures used for verifoing savings through an engineering analysis depended on the type of measure
analyzed. A list below presents analytical methods used in this evaluation, with descriptions in the
following sections:
o Prescriptive deemed savings
o Short-term metering
o Calculationspreadsheets
. Energy simulation modeling
Prescriptive Deemed Savings
For most prescriptive measures, Cadmus verified the deemed savings estimates that Avista used for
savings calculations. Verification activities focused on:
o The installed quantity;
o Equipment nameplate data;
o Proper equipment installation; and
o Operating hours.
Where appropriate, Cadmus used data from site verification visits to reanalyze prescriptive measure
savings using Avista's Microsoft Excel calculation tools, ENERGY STAR calculation tools, RTF-deemed
savings, and other secondary sources.
Short-Term Metering
Depending on the site and measure, Cadmus determined short-term light logging over a period of two
to four week presented the most effective method for achieving precision on four lighting projects'
energy-saving ca lcu lations.
Calculotion Spreodsheets
Avista developed calculation spreadsheets to analyze energy savings for a variety of measures, including
building envelope measures, such as ceiling and wall insulation. Calculation spreadsheets required input
of relevant parameters (e.g., square footage, efficiency values, HVAC system details, and location
details). From these data, energy savings could be estimated using algorithms programmed by Avista.
Cadmus reviewed input requirements and output estimates for each spreadsheet and determined if the
approach proved reasonable.
Energy Simulation Modeling
Avista determined savings for several site-specific HVAC and shell projects with energy simulation
modeling (using eQuest software). Avista chose this method due to the complex interactions between
Exhibit No.3
Case Nos. AW-E-13 AW-G,I3
L.Hermanson, Avista
Schedule3, Page 54ot75
heating and cooling loads and building envelopes. lmplementation staff provided the original energy
simulation models, and Cadmus reviewed the models to determine relevant parameters and operating
details (such as temperature set points) for the applicable measure, and then updated the models as
necessary, based on on-site verification data.
2.3 Results and Findings
2.3.1 Overview
Cadmus adjusted gross savings estimates based on evaluated findings. The following sections discuss
further details by program. The ldaho evaluation sample included 18 projects, divided into the following
program subsectors:
o Prescriptive: four projects
o Energy Smart Grocer: four projects
o Site Specific: 10 projects
2.3.2 Prescriptive
Cadmus evaluated savings for a sample of sites across nine prescriptive programs for the combined
ldaho and Washington sample. The ldaho sample only included projects from two programs-Lighting
and PCN. Table tt4 shows evaluated results by program. Further evaluation details for each
program follow.
Table tt4. Evaluated Results for PY12 Nonresidential Electric Prescriptive Sample-ldaho
1,637,O27
Table 45 shows the combined ldaho and Washington prescriptive resuhs, which the study used for final
extrapolation (as the sample derived from a combined sampling methodology).
Table 45. Evaluated Results for PY12 Nonresidential Electric Prescriptive Sample-
00%
1,849
qL
W"
95%
,_ 1,84! I 100%
1
- _ t,Ett5,74g_l* gsx
]
Washington and ldaho
50
1,939,067_
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 55 of75
Cadmus identified severaldiscrepancies between the evaluated results and Avista's savings calculations.
These often relied on reported equipment and operations data which could vary from parameters
identified during on-site verification visits and metering.
Applied adjustments decreased savings by 5%for ldaho projects, described as follows:
e Cadmus used lighting logging and verification data to confirm or adjust operating hours for three
projects. These adjustments, in addition to those made from verified fixture counts, reduced
energy savings by 4%.
o The evaluation addressed one PCN project. The participant installed the system in 2009 and
applied for an incentive in December 2009. Project files indicated Avista continued to seek
output reports from the control system to verify savings in 2011 and 2OL2. The incentive was
approved in early 2012. Cadmus contacted the facility in October 20L2 and learned the
participant had deactivated the PC network control system. Consequently, savings could not be
assigned for this project.
2.3.3 Energy Smart Grocer
Cadmus performed on-site visits to four Energy Smart Grocer program projects in ldaho: two
refrigeration case lighting projects and two walk-in case ECM projects. The study calculated an overall
realization rate for all projects in ldaho and Washington, and then applied the resulting realization rate
to savings for each state. Table 45 shows evaluated program results for ldaho, and Table 47 shows
combined results for both states.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 56 of 75
51
Table 46. Evaluated Results for PY12 Nonresidential Energy Smart Grocer Sample-ldaho
Table 47. Evaluated Results for PY12 Nonresidential Energy Smart Grocer Sample-
Combined Washington and ldaho
Adjustments decreased Idaho savings by 1%. Cadmus applied a calculation algorithm from the
Pennsylvania TRM for ECMs, which resulted in saving slightly below the reported values.
2.3,4 Site Specific
Cadmus performed site visits for 10 site-specific program pfiects, representing a variety of measure
types. The study included calculating an overall realization rate for all projects in ldaho and Washington,
and then applying the resulting realization rate to savings for each state.
Table 48 lists the different measure types evaluated as well as the number of projects and reported
savings. Table 49 shows the combined ldaho and Washin4on site-specific results. The final
extrapolation used these results as the sample drew upon a combined sampling methodology.
Table tl8. Evaluated Results for PY12 Nonresidential Electric Site-Specific Sample-ldaho
254
151,006z I L,079,628
2 1, s5ra6o
L,438,795
1,783,413 lll%_i
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G13
L.Hermanson, Avista
Schedule 3, Page 57 o175
52
Table 49. Evaluated Results for PY12 Nonresidential Electric Site-Specific Sample-
i SSHVAC 2,37L,55O
The reported savings methodology and estimates proved accurate for eight of ldaho's 10 site-specific
projects. Site-specific projects tend to be more complex, making energy-savings parameters and impacts
more difficult to estimate. The calculations also often rely on participant-supplied building, equipment,
and operations data, which may vary from parameters identified during on-site verification visits.
Cadmus found it notable that such a large portion of the projects achieved the reported savings.
Two adjustments increased ldaho savings by 2O%, driven primarily by the high realization rate for a
census-level Site Specific Other project: an industrial process measure. Cadmus conducted power
metering for several months on the project, with metering data showing retrofit power consumption
less than the reported estimate, resuhing in higher energy savings.
The other adjustment involved a Site Specific Lighting project, where Avista's documentation listed
energy savings of 151,181kWh. The tracking database reported a value of L48,O7L kWh. Cadmus
calculated energy savings of 151,005 kWh, based on the on-site verification resuhs. Comparing Cadmus'
those results with those from the tracking database resulted in a LO2% realization rate.
2.3.5 Extrapolation to Program Population
ln evaluating the nonresidential electric programs, Cadmus selected sites that could provide the most
significant impacts. As discussed, site visits sought to achieve a statistically valid sample for the major
strata. For measures in the random (non-census) sample, Cadmus calculated realization rates to apply to
programs at the remaining non-sampled sites. These realization rates were weighted averages, based on
the random verification sample, and using the following four equations:
Evaluated,,
RR,, - 7! ;.for measure j at site i' Trackedu
\Evaluated,
Rr?, ={a7 ; for measure j across all sample sites' ),Tracked,i
(1)
(2t
Washington and ldaho
3
Exhibit No.3
Case Nos. AW-E-I 3 AVU-GI 3
L.Hermanson, Avista
Schedule 3, Page 58 of75
53
\Evaluated o : kRrx),Trackedr; for measure j across all sites in measure population (3)
k
\Evaluated o
m, =fu; for the population (all sites and measures)
k
Where:
RR = the realization rate
i = the sample site
j = the measure type
k - the total population for measure type J'
| - the total program population
Cadmus calculated realization rates for each individual site in the sample based on the measure type
(Equation 1). The realization rates could then be calculated for the measure types using the ratio of the
sum of evaluated savings to the sum of tracked savings from the randomly selected sample for each
measure type (Equation 2). Non-census population evaluated savings could be determined by
multiplying the measure type realization rate from the random sample by the tracked savings for the
non-census population of each measure type (Equation 3). Adding the tracked and evaluated savings
from census stratum measures produced the total tracked and evaluated savings for each program. The
program realization rate derived from the ratio of all evaluated savings to all tracked savings
(Equation 4).
Table 50 summarizes the results for all prescriptive and site-specific programs in ldaho. The state
achieved a 95o/o overall nonresidential electric portfolio gross realization rate.
(4)
Table 50. PY 2Ot2 Gross Electric Program Realization Rates
Plescrlptly9
E!c
!SHVAC
SSL
sso
sss
Total
L2,778,400
1,585,096
t,679,069
2,735,976
L,L24,082
256,296
20,159,919
LL,746,128.
1,699,596
Lt455-,?O:8
2,57O,650
L,487,938
211,803
1erl19r-6!-1
92%
LO7%
87%
94%
t32%
83%
95%
*Full program realization rates vary from the sample realization rates above because of sample extrapolation of
the non-census level projects.
54 Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule3, Page59of75
2.3.6 HVAC/Liehting lnteradive lmpacts
The Avista portfolio results did not account for gas heating penalties caused by increased lighting
efficiency. Lighting systems convert a large portion of their input energy to useful light output, but a
substantial portion also converts to heat. Any reduction in lighting input energy also reduces waste heat.
Reducing waste heat lowers the site's required cooling load, but increases its heating load.
Cadmus noted that Avista tracked and recorded these HVAC interactive effects for many projects to
determine program cost-effectiveness. Most interactive effects involved prescriptive or site-specific
lighting projects, although some therm penalties resulted from the Energy Smart Grocer (in Avista's
electric portfolio) and site-specific HVAC program projects.
Typically, Cadmus applies interactive factors, based on values supplied by the Northwest Power and
Conservation Council's RTF. Those values vary by fixture savings, building types, and HVAC systems. Such
information, however, could not be procured for most of the affected projects evaluated. Avista
acknowledged it did not use as robust of a methodology for calculating interactive effects as that used
for its energy-savings methodology.
2.4 Conclusions
Cadmus evaluated L8 of L,49L measures installed through the program in Idaho, representing L7% ot
reported savings. Extrapolation was based on the combined, Idaho and Washington sample.
Generally, the evaluation results indicated that Avista implemented the programs well. The overall
nonresidential electric portfolio achieved a95o/o realization rate, upon comparing gross evaluated
savings to gross reported savings.
Cadmus identified the following key issues that adjusted energy savings:
o Power metering on one industrial process measure indicated lower-than-expected post-
installation power consumption, which increased energy savings.
o Li8ht logging on three projects identified a slight decrease in operating hours from the
reported values.
o Cadmus applied algorithms different from those used by PECI to determine energy savings for
ECMs. This resulted in a slight decrease in energy savings.
Cadmus identified an implementation issue affecting the impact evaluation:
o One project installed PCN in 2009, but did not provide the final data demonstrating reduced
consumption until 2012. Avista paid the incentive in20L2, but the participant reported
deactivating the system soon after.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 60 of 75
55
2.5 Recommendations
Cadmus recommends that Avista continue to offer incentives for measure installations through the
evaluated programs. Based on results from the ldaho projects, the following recommendation has been
designed for improving program energy-savings impacts and the effectiveness of evaluation:
e Avista should work with participants to accelerate the process for claiming energy savings and
paying the project incentive. Preferably, this should occur within one year of measure
installation, depending on Avista's requirements for post-installation data for a particular
project.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 61 of 75
56
3. LOW INCOME ETECTRIC IMPACT REPORT
3.1 lntroduction
ln 2011, Cadmus conducted a statistical billing analysis of 2010 low income participants, determining
adjusted gross savings and realization rates for energy-efficient measures installed through Avista's low
income weatherization program. The study examined analysis and results at the household or
participant level, ratherthan the measure level.
This report section addresses the following:
o Application of the 2010 billing analysis savings estimates to the 2012 participant population; and
o Reporting total electric impacts associated with the 2OL2 program year in ldaho.
tn the first quarter of 2OL4, Cadmus will perform a new billing analysis on 2012 participants, using pre-
period data from 2011and post-period data from 2013. ln the interim, this evaluation extrapolates
results from the recent zOilFzOLL electric impact analysis to20L2 participants.
To estimate 2010-2011 energy savings resulting from the program, Cadmus used a pre- and post-
installation, combined Conditional Savings Analysis (CSA) and Princeton Score-Keeping Method (PRISM)
approach, utilizing monthly billing data. This approach involved:
o Analyzing savings estimates for ldaho and Washington;
o Running a series of diagnostics (such as a review of savings by pre-consumption usage
quartile); and
o Conducting outlier analysis.
Avista's 207o_2077 Multi-Sector Electric lmpoct Evoluotion Report presents a detailed discussion of the
regression mode! and methodology used for this analysis.
3.1.1 Program Description
Five programs, listed in Table 51, make up Avista's Low Income Weatherization Program. Local
Community Action Partners (CAPs) within Avista's ldaho and Washington service territories implement
these low income programs. CAPs holistically evaluate homes for energy-efficiency measure
applicability, combining funding from different programs to apply appropriate measures to a home,
based on the results from a home energy audit.
Table 51 also describes measures installed under each program component, along with counts of
electric measures installed in PY 2012 and included in our electric impact analysis (a separate report
contains findings on evaluated gas measures).26
26 Cadmus. Avisto 2072 tdaho Gos Portfotio tmpoct Evoluotion Report.July 3O 2013.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 62 of75
57
Table 51. 2012 Electric-Efficiency lnstallations by Program Component
ShelUWeatherization
Fuel Conversion*
lnsulation, window/door, air infiltration,
programmable thermostat
Electric furnace, heat pump, or water heater
replacement with gas units
180
28
6I,i:[1,':fl'1il;:"." -Xi:i:flH::lHi;j:il:ffii]::ffii' N/A
N/AHVAC Efficiency High-efficiency gas furnace rep,lg,gement
*The Avista portfolio considers (and reports) fuel conversion measures as electric-saving measures.
3,2 Data Collection and Methodology
Cadmus primarily drew impact evaluation data from the program participant database. Avista provided
information regarding program participants and installed measures for ldaho. Specifically, these data
included:
o Lists of measures installed per home; and
o Expected savings from each completed measure installation.
The data, however, did not include the quantity of measures installed (such as the square footage of
installed insulation) or per-unit savings estimates.
Starting in 20L2, Avista incorporated TRM savings estimates, developed by Cadmus and specific to
Avista's low income customer segment. These measure-specific savings estimates incorporated data
from regional and secondary research (e.g., RTF, DOE) as well as input assumptions derived from
analysis of low income weatherization program participant consumption (e.g., pre-period heating
consumption).
3.2.L Documentation RevieflDatabase Review
Cadmus used the 2012 ldaho and Washington program participant database, provided by Avista, to
develop a complete population for applying the 2010-2011 billing analysis results. Participant
data included:
o Customerinformation;
o Account numbers;
o Types of measures installed;
o Rebate amounts;
o Measure installation costs;
o Measure installation dates; and
o TRM savings per measure.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 63 of75
58
3.2.2 Sampling
ln applying the 201G-2011 electric billing analysis results, Cadmus used a census of 2Ot2 program
participants, containing 81 electric accounts, including 16 electric participants receiving
conversion measures.
3.2.3 Billing Analysis-CSA Modeling Approach
To estimate energy savings from this program, Cadmus used a pre-post CSA fixed-effects modeling
method, which utilized pooled monthly time-series (panel) billing data.
The fixed-effects modeling approach corrected for differences between pre- and post-installation
weather conditions as well as for differences in usage consumption between participants (i.e., including
a separate intercept for each participant). The modeling approach ensured model savings estimates
would not be skewed by unusually high-usage or low-usage participants. Pairing monthly consumption
between pre- and post-months maintained the same time frame for evaluating unique participants.
Additional details regarding the 2010-2011 billing analysis can be found in the Avr3to 207F2O77Multi-
Sector Electric lmpact Evoluotion Report.
3.2.4 Estimating Conversion Participant Savings
While the program historically installed electric to gas fuel-conversion measures in Washington, Avista
introduced these measures to ldaho participants starting \n2OL2. Given the 201f2011 analysis of
conversion measures only addressed Washington installations, this study scaled these savings estimates
using average heating degree days to apply to ldaho customers. This approach assigned savings to
conversion participants (n = 15), based on the specific electric to gas conversion measures installed.
Table 52 provides energy savings estimates assigned to ldaho conversion measures.
Table 52. ldaho Electric Conversion Energy Savings
*Given the low precision in modeling furnace-only impacts in the 201(F2011 study, reported savings represent the
difference between modeled combination participant savings (those receiving both furnace and water heater
conversions) and water-heater only participant savings.
3.3 Results and Findings
3.3.1 Summary of Program Measures
Table 53 shows the count and average reported TRM savings for 2OL2 electric-saving measure
installations in ldaho (including non-conversion and conversion participants). lnfiltration measures
exhibited the highest count, followed by windows and floor insulation. Duct insulation achieved the
highest average reported TRM savings.
Exhibit No.3
Case Nos. AVU-E-13 AVU-C-I3
L.Hermanson, AMeta
Schedule 3, Page 64 of 75
59
Table 53. Average Reported Savings and lnstallation Count by Measure
Attic insulation
Doors
Duct insulation
Floor insulation
Furnace replacement (conversion)
lnfiltration controls
Refri gerator replacement
Wall insulation
Water heater.replacement
Water heater replacement (conversion)
Windows
20
29
6
32
L4
57
N/A
2
6
t4
34
t,478
287
5f 48-5
4,408
N/A
L,87L
N/A
3,466
299
N/A
2,432
To highlight some distinctions in Avista's reported savings that contributed to changes in realization
rates, Cadmus compared average expected measure savings from the 2OLO-1OLL period to the 2012
TRM estimates. Figure 10 highlights the differences between reported average savings.
Figure 10. Comparison of 2010-2011 and 2012 Average Reported Savings by Measure
.20LO-20LL r 2012 (TRM)
Attic insulation
Duct insulation
Floor insulation
lnfiltration controls
Wall insulation
Windows
1,000 2,000 3,000 4,000 5,000
Reported Electric Savings (kWh)
6,000
A number of measures reported considerably lower savings in 2012 using TRM estimates than the
2010-2011 average savings, most notably: insulation measures, windows, and infiltration controls. The
different years, however, generally offered a relatively similar mix of measure installations, with
infiltration controls and window replacements the most frequently installed measures for electric-
saving participants.
N o n -Co nve rsi o n Po rti ci pa nt Re su lts
Applying savings estimates from the billing analysis to the electric-saving participant program population
produced total savings of L,602 kWh per participant. Cadmus applied these modeled savings estimates
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 65 of 75
60
to electric-savings participants not receiving conversion measures, and calculated average reported TRM
savings by: summing measure savings at each household; and then taking the mean household savings
across individual participants. Table 54 compares average participant TRM savings and modeled savings
for non-conversion customers.
104130
Conve rsi on Pa rti ci po nt Res u lts
Of the Sl total ldaho gas-savings participants, 16 received electric-to-gas conversion measures, including
electric-to-gas furnace and water heater replacements. The analysis considered these participants
separately, as the methodology for estimating evaluated savings differed slightly different from the non-
conversion participant Broup. Table 55 provides a distribution of all Avista-funded conversion measure
installations and their associated energy savings. Each group of conversion participants exhibited a high
realization rate, with an overall realization rate of 103%.
Table 54. Non-Conversion Gas Savings
Table 55. Measure lnstallations for Conversion Participants
Furnace Only
DWH Only
Combo
Total
2
2
L2
15
!6,428
8,026
146,938
17,-,392
8,506
4,L62
L2,658
L7,0t2
8,324
Ls2pL7
L77,353
L04%
L04%
L03%
to3%
3.3.2 Overall Participant Results
Table 56 provides overall electric savings, including savings attributed to fue! conversion participants.
104,130 L77,353 28L,483 274,9L3
3.4 Conclusions
Upon comparing the 20LF2OLL and 2012 results, changes in Avista's expected savings calculations led
to differences in realization rates. Average reported electric savings per (non-conversion) participant
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 66 of 75
Table 55. Overall2012ldaho Electric Savings
51
decreased by 56% between the examined periods, falling from 3,525 kWh in 20Lrl20LL to 1,593 kWh in
2012 (based on the TRM). This appeared to primarily drive shifting realization rates for non-conversion
participants, lrom 27% for ldaho in 2010-2011to LOl% in 2012.
As shown in Figure 10, all measure-level estimates observed significant changes in kWh savings between
the 201G-2011 reporting and the 2012 TRM estimates, with these decreases in average savings ranging
from three to approximately 100 times the previously reported estimates, most notably for windows
and duct insulation measures.
3.5 Recommendations
Cadmus' recommends the following enhancements to improve program impact results:
o ln future billing analyses, use a control or comparison group. For upcoming impact evaluations
revisiting the billing analysis, Cadmus suggests using 2013-2014 participants as a control group
to analyze the treatment group of 20t2 participants. For such analysis, 2011 and 2013 annual
participant consumption histories would serve as the pre- and post-periods. Using a control or
comparison group of nonparticipants would allow analysis to control for exogenous factors
(e.g., macroeconomic, rate changes, technologicaltrends)that could result in trends affecting
consumption. Controlling for these trends using a control/comparison group reflects a more
robust experimental design and defensible methodology for estimating accurate energy-
savings impacts.
o Work with ldaho agencies to provide refriterator replacements. Refrigerator replacements can
result in significant electric savings. Avista should work with local CAP agencies and other ldaho
stakeholders to identify the best ways to encourage integrating these measures into
program delivery.
o Include high-use customers in program targeting. While prioritization guidelines for targeting
low income weatherization participants are set at the federal level, some utilities, for targeting
purposes, actively track customer usage and provide agencies with lists of customers
experiencing particu larly high energy consumption.
r Notably, DOE protocols list high-energy consumption as a factor allowed in participant
prioritization. ln such cases, along with other targeting criteria (e.g., families with children,
senior citizens), agencies may incorporate energy-consumption characteristics into their
program participant prioritization. Not only would weatherizing high-use customers likely
result in higher energy savings, but the program could provide some financial relief for
customers overly burdened with energy bills due to their housings' characteristics.
r Methods exist for identifying high-usage customers while controlling for factors contributing
to consumption (e.9., square footage, income, numbers of people per household). Avista
should utilize such approaches.
! Given reductions in federal funding for weatherization and associated reduced agency
capacities resulting in more limited leveraging opportunities, Avista has an opportunity to
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 67 of 75
62
lead new efforts for continued delivery of energy-savings resources to low income
residential customers. By considering high-usage targeting, potential exists to secure cost-
effective energy savings through one segment of this population, while continuing to
support weatherization for income-qualified customers, which may result in lower savings
and prove less cost-effective. Efficient targeting can aid in balancing these efforts to provide
whole-house weatherization, while continuing to leverage the agency network as a resource
for outreach and delivery.
Track and compile additional data from agency audits. These data include information on
primary and secondary heating and cooling, and on the size of a home. As an inexpensive
alternative to gas heat, gas customers may turn to electric room heaters and wood stoves,
reducing the impacts of weather-sensitive measures installed through weatherization (e.g.,
insulation). Collecting information on customers' primary heating usage during weatherization
would provide more reasonable savings estimates.
Cadmus recommends Avista work with agencies to develop explicit, on-site tracking protocols
for collecting information on participant heating sources. Agencies should collect the following
information to better inform heating (and cooling) sources:
r Visual inspections of all heating equipment found on site;
r Participant-reported primary and supplemental heating sources used;
. Quantities of secondary heating, if applicable (e.g., numbers of electric room
heaters); and
! Any indicators suggesting discrepancies between actual and reported primary heating.
Consider performing quantitative, nonenergy benefit analyses. With respect to ongoing,
Advisory Group discussions that address quantifying non-energy benefits, Cadmus recommends
Avista consider pursuing additional analyses, aimed at quantifying nonenerg'y benefits
associated with low income weatherization and applicable to the TRC test. ln particular,
analyses of economic impacts and payment pattern improvements (including reduced
arrearages and collections costs) can provide program stakeholders with the monetized values
of benefits. CIher utilities have used such analyses in reporting low income weatherization cost-
effectiveness in the Northwest (e.g., ldaho, Washington). Standard cost-effectiveness testing,
using the TRC test, accounts for all program costs (only including energy savings as program
benefits), but clearly omits some genuine nonenergy benefits experienced by participants (as
discussed in greater detail in the 2070 Process Evoluotionl.
Exhibit No. 3
Case Nos. AW-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 68 of 75
53
4. CFL CONTINGENCY PROGRAM
4.1 lntroduction
Cadmus' previous evaluation2T estimated the percentage of butbs installed by the end of calendar year
2011 and only provided the savings associated with these bulbs. This report provides total energy
savings achieved by the program in the first year and calculates energy savings installed in 2OL2 as the
difference between the total program savings and evaluated PY 2011 savings.
4.L.1 Program Description
The CFL Contingency program's design intended to deliver highly cost-effective, energy-efficiency
resources to Avista's customer base (both residential and small commercial), while simultaneously
maintaining the utllity's flexibility to meet anticipated energy acquisition targets at a lower ratepayer
cost and with a minimum of uncertainty.
Starting in July 2011 and continuing through November 2011, Avista sent residences and small
businesses within the utility's territory a box of eight ENERGY STAR CFLs of varying sizes, accompanied
by literature on the benefits of their use and instructions on proper disposal and bulb placement.
Customers also received information about returning the CFLs, at no cost to the customer, should they
decide not to keep them. Customers also could request additional bulbs.
4.2 Methodology
For evaluating the savings achieved by the CFL Contingency Program, Cadmus completed an engineering
review, based on the previous evaluation analysis, but updated to include recent evaluation results and
expected regional decisions.
Six parameters informed the calculation of gross savings for the lighting component:
@r@r@r@x@r@=@
Wattage of the mailed ENERGY STAR CFL
The difference in wattage between baseline bulb and the CFL, divided by the
wattage of the CFL
Daily lighting operating hours
27 Cadmus. Avista 207(>2077 Multi-Sector Electric lmpoct Evoluotion Report. May 20L2.
Where:
CFL Watts =
DWM =
HOU =
Exhibit No.3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 69 of 75
64
DAYS =Days per year (355)
An adjustment representing the interactive effects of lighting measures on
heating and cooling equipment operations
The percentage of units installed
The annual savings algorithm derived from industry-standard engineering practices, consistent with the
methodology used by the Northwest RTF. Discussions of each input follow.
4,2,1 CFL Wattage
This assumption did not change from the previous analysis. The program delivered over 2.3 million CFLs
to residential and commercial customers in Avista's territory, with the distribution shown in Table 57.
The CFL wattage derived from the weighted average of delivered units to each sector. The residential
sector had an average delivered CFL wattage of 18.30 watts, and the commercial sector had an average
delivered CFL wattage of 18.25 watts.
4.2.2 DWM
This assumption did not change from the previous evaluation. Cadmus relied on the RTF (for residential)
and the 5s Power Plan (for commercial) to determine the DWM. Adjusting the RTF's residential DWM
allowed incorporation of Avista's survey results that documented the room distribution of installed
bulbs. Thus, the DWM for residential installation changed from the RTF's 2.50 to 2.63.28 The commercia!
DWM was 2.70, based on the 6s Power Plan lighting workbook.
This analysis did not consider EISA's potential impact. EISA could only impact the baseline for the 55,115
23-Watt CFLs mailed to Washington residents. Only the first round of packages included these bulbs,
which appeared to have almost achieved the maximum ISR by the end of 2011 according to surveys.
4.2.3 HOU
This residential assumption has been updated, based on recent evaluation results. Cadmus estimated
the CFL HOU for residential installations using: Avista's survey of room types; and a multistate modeling
approach, built on Iight logger data collected from five states (Missouri, Michigan, Ohio, Maine, and
2a The RTF DWM represents the 2011 baseline, and does not include federal EISA impacts startint in20t2.
Exhibit No.3
Case Nos. AW-E-13 AVU-GI3
L,Hermanson, Avista
Schedule 3, Page 70 ot75
Table 57. Total Units of Delivered CFts by State and SectorType
_ $92!q
55,116
65
Maryland).ze The Maine HOU study, completed in the past year, was added to the model used for the
previous evaluation. The average HOU was calculated using a regression statistical modelthat combined
multistate, multiyear data. Cadmus used the multistate model's estimate of HOU by room type,
weighting this based on Avista's survey results to determine an overall HOU average of 2.38, a 3o/o
reduction from the 2.45 estimated previously.
For commercial HOU, Cadmus used the 6th Power Plan's documented lighting hours of operating for
each building. After gathering building type information from Avista's survey of commercial participants,
Cadmus weighted the 10.16 lighting hours from the 6th Power Plan to calculate 10.02 for Avista's
commercia! HOU. The assumed commercial HOU did not change from the previous analysis.
4.2.4 WHF
This assumption did not change from the previous evaluation. The WHF accounts for changes in annual
HVAC energy (lost or gained) due to reductions in facility lighting energy. Cadmus based the WHF on
SEEM building models, developed by the Northwest Power and Conservation Council. These SEEM
building models estimated the change in HVAC equipment energy use due to a change in lighting
technology (e.g., incandescent lamps to CFLs). ln general, the models accounted for interactions using
load-shape profiles of the HVAC and lighting equipment, based on dwelling occupancy.
The Northwest Power and Conservation Council uses an inherently conservative method that assumes a
closed shell (i.e., all interior lamps, including ceiling recessed cans would be contained in a closed
system, hence any heat generated by the bulbs would go into the building). ln reality, waste heat could
transfer out ofthe conditioned space.
Cadmus based the WHF calculation on Avista's share of electric heating equipment,30 along with its
associated efficiencies and its surveys of interior and exterior distribution, to obtain a WHF of 89.8%.31
Cadmus used the commercial WHF of 85.5% provided in the 6th Power Plan.
4.2.5 tSR
An update to this assumption allowed estimates of the percentage of bulbs installed. The ISR used in this
analysis represented the percentage of bulbs believed to be installed within one calendar year of the
receipt of the CFL package.
The Cadmus Group, lnc. 2070 Evaluation, Meosurement, dnd Verificotion Report. Dayton Power and Light.
March 15, 2011,
Saturations of Avista equipment types are based on the 2011 participant survey for the CFL Contingency
Program.
The RTF WHF is 85.4Yo; the adjusted Avista wHF is 89.8%.
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 71 ol75
66
ln December 2012, the RTF approved the Reside ntiol: Lighting-Speciatty CFts workbook,32 the only
residential CFL workbook reviewed by the RTF since the Northwest Energy Efficiency Alliance (NEEA)
RBSA data became available. Based on the RBSA results, the approved workbook assumed a24% storage
rate for residential specialty CFLs. Cadmus assumed that, since the data used to develop this storage
rate was not specific to specialty CFLs, the RTF will update its storage rate assumption for all CFLs to this
value upon updating the Residentiol: Lighting-CFts workbook later this year. When combined with an
assumed 3.57% removal rate, a 73.6% first-year ISR results for direct mail CFLs,
4.3 Overall Program Savings
Cadmus calculated PY 2072 savings by subtracting the PY 2011 evaluated savings, calculated in the
previous evaluation, from the total program savings calculated in this evaluation. Table 58 shows
achieved annual savings by year, state, and sector.
Table 58. CFL Contingency Program Evaluated and Expected Savings by State and Year
WA
Residential lD
Total
WA
Commercial lD
Total
Tota!
42,95L,93L
18,698,560
61,650,591
g,609,gg3
7,079,548
t5,689,4t
23,347,564
10,L43,973
33A9t,536
3,826,229
3,L46,L45
6,972,374
77,y10,o32
L9,604,367
8,554,687
28,159,054
4,783,664
3,933,403
8,717,O57
q,453,gto
32 htto://rtf . nwcou nci Lorelm easures/measure.aso?id= 142
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 72 oI 75
67
5. PORTFOTIO GROSS AND NET SAVINGS
5.1 Gross Portfolio Savings
The 2012 ldaho electric portfolio consisted of several sectors and many program delivery streams. In
total, the programs achieved a 98.7% gross realization rate and total gross savings of 37 ,483,952 kwh
(Table 59).
*lncludes CFL Contingency savings.
5,2 NTG Adjustment
Cadmus evaluated NTG through customer self-reports, utilizing different methodologies and data
sources for the different programs, as detailed below.
5.2.t Residential NTG
NTG values were updated for the 2012 residential population. Freeridership and participant spillover
was determined from participating customer self-reports from 274 phone surveys performed during Q2
2013. The methodology is consistent with that described in detail in a full NTG report published last
year."
Non-participant spillover was calculated from 1,051 completed surveys (380 in ldaho) from our multi-
method General Population survey. 3,000 paper surveys were mailed to randomly selected residential
customers in both lD and WA. These mailings included a website to complete the survey online, and
finally, a subset of the sample was called with a traditional phone survey. This multi-media method
helps reduce survey bias. The Second Refrigerator and Freezer Recycling program has a specific NTG
methodology that is discussed in detail in Section t. Table 60 outlines the NTG components and
resulting program level NTG.
Residential*
Nonresidential*
Low lncome
Total
ENERGY STAR Products
Heating_ and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
Space and Water Conversions
L3,627,696
24,093,322
274,9L3
37,*5,931
77%
6tYo
.59%
77%
63%
14,098,455
23,LO4,034
281,483
37,48.3,952
L035%
9s.9%
L02.4%
98.7%
0%
0Y"
L,4Yo
,
o%
0.t%
0%
L.9%
4.6%
0o/o
0%
23%
4LYo'
47Yo
,
23%:
37%
33 Cadmus. Net-to-Gross Evaluotion of Avistd's Demand-Side Manogement Progroms.June 2012.
Table 59. 2012 ldaho Gross Savings
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 73 ol 75
58
Table 51 shows the NTG values and resulting net savings for Avista's residential downstream programs.
Table 51. Residential NTG and Net Savings
Second Refrigerator and Freezer Recycling
ENERGY STAR Products
Heating and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
Space and Water Conversions
ENERGY STAR Homes
Total
350,968 44%
193,963 23%
67L,428 4t%
37,373 47%
8,861 23%
34L,977 37Y:
24,698 74%
1,629,268 39%
154,811
44,6L1
274,4L3
.t7,535
2,039
L27,OLo
L8,277
538,695
5.2.2. Nonresidential NTG
To reduce survey fatigue for Avista's nonresidential customers, Cadmus did not perform any data
collection with 2012 program participants, and does not have updated NTG information. Surveys,
planned for the 2013 participant population, will be performed in Q1 2014. This report uses NTG values
from the 2011 analysis,3a which can be found in Table 52, along with the resulting net savings. The
nonresidential sector exhibited a weighted nonresidential NTG of 75%.
Energy Smart Grocer
Prescriptive
Site-Specific
Total
5.2.3 No NTG Adjustment
1,698,686 96%
LL,746,328 67%
5,725,6L7 83%
19,L70,53L 75%
1,630,739
7,9L7,025
4,769,439
L4,3L7,2O3
The following programs did not receive a NTG adjustment as the original savings analysis methodology
accurately reflected net market characteristics: Low lncome, Simple Steps, and CFL Contingency.
Low lncome
Commonly, low income programs receive a LOO% NTG, as the energy-efficient upgrades are performed
at no cost to the home owner, and are considered a social good.
Simple Steps
The savings analysis methodology for Avista's upstream lighting program follows the RTF-an
organization that does not differentiate between gross and net savings in favor of using an adjusted
market baseline approach. As discussed in Section 1, the various inputs to the savings calculation either
34 Cadmus. Net-to-Gross Evaluotion of Avista's Demond-Side Monogement Progroms.June 2012.
Table 62. Nonresidential NTG
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L.Hermanson, Avista
Schedule 3, Page 74 ot 75
59
used direct RTF values or RTF methods with Avista-specific data. To assign an additional NTG value to
this program would, in effect, be double counting.
CFL Contingency
The CFL Contingency program sent bulbs at no cost to Avista customers. As consistent with the 2011
evaluation, no NTG adjustment is applied to these bulbs.
5.3 Net Portfolio Savings
The portfolio achieved an overall NTG ratio of 84% and 31,539,951 kwh of net savings. Table 53 shows
verified gross and resulting net savings for ldaho's 2012 DSM programs. Note that the residential and
nonresidential NTG values are higher here because of the inclusion of the CFL Contingency savings that
receive 100% NTG.
Residential* L4,098,435 93%
Nonresidential* 23,L04,034 79%
Total 37N3,9?2 U%
* lncludes CFL Contingency savings.
Residential
ttonreiidential
Low lncome
Total
7,495,L08
8,423,(X)0
L,L96,892
17,115,000
L3,t07,862
181250,606
28L,483
31,639,951
L3,1O7,862
18,2=5_0,606
281,483
31,539951
t74.9%
2L6.7%
23.5%
t8r..9%
5.4 IRP Goals Achievement
Table 54 shows net verified savings, as compared to the lntegrated Resource Plan (lRP) goal of
17,115,000 kWh. The IRP states its goal as a portfolio-level target; so, for purposes of sector-level
comparison, Cadmus adopted the Avista Business Plan goals by sector, and applied those proportions to
the IRP target. The 2012 program year achieved 184.9% of the IRP target in ldaho with 31,539,951 kwh.
Even excluding the CFL Contingency savings, ldaho still surpassed the IRP goal, at 111.9% with
19,151,861kwh.
Table 63. 2012 ldaho Net Savings
Table 64. 2012 Reported and Gross Verified Savings for ldaho
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L.Hermanson, Avista
Schedule 3, Page 75 ot75
70
KE
FINAT REPORT
Avista 2Ot2ldaho Gas Portfolio
fi[fli t'.::- :]i
?l}r3 s[P 30 fr?i lt]: lB
lil'i:-: -i ;''--i'" ',r , -rlTIL.t-l 1,:ii i l:,ii:i.riii.i. ji i
lmpact Evaluation Report
July 30, 2013
Avlsta Corporation
1411 E Mission Ave
Spokane, WA 99220
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 1 of45
Prepared by:
Danielle C6t6-Schiff Kolp, MESM
AndrewWood
Jeff Cropp, P.E.
Scott Reeues
M. SamiKhawaja, Ph. D.
Cadmus
EfiibitNo.3
Cas6 Nos. AW-E-13AVt -G,13
L. Hermamon, Aviata
Sdredub 4, Pagc 2 of45
Savings Resuhs
1.1 lntroduction..........,....5
L.2 Methodology............................. 6
2.3 ResultsandFindings....................................24
3.5 Recommendations....
Appendix 1A: Residential ENERGY STAR Home Model |nputs............... .........'.,...'...........41
Appendix 18: Electricity Savings Achieved by Residential Gas Programs... ......................42
Exhibit No. 3
Case Nos. AW-E-I 3 AVU-G1 3
L. Hermanson, Avista
Schedule 4, Page 3 of45
Portfolio Executive Summary
Avista Corporation contracted with Cadmus to complete process and impact evaluations of the
company's 2OL2 gas demand-side management (DSM) programs. Avista has been administering DSM
programs to reduce energy use of electricity and natural gas for its portfolio of customers for several
decades. Most programs are implemented in-house, but a few utilize external implementers. This report
presents our impact findings for the PY 2OL2 gas portfolio in the state of ldaho.
Evoluation Activities
For each of the three sectors-residential, nonresidential, and low income-we employed a variety of
evaluation methods and activities, as shown in Table 1.
ENERGY STAR
Products
Heating and Cooling
EfficiencvResidential weatherization/shell
Water Heater
Efficiency
ENERGY STAR Homes
Nonresidentiat ll"ttjiptlY: Programs
5tte-5peclflc
. Low lncomeLow lncome Programs
Savings Results
Table 2 presents sector-level reported and gross verified savings values and realization rates. Overall,
the ldaho portfolio achieved a 94.4Yo realization rate, and acquired 2L6,766 in annual therm savings.
{
,/ ,/
Residential
Nonresidential
Low lncome
Total
L23,696
96,452
9,363
229,508
t21,978
83,729
11,059
2t6,766
98.6%
86.8%
118j%
94.4%
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 4 of 45
Table 3 shows the gross verified savings, compared to the lntegrated Resource Plan (lRP) goal of 746,728
therms. The IRP states its goal as a portfolio level target; so, for purposes of sector-level comparison,
Cadmus adopted the Avista Business Plan goals by sector, and applied those proportions to the IRP
target. The 2OL2 program year achieved 29.0% of the IRP target in ldaho.
Table 1. 2012 Gas Programs Evaluation Activities
Table 2. 2012 Reported and Gross Verified Savings for ldaho
Table 3. 2012 lRP Goals and Gross Verified Savings for ldaho
281,039
4/;0,478
25,2t2
746,728
727,978
83,729
11,059
216,766
Key Findings ond Conclusions
Residential
For PY2OL2, Avista's residential gas programs produced L2L,978 therms in savings, yielding an overall
realization rate of 98.6%. Residential gas savings achieved 43% of Residential IRP goals.
The evaluation produced the following, major, residential program conclusions:
o Overall, residential gas customers responded well to the programs, and often installed several
measures within the same year.
o Avista's program and tracking databases were adequate for evaluation purposes, providing
sufficient contact information, and measure and savings information. The database review
confirmed the information was reliable and accurate.
o All measures rebated through the program had been installed and continued operating, With
one exception, all measures reviewed met the program qualification standards.
Nonresidential
Cadmus evaluated Ll of 77 measures installed through the nonresidential energy-efficiency programs,
representing 39%of reported savings. ForPYZOLZ, Avista's nonresidential gas programs produced
83,729 therms in savings, which yielded an 86.8Yo overall realization rate. Nonresidential gas savings
achieved L9% of Nonresidential IRP goals.
Though Cadmus determined that Avista generally implemented the programs well, the following key
issues reduced the evaluated energy savings below the reported value:
. At times, the programs provided incentives for measures that may not have been appropriate,
such as a night-time temperature setback for a laboratory operating at consistent temperatures.
o Post-installation inspection process may not have always identified operational issues with
rebated equipment. An example is the Site-Specific HVAC census project, for which Avista staff
verified the lighting measure but performed only cursory review of the HVAC measure.
Low lncome
For PY2OL2, Avista's low income gas programs produced 11,059 therms in savings, yielding an overall
realization rate of 118.1%. Low income gas savings achieved 44% of Low lncome IRP goals.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 5 of45
When state-level ldaho savings estimates from the 2010 gas billing analysis were applied to 81 gas-
saving 2012 program participants (not receiving fuel-conversion measures), 123 therms per home
resulted.
An additional 16 participants received fuel conversions for electric heating and/or water heating
equipment, along with bundles of other gas-saving weatherization measures (e.9., insulation). We
assigned savings to three categories for these conversion participants: full model savings; partial model
savings; and no model savings (only technical reference manual pass-through savings). ln total, we
estimated an additional 1,095 therms in savings for gas-saving conversion participants.
Recommendotions ond Further Anolysis
Residential
Based on the evaluation results, Cadmus offers the following recommendations to Avista:
o List energy factors (or, at least, model numbers) for appliances. lncluding more information
about the actual efficiency of equipment installed allows for greater accuracy in estimating gross
energy savings achieved.
o lf possible, include existing equipment information.
o lf the measure is reinstated, consider moving all ENERGY STAR Clothes Washer rebates to the
electric program.
The following research recommendations draw upon this impact evaluation's results and from known
future changes to program requirements:
r Perform a targeted billing analysis on weatherization participants who use both electricity and
gas to heat their homes.
o Perform a billing analysis on ENERGY STAR homes using a nonparticipant comparison group
once enough homes have participated under the new requirements to justify conducting
the work.
Nonresidential
Cadmus offers the following recommendations for improving program energy-savings impacts and
evaluation effectiveness:
o Review whether reported HVAC measures are appropriate for facilities with consistent space
conditioning requirements, such as laboratories.
e Consider focusing post-installation inspections on projects with the highest level of tracked
energy savings.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 6 of 45
Low lncome
The impact evaluation revealed several areas where program performance and savings calculation
accuracy could be improved. Consequently, we recommend that Avista consider the following:
r lnclude a control/comparison group in future billing analyses.
o Consider targeting high-use customers.
o Track and compile additional data from agency audits.
o Consider analyzing easy-to{uantify, non-energy benefits, which can be added to program cost-
effectiveness reporting.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 7 of45
L 2OL2 ResidentialGas Impact Report
T.T lntroduction
During the 2012 program year, Avista's residential gas demand-side management (DSM) programs in
ldaho reported savings of 123,693 therms for 1,802 measures. Avista's 2012 DSM residential gas
programs included:
r ENERGY STAR Products
o ENERGYSTAR Homes
o Heating and Cooling Efficiency
. Water Heating
o WeatherizationMeasures
This report explains the methods used to qualify and verify these savings.
1.1.1 Evaluation Methodology
We designed our impact evaluation to verify tracked program participation and energy savings using:
. Data collected in the tracking database;
o online application forms;
r Phone surveys; and
. Applicable deemed values developed for Avista's technical reference manual (TRM).1
As shown in Table 4, Cadmus employed up to two evaluation methods and activities for
each program.
ENERGY STAR Products
Heating and Cooling Efficiency
Residential Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
1.L,2 Energy Savings
Table 5 shows aggregated adjusted gross savings and resulting realization rates by program.
' ln 2OU's first quarter, Cadmus created a TRM for use in deemed measure savings calculations, and updated it
where necessary for the 2012 program year.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 8 of45
Table 4. Evaluation Methodology
Table 5. Reported and Adlusted Gross Savings
ENERGY STAR Products
Heating and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
Total
3,2s6
106,591
Lt,448
468
L,829
123,693
2,490
10s,837
LL,357
455
L,829
Lzt,97A
76.5%
99.2%
99.2%
99.-2%
t00.0%
98.6%
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 9 of 45
Table 6 shows reported measure counts. We verified savings of 121,978 therms through the installation
of 1,802 measures during PY 2012. Overall, residential gas programs achieved an adjusted gross
realization rate of 98.6%.
ENERGY STAR Products
Heating and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
Total
532
t,037
t72
52
I
1,8o2
7.2 Methodology
1.2.1 Sampling
Cadmus randomly sampled program participants to complete surveys. Cadmus also randomly sampled
participant applications to be reviewed for this evaluation. The following subsections describe methods
used to select the required samples.
Record Review Sampling
To determine the percentage of measures incented that qualified for the program, Cadmus designed
sample sizes to yield significance at the 90% confidence and t10% precision levels for each application
type, across both states and fuels. Cadmus randomly selected participant measures for a record
qualification review from the 2012 gas and electric program populations. We sampled participants using
a single measure record. However, if a customer applied for multiple rebates on the same application
form during the program year, the record review checked all measures included in the application for
qualification, whether for electric or gas.
Table 7 shows the number of record reviews completed for unique accounts and unique measures.
Table 7. Measure level Record Review Completes
Total Participants Reviewed
Total Measures Reviewed
277
250
Suruey Sampling
For program-level survey results, Cadmus designed participant survey sample sizes to yield significance
at the 90% confidence and +7O% precision levels for each program within each ldaho and Washington,
The participant survey sampling plan drew upon on multiple factors, including:
o The feasibility of reaching customers;
o The program participant population; and
. Research topics of interest,
Customer fuel types did not factor in survey sampling.
Cadmus did not survey home buyers for the ENERGY STAR New Homes program because home builders
received the rebates. The evaluation completed a total of 274 surveys with ldaho participants. Table 8
shows: the number of surveys achieved; and the resulting absolute precision for each program. Note
that the absolute precision achieved did not always meet the t10% goal, but is safely within the portfolio
precision goal of 90/10.
Table 8. Participant Survey Sample Sizes and Savings-Weighted Precision Estimates by Program
Space and Water
Conversions
Water Heating
] ENERGYSTAR
Products
Heating and Cooling
Efficiency
Second Refrigerator
and Freezer Recycling*
Weatherization and
Shell Measures
38
L27
11
26
73
7t
30
50
70
78.9%
39.4Yo
3.0%
3.9/o
L7.3%
27.7%
t20%
!73%
t9%
tto%
t9%
tt3%
Exhibii No. 3
Case Nos. AW-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 1 0 of 45
2,323
1,805
345
22L 31
70
60
60
*This program did not claim therms savings.
Cadmus randomly called program participants included in survey sample frames. Geographic
distributions of survey respondents clustered around urban centers within Avista's service territory (for
both states); specifically the cities of Spokane, Pullman, Moscow, and Lewiston, as shown in Figure 1.
Figure 1. Geographic Distribution of Participant Survey Completes
Pond OrEle
ontan
ldaho
Oregon
i-----.l#
L.2.2 Data Collection and Analysis
Record Review
Cadmus reviewed all records for the selected sample of accounts, checking them for completeness and
program compliance using the data they contained. Measures qualified if all data found in the
application complied with the program specifications. As the evaluation randomly sampled customers
by application type (several measures can be found on different application forms), we tracked
qualification rates at the application type level.
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 11 ot 45
The review revealed one improperly issued insulation rebate on a Home lmprovement application, as it
had an existing R-value above the participation requirements (the applied qualification rates include
this result).
Surveys
Cadmus contracted with Discovery Research Group (DRG), a market research firm, to conduct surveys
with sampled participants. To minimize response bias, DRG called customers during various hours of
days and evenings (includinB weekends), and made multiple attempts to contact individual participants.
Cadmus monitored survey phone calls to ensure accuracy, professionalism, and objectivity. We analyzed
the survey data at the program level rather than the measure level, and weighted survey results at the
portfolio level by program participation to ensure proper representation.
Dotabose Analysis
Cadmus reviewed the participant database Avista provided to check for inconsistencies in tracked
savings and measure duplications, This review did not identify inconsistencies in data tracking. All
tracked savings were based on the 2012 Avista TRM.
Unit Energy Sovings Analysis
Cadmus updated the unit energy savings achieved by ENERGY STAR Clothes Washers, based on new
survey data of Avista participants. We did not update other unit energy savings in the TRM.
1.2.3 Verification Rates
Cadmus determined verification rates for each program, but not for each measure. Where applicable,
the review covered the following topics:
o Checking that the database tracked the correct measures;
. Accounting for correct quantities; and
. Determining whether units remained in place and were operable.
All measures researched remained in place and were operable, resultin gin a LO0Yo verification rate.
L.2.4 Measure Qualification Rates
Cadmus considered a measure qualified if it met the various requirements particular to its category,
such as receiving an ENERGY STAR certification or achieving program minimum efficiency standards.
When necessary we conducted online database searches for model numbers, and noted necessary
characteristics to verify achievement of all qualifications.
Out of the entire verification sample, we identified one nonqualified measure:
o An attic insulation project had a base case condition that should have prevented it
from qualifying.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 12 ot 45
7.3 Progrom Results and Findings
1.3.1 Overview
End results from the review produced total adjusted gross savings for each measure and program as
well as overall realized savings for each program. The following sections describe each program, explain
analysis steps taken, and discuss results and findings.
Calculating the measures' adjusted gross measure savings required the following steps:
1. Reviewing the database to determine whether adjusted measure counts correctly represented
the number of measures installed.
2. Conducting a phone survey with a sample of customers to verify measure installations.
3. Reviewing records to determine measure qualification.
4. Calculating verification and qualification rates.
5. Calculating deemed measure savings for rebated products.
5. Determining adjusted gross savings for each measure by applying the above-calculated rates
and deemed savings to measure counts.
1.3.2 ENERGY STAR Products
Progrom Description
The ENERGY STAR Products program included the following gas measures:
o Clothes washer (gas)
o Dishwasher (with gas water heater)
The program offered direct financial incentives to motivate customers to use more energy-efficient
appliances. The program indirectly encouraged market transformation by increasing demand for
ENERGY STAR products. The program included electric and gas measures, though this report focuses on
gas savings.2
Anolysis
Energy savings credited to the ENERGY STAR Products program had to meet multiple criteria:
o Measures had to remain in place and operate properly at the time of verification;
o Numbers of installed equipment pieces and their corresponding model numbers in the
applications had to match the database; and
o Units must have been ENERGY STAR-qualified at the time of the program offering.
' See Appendix 18 for the electricity savings achieved through the gas program.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 1 3 of 45
Cloth6 Washe6
Energy saving calculations drew upon a 20O9 Cadmus metering study,3 which metered more than 10O
clothes washers in California homes for three weeks; the largest ,n situ metering study on residential
clothes washers and dryers conducted in the last decade. The study indicated higher consumption and
savings values than those often estimated.
Dryers produced the majority of energy consumption and savings, as high-efficiency washing machines
removed more moisture from clothes, allowing shorter drying times. As most energy savings resulted
from decreased dryer use, the study had to estimate the percentage of homes using gas domestic hot
water heaters and electric dryers. The Regional Technical Forum (RTF) advocates an 82% assumption,
which this analysis used. Consequently, 82% of installations of ENERGY STAR clothes washers in homes
with a gas domestic hot water heaters achieved significant amounts of electricity savings.
oetermining adjusted gross savings required using the following, additional input assumptions:
. Recent independent evaluation surveys from the Residential Building Stock Assessment (RBSA)
and 2OL2 Avista Participant surveys estimated 252 washing cycles per year. Unit energy savin8s
values have been adjusted accordingly, as reflected in the realization rate for this measure.o
o Cadmus utilized the California metering study to estimate consumption per wash and dry cycle
for the base and efficient equipment.
Dlshwashers
Cadmus estimated dishwasher savings based on methods currently used in the ENERGY STAR Calculators
(the only calculator available providing consistent energy-savings estimates in the presence of a gas or
electric domestic hot water heater). The following input assumptions were applied:
o Cadmus calculated the average base case and efficient case Energy Factor (EF), with both based
on data utilized by the RTF. The baseline EF equaled the average market efficienry of units not
qualifying for the program. The efficient EF equaled the average market efficiency of units
qualifying for the program at the time of their rebate.
. Recent evaluation surveys conducted in the region estimated 245 washing cycles per year.t'7
The Cadmus Group, lnc. 2010. "Do the Savings Come Out in the Wash? A Large Scale Study of ln-Situ
Residential Laundry Systems."
http://www.cadmusgroup.com/pdfs/Do-the Savings-Come-Out-in-the-Wash.pdf
Ecotope lnc. 2012.2Otl Residential Building Stock Assessment: Single-Fomily Chorocteristics ond Energy Use.
Seattle, wA: Northwest Energy Efficiency Alliance.
http://www.enerSystar.gov/ialbusiness/bulk-purchasing/bpsavings-calc/
Ca lcu latorConsu merDishwasher.xls?7 L82-1(9z
Pacific Power Woshington 20092070 Residentiol Home Energy Savings Evoluotion,, January 2012.
Rocky Mountain Power 2009-2070 ldoho Residentiol Home Energy Sovings Evoluotion., February 2012.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 14 oi 45
. Water heating consumed 56% ofelectricity required to run a dishwasher connected to an
electric domestic hot water heater.8
Results ond Findings
Table 9 shows: total reported and qualified counts, savings, and realization rates of gas ENERGY STAR
Products measures in ldaho.
G Clothes Washer-Nat Gas H20
G Dishwasher-Nat Gas H20
Program Total
2,298
L92
2,490
383
149
s32
3,064
L92
t,255
2,298 100.0% 100.0%
L92 Lfi,O% 100.0%
2,490 100.016 1@.0%
75.Oy"
100.0%
76.5%
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 1 5 of 45
Appendix 1B addresses electricity savings achieved by the installation of ENERGY STAR products in
homes with a gas domestic hot water heater.
The program achieved a 76.5% realized adjusted gross savings rate, a result driven by the reduction in
assumed clothes washer cycles per year.
1.3.3 Heating and Cooling Efficiency
Program Description
The Heating and Cooling Efficiency program included the following gas measures:
. Gas Boiler
. Gas Furnace
The program offered a S4O0 direct financial incentive to motivate customers to use more energy-
efficient heating and cooling equipment. Participants could receive the incentive for installing a high-
efficiency natural gas furnace of 90% AFUE (heating efficiency) or greater, or a natural gas boiler of 90%
AFUE or greater.
Analysis
The PY 2010 gas impact evaluation report documented a census billing analysis Cadmus performed to
determine the change in energy consumption due to the installation of a high-efficiency gas furnace. As
8 htto://www.enerwstar.sovlialbusiness/bulk ourchasinr/bosavines calc/CalculatorConsumerDishwasher
.xls?7182-1c92
Table 9. ENERGY STAR Products Program Results
the billing analysis provided the best information on this measure, Cadmus continued tracking results
for the 2012 program year.'
We calculated energy savings achieved through installations of high-efficiency gas boilers by adjusting
the billing analysis results to the typical participant home installing a high-efficiency boiler.
Results and Findings
Table 10 shows total tracked and qualified counts, savings, and realization rates of gas Heating and
Cooling Efficiency measures in ldaho.
G Nat Gas Boiler
G Nat Gas Furnace
Program Total
72 L,Lr6 1,115L,9?s 105,s75 10s,s7sr,or7 16,691 16,691
99.2% 7@/o L,LO7 99.2%gg.2% tctr,Yo LO4,73o gg.2%
99.2% Lo,J./" 101837 99.2%
The program achieved a 99.2% rcalized adjusted gross savings rate, reduced slightly due to qualification.
1.3.4 Weatherization/Shell
Progrom Description
This program incented five categories of measures, available to residential electric and gas customers
with homes heated with fuel provided byAvista:
o lnsulation-Ceiling,/Attic
o lnsulation-Floor
. lnsulation-Wall
The program incented qualifying ceiling and attic insulation (both fitted/batt and blown-in), which
increased the R-value by 10 or more, at 50.25 per square foot of new insulation, and up to 50% of
installation costs. Homes qualified if they had existing attic insulation less than R-19.
The program incented floor and wall insulation (both fitted/batt and blown-in), which increased the
R-value by 10 or more, at 50.50 per square foot of new insulation, up to 50% of the installation cost.
Homes qualified if they had existing floor and/or wall insulation less than R-5.
Anolysis
The PY2011 gas impact evaluation report documented a census billing analysis Cadmus performed to
determine the change in energy consumption resulting from installation of weatherization and window
" Auista 2O7O Multi-Sector Gos lmpoct Evaluotion Report. August 2011.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 16 of45
Table 10. Heating and Cooling Efficiency Program Results
measures. As the billing analysis continued to provide the best information on this measure, results
were maintained for the 2012 program year.o
Table 11 shows total reported and qualified counts, savings, and realization rates of gas Weatherization
program measures,
t72
L72
LL,448
Ll,48
7L,444 99.2% 1m.0% 11,357 99.2%1L44E !D.Ti.6 100.0t6 tr,?57 $.216
1.3.5 Water Heater Efficiency
Progrom Description
The Water Heater Efficiency program includes the following gas measures:
o High-Efficiency 40-Gallon Water Heater
o High-Efficiency 50-Gallon Water Heater
Through this program, Avista offered a 550 incentive to residential customers installing eligible high-
efficiency water heaters. To qualify for the program, natural gas water heaters with tanks had to have a
0.60 EF or greater for a SGgallon tank, and a 0.52 EF or greater for a 40-gallon tank.
Analysis
Deemed unit energy savings remained consistent with those used in the 2011 program year, thus no
changes were necessary.
Results ond Findings
Table 12 shows total tracked and qualified counts, savings, and realization rates of gas Water Heater
Efficiency measures in ldaho.
'o Avisto 2077 Muttfsector Gas t mpoct Evoluotion Report. May 2OL2.
Exhibit No. 3
Case Nos. AW-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 17 ol 45
Table 11. Weatherization Program Results
Table 12. Water Heater Efficiency Program Results
G 40 Gallon Nat Gas Hot
Water
G 50 Gallon Nat Gas Hot
Water
Program Total
Home-Gas Only
Elec/Gas (Gas)
Program Total
7
45
52
61
404
465
62
407
468
62
407
468
7,423 700.0%
406 100.0%
L,829 t@.O%
99.2%
99.2%
99.2%
7,423 100.0%
406 100.0%
7.,829 100.0%
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 18 of45
99.2%
99.2%
99.2%
LOO.O%
100.0%
ldr.0%
1.3.6 ENERGY STAR Homes
Progrom Description
The ENERGY STAR Homes program offered incentives to builders constructing single-family or
multifamily homes complying with ENERGY STAR criteria (and verified as ENERGY STAR Homes). Avista
provided a 5900 incentive for homes that have Avista electric or electric and natural gas service for
space and water heating. Avista provided a 5650 incentive for homes that only have natural gas service
(both hot water and space heating had to be natural gas),
Analysis
The PY2011 gas impact evaluation report documented the simulation modeling Cadmus performed to
determine the energy savings achieved by these measures. As the simulation results continue to provide
accurate estimates of savings, results were maintained for the 2012 program year.11
Results and Findings
Table 13 shows total tracked and adjusted counts, savings, and realization rates for gas measures within
ENERGY STAR Homes. The electric and gas programs funded participating homes using both Avista
electric and gas. The associated electric impact evaluation report will address electric savings associated
with these homes.
700.0%
100.0%
100.0%
" Avisto 2077 Multi-sector Gos tmpoct Evoluotion Report. May 2012.
Table 13. ENERGY STAR Home Program Results
t.3.7 Residential Programs Confidence and Precision
Cadmus determined the overall precision of the adjusted gross savings by estimating the standard error
associated with each measure. For measures only based on deemed savings estimates, error in the
deemed savings resulted from error in each of the input assumptions.
Typically, the error for each savings estimate results from the sampling error associated with the
research into each savings equation input. To simplify this analysis, Cadmus conservatively estimated a
standard error associated with each deemed measure as20% of the unit energy savings, unless recent
evaluation research developed a more accurate estimate. Though a greater estimate than the values
Cadmus typically determines, this provided a conservative estimate of program precision.
Two programs used more accurate estimates of error, based on recent research. The standard error for
the Heating and Cooling efficiency program drew upon the billing analysis performed in 2011.12 The
standard error for the Weatherization/Shell program drew upon the billing analysis performed in 2Ot2,r3
Following determination of program measure savings-based error, Cadmus applied the verification error
determined through this year's surveys to each program, except for the two using billing analysis results.
We did not apply verification survey error to savings determined through a billing analysis as their
results included homes where installations were stated to have occurred, but did not occur. Table 14
shows the program level error and precision for the portfolio's residential portion. Overall, the
residential programs achieved 3.6% relative precision at the 90% confidence interval.
ENERGY STAR PToducts
Heating and Cooling Efficiency
Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
Total
9,547
335,775
50,369
3,L64
4,469
403,324
2,387
8,082
2,754
564
634
&905
4L.O/o
4.0%
9.0%
29.3/o
23.4%
t.6%
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page I 9 of 45
7.4 Conclusions
Overall, the 2012 residential gas programs in the state of ldaho produced 12L,978 therms in savings. As
shown in Table 15, the evaluation yielded a realization rate of 98.6%.
Avisto 2070 Multi-Sector Gos lmpoct Evoluotion Report August 2011.
Avisto 2077 Multi-Sector Gos lmpoct Evoluotion Repoft. May 2OL2.
t2
13
Table 14. Program Savings Precision at the 90% Confidence lnterval
Table 15. Program Reported and Verified Gross Verified Savings and Realization Rates-ldaho
ENERGY STAR Products
Heating and Cooling
Efficiency
Weatherization/Shell
Water Heater Efficiency
ENERGY STAR Homes
Total
3,255 2,490 100.0% 100.0%
99.2% LOO.OYo
99.2% L00.0%
99.2% LOO.O%
100.0% Loo.o%
99.2% 100.0%
2,490 76.5Yo
105,837 99.2Yo
rL,357 99.2%
465 99.2%
1,829 tOO.OYo
12t,978 98.6%
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 20 ot 45
106,591
tL,448
458
L,829
L23,693
106,591
Lt,448
458
L,829
t22,927
Table 16 shows the achievement rates for gross savings compared to the IRP goals for the residential
sector.
727,978
7.5 Recommenddtions
Cadmus offers the following recommendations, based on evaluation results:
r Avista should collect and record equipment efficiency information in the database tracking
system, or at least record the model numbers for appliances. lncluding equipment-specific
information addressing the actual efficiency of the equipment installed would allow greater
accuracy in estimating the gross energy savings achieved. Future evaluations could use collected
information to determine savings, rather than relying on regional market average estimates,
which do not account for the self-selection inherent in rebate programs.
o lf the Clothes Washer measure is reinstated, Avista should consider moving all rebates to the
electric program, as the majority of savings will likely result from a reduction in consumed
electricity from the dryer. Qualifying for the program should be based on the presence of an
electric dryer in the home. Given the large percentage of savings achieved through reduced
dryer energy, and because of the high likelihood that most participants have an electric dryer,
this measure predominantly produces electric energy savings.
1.5.1 Future Research Areas
The following research recommendations draw upon this impact evaluation's results and on known
future changes in program requirements:
o Perform a targeted billing analysis for weatherization participants using both electricity and gas
to heat their homes.
Table 16. Overall Evaluated Gas Savings and IRP Goals
o Perform a billing analysis for ENERGY STAR homes using a nonparticipant comparison group,
once enough homes have participated under the new requirements to justify the work.
Exhibit No.3
Case NoS. AVU-E-13 AVU-GI3
L. Hermanson, Avista
Schedule4, Page21 of 45
2 2012 Nonresidential Gas lmpact Report
2,T lntroduction
Avista's nonresidential portfolio of programs promotes the purchase of industry-proven, high-efficiency
equipment for its commercial customers. Avista provides rebates to partially offset cost differences
between high-efficiency equipment and standard equipment, reducing first-cost barriers and making
high-efficiency equipment a more viable option for commercial customers.
Six programs make up the nonresidential gas portfolio, divided into two major categories:
o Prescriptive (five programs)
o Site-Specific (one program)
z.L.L Prescriptive
Prescriptive Commerciol Clothes Washer (PCW)
To encourage customers to select high-efficiency clothes washers, this program targets nonresidential
electric and natural gas customers in multifamily or commercial Laundromat facilities. The program's
streamlined prescriptive approach, designed to reach customers quickly and effectively, promotes
ENERGY STAR or Consortium for Energy Efficiency (CEE) listed units.
Prescriptive Commerciol HVAC (PCH )
Beginning in January 2011, installations of efficient HVAC systems have been processed through a
prescriptive program rather than through the site-specific program. The prescriptive program limits
eligible measures to the following:
o Furnaces under 225 kBtu, with an efficiency greater than 90% AFUE.
o Furnaces between 225 kBtu and 300 kBtu, with an efficiencygreaterthan 85% AFUE.
Prescriptive Commercial Windows & lnsulation (PCS)
Beginning in January 2011, installation of commercial insulation has been processed through a
prescriptive program, in addition to the site-specific program. Projects qualify for the prescriptive
program if they meet the following, pre-existing qualities:
o Wall insulation levels of less than R4, improved to Rl1 or better.
. Attic insulation of less than R11, improved to R30 or better,
o Roof insulation of less than R11, improved to R30 or better.
Prescriptive Food Seruice Equipment (PFS)
Applicable to nonresidential electric and gas customers with commercial kitchens, this program provides
direct incentives to customers choosing high-efficiency kitchen equipment. To qualify for an incentive,
the equipment must meet ENERGY STAR or CEE tier levels (depending on the unit).
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule4, Page22ol45
Energy Smort Grocer (ESG)
Though refrigeration offers potentially high energy savings, the technical aspects of the equipment often
cause it to be overlooked. The Energy Smart Grocer program assists nonresidential grocery store
customers with the technical aspects of their refrigeration systems, while clearly presenting the savings
they can achieve. A field energy analyst provides customers with technical assistance, produces a
detailed report of potential energy savings at a facility, and guides customers through the process, from
inception through the payment of incentives for qualifying equipment.
2.1.2 Site-Specific
The site-specific program addresses nonresidential measures that do not fit the prescriptive
applications; thus, they must be considered based on their project-specific information. Measure eligible
for consideration must produce demonstrable kWh and/or therm savings, and are available to
commercial, industrial, or pumping customers: receiving electric or natural gas service from Avista; and
seeking to make cost-effective, energy-efficiency improvements to their businesses. The program
includes the following electric- and gas-saving measures:
o Site-specific HVAC (SSHVAC)
. HVAC combined
. HVAC heating
o Site-specific other (SSO)
' APP|iances
r Motors (demand controlled ventilation)
o Site-specific shell (SSS)
Avista designs, manages, and implements the prescriptive and site-specific programs. lt has also
developed algorithms it uses to calculate measure savings and to determine measure and customer
eligibility.
Avista staff fields inquiries from potential participants and contractors, and maintains a tracking
database for projects. Throughout the program, Avista manages projects by reviewing and approving
applications at all stages of the process, calculating project savings, and populating the database with
relevant information.
2.2 Methodology
Cadmus designed the impact evaluation to verify tracked program participation and to estimate energy
savings. We determined gross savings using: engineering calculations, desk reviews, verification site
visits, and some project-level billing analysis.
Cadmus reviewed Avista's tracked gross energy savings and available documentation, such as audit
reports and savings calculation work papers, for a sample of sites, particularly focusing on calculation
procedures and documentation for savings estimates. We also verified the appropriateness of Avista's
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 23 ot 45
analyses to calculate savings, and the operating and structural parameters of the analyses. Through site
visits or desk reviews of a sample of projects, we collected data and evaluated gross energy savings
through engineering calculations.
Cadmus collected baseline, tracking, and program implementation data through on-site interviews with
facility staff. During on-site visits, we verified measure installations and determined changes to the
operating parameters occurring since measure installation. Facility staff interviews included questions
regarding the installed systems' operating conditions, additional benefits, or shortcomings. We used the
savings realization rates from sample sites to estimate savings and to develop recommendations for
future studies.
2.2.1 Sampling
Avista reported planning to phase out the gas programs due to cost-effectiveness concerns associated
with the declining price of natural gas in 2011. Consequently, Cadmus and Avista found it appropriate to
apply a lower rigor level for sampling than that used in the 2010 and 2011 evaluations. Cadmus selected
a precision target of 80% confidence and a ZOYo confidence interval for the 2012 program sample. We
developed a sampling calculation tool to estimate the number of site verifications and desk reviews
required to achieve the precision target's rigor levels.
Using program population data provided by Avista, we determined 43 sites would require evaluations
across Washington's and ldaho's program populations for both years. Cadmus will calculate the
combined 2OL2 and 2013 evaluation precision following the 2013 program evaluation.
Table 17 shows the proposed precision targets for the site verification and desk review evaluation
activities.
Table 17. Proposed PY 2OL2-2OL3 Nonresidential ldaho and Washington Gas Evaluation Sample
Prescriptive
SSHVAC
sso
sss
Tota!
We assigned a census and a random sample for each stratum, The census stratum represented the six
projects with the highest overall gas savings, with one of the six census sites located in ldaho. Each
census site reported over 10,000 therms in savings and combined to represent 24yo of tolal program
reported savings. For the non-census stratum, we randomly selected additional participants from the
remaining project population.
Cadmus found the database extract from Avista provided program-level but not measure-level
information (e.9., boilers, chillers, LED lighting fixtures). Therefore, we sought to verify savings for every
incented measure at each site, regardless of whether it achieved gas or electric savings. Establishing
24
7
8
4
43
80/20
80/20
80/20
80/20
80l20
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page24 of 45
whether we evaluated an accurate distribution of specific measure types within each program would
have required an exhaustive review of project files, which fell outside of the evaluation's scope.
2.2.2 Data Collection
Cadmus collected data from one on-site verification in ldaho and conducted 10 desk reviews. For each,
we first conducted a document review to determine measure types, quantities, operational parameters,
and calculation methodologies.
Document Review
Avista provided Cadmus with documentation on the sample sites'energy-efficiency projects, including:
program forms, the tracking database, audit reports, and savings calculation work papers for each
rebated measure. Our review of calculation spreadsheets and energy simulation models paid particular
attention to calculation procedures and documentation for savings estimates.
Cadmus reviewed each application for the following information:
. Equipment replaced: descriptions, schematics, performance data, and other supporting
information.
o New equipment installed: descriptions, schematics, performance data, and other supporting
information.
r Savings calculation methodology: the methodology type used, specifications of assumptions,
sources for these specifications, and correctness of calculations.
Site Visits
Cadmus performed on-site visits to verify the three primary tasks that follow:
1. Verifying the implementation status of all measures for which customers received incentives.
This required verifying energy-efficiency measures had been installed correctly and functioned
properly. We also verified the operational characteristics of the installed equipment, such as
temperature set points and operating hours.
2. Collecting the physical data, such as boiler capacities or operational temperatures, and analyzing
the energy savings realized from the installed improvements and measures.
3. Conducting interviews with facility personnel to obtain additional information regarding the
installed system, thus supplementing data from other sources.
Desk Reviews
For some prescriptive and site specific projects, we analyzed and evaluated energy savings by reviewing
calculation spreadsheets and documentation submitted with the rebate application. These 10 projects
experienced smaller therm savings compared to census-level projects we selected for site visits. For the
analysis, Cadmus verified the equipment efficiency, based on equipment model numbers provided in the
rebate applications and the savings calculation methodology.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 25 ol 45
2.2.3 Engineering Analysis
Nonresidential prescriptive and site-specific programs required significantly different methods
of analysis.
Overuiew
Procedures used for verifying savings through an engineering analysis depended on the type of measure
being analyzed. This evaluation used the following analytical methods, with descriptions included in
their respective sections:
o Prescriptive deemed savings
e Billing analysis
r Calculationspreadsheets
. Energy simulation modeling
Pre scri ptive Dee me d Savi n gs
For most prescriptive measures, we verified the deemed savings estimates that Avista used for savings
calculations, and then compared these with the values we developed for the TRM. We focused our
verification activities on:
o The installed quantity;
o Equipment nameplate data;
r Proper installation of equipment; and
o Operating hours.
Where appropriate, we used data from site verification visits to reanalyze prescriptive measure savings
with Avista's Microsoft Excel calculation tools, ENERGY STAR calculation tools, RTF deemed savings, and
other secondary sources.
Billing Analysis
Cadmus analyzed Avista's metered billing data for two site-specific HVAC projects. Using a pre- and post-
modeling approach, we developed retrofit savings estimates for each site. This modeling approach
accounted for differences in heating degree days (HDDs), and determined savings based on normalized
weather conditions, as actual weather conditions may have been milder or more extreme than the
TMY3's (typical meteorological year) 15-year normal weather averages from 1991-2005, obtained from
the National Oceanic and Atmospheric Administration (NOM).
NOAA also provided daily weather data for each weather station associated with the participant
projects, and we calculated the base 65 reference temperature HDDs. We matched participant billing
data to the nearest weather station by ZIP code, and matched each monthly billing period to the
associated base 65 HDDs.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 26 ol 45
ln developing the analysis models, we followed a modified PRISM approach, which normallzed all
dependent and independent variables for the days in each billing period, and allowed model coefficients
to be interpreted as average daily values. This methodology accounted for differences in the length of
billing periods. For each project, we modeled average daily consumption in kWh as a function of some
combination of the average standing base load, HDD, and (where appropriate) daily consumption.
For each site, Cadmus estimated two demand models: one for the pre-period; and one for the post-
period. We chose this methodology over a single standard treatment effects model to account for
structural changes in demand that might occur due to retrofits.
After estimating model coefficients for each site, Cadmus calculated three scenarios:
o We estimated a reference load for the previous 12 billing cycles, using the pre-period model.
This scenario extrapolated the counterfactual consumption (i.e., what the consumption would
have been in the program's absence). We calculated energy savings as the difference between
the counterfactual scenario and the actual consumption.
o We estimated two normalized scenarios: one using the pre-model; and one using the post-
model, Both scenarios used 15-year TMY3 data as the annual HDD and mean annual values for
the usage data. The difference between these two scenarios represented the long-term
expected annual savings.
Calculation Spreadsheets
Avista developed calculation spreadsheets to analyze energy savings for a variety of measures, including
the construction of envelope measures (such as ceiling and wall insulation). The calculation
spreadsheets required entering relevant parameters, such as square footage, efficiency values, HVAC
system details, and location details. From these data, energy savings could be estimated using
algorithms programmed by Avista. For each spreadsheet, we reviewed input requirements and output
estimates, and determined if the approach proved reasonable.
Energy Si mulation Modeling
Avista determined savings for many site-specific HVAC and shell projects using energy simulation
modeling (chosen due to the complex interactions between heating and cooling loads and the building
envelope). Avista provided the original energy simulation models, which we reviewed to determine the
relevant parameters and operating details (such as temperature set points) for the applicable measures.
We updated the models as necessary based on our on-site verification data.
2.3 Results and Findings
2.3.1 Overview
Cadmus adjusted gross savings estimates based on our evaluated findings. The following sections discuss
further details, by program.
Exhibit No. 3
Case Nos. AW-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 27 of 45
2.3.2 Prescriptive Programs
We evaluated savings for a sample of sites across five prescriptive programs. Table 18 shows the savings
and realization rates by proBram. Further evaluation details for each program follow. Table 19 shows the
combined ldaho and Washington prescriptive results. These results were used for final extrapolation
because the sample was chosen from a combined sampling methodology.
Table 18. Evaluated Results for PY2012 Nonresidential Gas Prescriptive Sample-ldaho
0
24
21
8
1
54
I
2
55
90
25
184
0
2
2
0
1
5
N/A
598
95
N/A
900
1,594
N/A
670
153
N/A
1053
1,876
1,053
N/A
2,304
L,728
4,677
9,762
N/A
LL2%
159%
N/A
L77%
Lt8r6
L77%
N/A
t04%
t@%
91%
9A16
Exhibit No. 3
Case Nos. AVU-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 28 of 45
Table 19. Evaluated Results for PY2012 Nonresidential Gas Prescriptive
ESG
PCW
PCH
PCS
PFS
Total
1
0
6
8
2
t7
900
N/A
2,224
L,736
5,135
9,9!r5
Cadmus identified several discrepancies between the rebate application information and inputs used in
Avista's savings calculations. The calculations often relied on reported equipment and operations data,
which could vary from parameters identified during on-site verification visits and metering.
Our adjustments increased savings by 18% for ldaho projects. The combined adjustments reduced
savings by 2Yo.Typical adjustments corrected equipment efficiencies, fuel types, operating schedules,
and operating parameters, as described below:
o For one prescriptive boiler replacement project, Cadmus found the proposed efficiency used in
the Avista savings algorithm was lower than the installed unit (90% versus 93%). By adjusting
the efficiency, the realization rate for that project increase d to 12L%.
. For one prescriptive window replacement project, the proposed solar heat gain coefficient
(SHGC) used in the savings calculation did not match the actual SHGC. The revised SHGC resulted
in higher gas savings (186% realization rate), but decreased electric savines (38%).
Sample-Combined Washlngton and ldaho
. One prescriptive EnergySmart project installed doors on medium-temperature refrigerated
display cases in a store. Cadmus used an industry standard tool to calculate savings based on the
linear feet of case retrofitted with doors. Avista savings calculations are hardcoded in the
spreadsheet and do not reference any savings algorithm. By using the Cadmus algorithm, the
realization project's rate increased to Ll7%.
2.3.3 Site-Specific
Cadmus evaluated the savings for six site-specific program projects, which represented a variety of
measure types. We evaluated the projects through on-site verification and desk reviews. We also
calculated an overall realization rate for all randomly selected (non-census) projects in ldaho, and then
applied the resulting realization rate to the non-census savings for each state and major measure type.
Table 20 shows our evaluated results for the program. Table 21 shows the combined ldaho and
Washington site-specific results. These results were used for final extrapolation because the sample was
chosen from a combined sampling methodology.
Table 20. Evaluated Results lor PY2OL2 Nonresidential Gas Site Specific Sample-ldaho
SSHVAC_
Census
SSHVAC
SSO
sss
Total
7
9
7
6
23
L
3
1
7
6
6
7
8
3
24
L8,267
10,535
9
7,344
36,155
95,999
24,950
8,363
26,673
155,985
3,195
L7,749
6
7,344
22,295
77,298
26,504
I,L87
26,818
138,807
t7%
Lt2%
69Y,
700%
62%
80%
106%
98%
liLlo
88%
Table 21. Evaluated Results lor PY2OL2 Nonresidential Gas Site-Specific Sample-Combined
SSHVAC_
Census
SSHVAC
sso
SSS
Total
6
35
33
26
100
Cadmus identified several adjustments to tracked savings from site-specific program project. Site-
specific projects tend to be more complex, making energy-savings parameters and impacts more
difficult to estimate. ln addition, the calculations often rely on participant-supplied building, equipment,
and operations data, which may vary from parameters identified during an on-site verification visit.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 29 ol 45
Washington and ldaho
ln aggregate, the site-specific program performed fairly well, achieving an overall combined realization
rate of 88%. The only census project in ldaho did not achieve savings. Though this reduced the overall
realization rate for ldaho projects significantly, higher-than-tracked savings for Washington site-specific
projects offset the project's losses. We made the following specific adjustments, based on our review of
rebate application and billing data:
o The ldaho Site-Specific HVAC census-level project (Table 20) retrofitted existing lighting and
installed a digital direct control system on the facility's HVAC system. During the on-site
verification, Cadmus verified the lighting retrofit had been installed as reported, but found
several discrepancies in the HVAC project's implementation. Cadmus's findings on the HVAC
system upgrade include the following:
. Nighttime temperature setback: The majority of the gas savings from this project derived
from implementation of a nighttime temperature setback strategy, but we found this
strategy had not been implemented. The building houses several laboratories, with material
testing stations and hood vents in every space, and sets its temperature to 72'F during
occupied and unoccupied hours due to controlled environment requirements and to
maintain the comfort of its occupants.
. HVAC system commissioning: Cadmus found the primary contractor performed ineffective
commissioning on the HVAC systern. Consequently, the building still experiences major air
balancing issues. The participant currently is working with a new contractor to re-
commission the HVAC system to resolve the balancing issues. The new contractor
conducted a thorough investigation of the HVAC system issues, reporting to Cadmus that
the building currently runs at a slightly negative pressure, an indication of poor balance
between supply air and exhaust. However, the new contracto/s work is not yet complete,
and these results did not factor into Cadmus' analyses.
r Billint analysis: Cadmus performed a linear regression with pre- and post-installation utility
billing data to determine the savings level for this project, as shown in Figure 2 and Figure 3.
This analysis confirmed the project realized a lower level of savings than was reported. The
slope for the regression equation represents the heating-dependent load, which is nearly
identical for the pre- and post-installation period. We noted the facility received a lighting
retrofit, which reduced the waste heat from inefficient lighting. This increased the heating
load for the facility, which is one reason why the post-installation regression analysis is
larger than expected. Cadmus calculated the additional heating load required as a result of
the lighting retrofit using values determined by the RTF. We added that value to the
difference between pre and post-installation linear regression to determine the evaluated
energy savings for the project, The project achieved a 17% realization rate.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 30 of45
Figure 2. Pre-lnstallation Linear Regression for Census-Level Project
Pre-lnstallation Therms vs. HDD
350
300
250
200
150
100
50
0
Y =7.256Lx+8.0377
R2 - 0.9911
O Therm/Day
-
Linear (Therm/Day)
Figure 3. Post-lnstallation Linear Regression for Census-Level Project
Post-lnstallation Therms vs. HDD
350
300
250
200
150
to0
50
0
y =7.2273x+20.491
R2 = 0.9913
O Therm/Day
-
Linear (Therm/Day)
As a result, Cadmus did not award gas savings to the project, which accounted for 40% of the tracked
savings for the site-specific HVAC program in ldaho.
Cadmus also revised gross energy savings for residential clothes washers installed at multi-family
facilities as a Site-Specific Other measure, as follows:
. Cycle: ln the previous evaluation, the washing cycles per year (377) were derived from Pacific
Power and Rocky Mountain Power Home Energy Savings participant surveys. Recent
independent evaluation surveys from the Residential Building Stock Assessment (RBSA) and
2012 Avista Participant surveys estimated 252 washing cycles per year. Unit energy savings
values have been adjusted accordingly, as reflected in this measure's realization rate.
. Consumption: Cadmus utilized the California metering study to estimate consumption per wash
and dry cycle for the base and efficient equipment.
Exhibit No. 3
Case Nos. AW-E-13 AVU-GI3
L. Hermanson, Avista
Schedule 4, Page 31 of45
. A69% realization rate resulted for one ldaho clothes washer project. This was the only Site-
Specific Other measure that Cadmus evaluated for ldaho. Cadmus evaluated the overall Site-
Specific realization rate based on the combined ldaho and Washington sample.
2.1.4 Extrapolation to Program Population
ln evaluating the nonresidential gas programs, we selected sites that could provide the most significant
impacts. We designed the site visits to achieve a statistically valid sample for the major strata, as
discussed. For measures in the random (non-census) sample, we calculated realization rates (the ratio of
tracked-to-evaluated savings) to apply to the programs at the remaining non-sampled sites. These
realization rates were weighted averages, based on the random verification sample, and using the
following four equations:
Evaluated..
RR,, =;j; for measure j at site i (1)- I racKeQ il
lEvaluated,Mi =fi**"a, ; for measure i across all sample sites
lEvaluatedo = RRixZTrackedr; for measure j across all sites inmeasure population (3)
kk
lEvaluatedrM, =fr**"a r ; for the population(all sites and measures) (4)
Where:
RR = the realization rate
i = the sample site
j = the measure type
k = the total population for measure type J'
| = the total program population
We calculated realization rates for each individual site in the sample based on measure type (1). We
then calculated the realization rates for the measure types using the ratio of the sum of evaluated
savings to the sum of tracked savings from the randomly selected sample for each measure type (2). We
calculated the non-census population evaluated savings by multiplying the measure type realization rate
(RR;) from the random sample by the tracked savings for the non-census population of each measure
(2)
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 32 of 45
type (3). We then added the tracked and evaluated savings from census stratum measures to calculate
the total tracked and evaluated savings for each program. The program realization rate derived from the
ratio of all evaluated to all tracked savings (4).
Table 22 summarizes the results for all prescriptive and site-specific programs in ldaho. The state
realized an 87% overall portfolio gross realization rate. Notably, during extrapolation ofgas savings to
the total gas measure population, the census-level site-specific HVAC project's realization rate was
excluded because it was not part of the random sample.
Prescriptive
SSHVAC
sso
sss
Total
32,6L5
47,951
2,499
73,387
96,452
31,852
3s,923
2,457
13,504
8t,729
98%
75%
98%
701%
a7%
2.3.5 Fuel Conversion and HVAC / tighting lnteractive lmpacts
The Avista natural gas portfolio reported savings do not include increases in gas consumption due to fuel
conversions from electric heating to gas heating or from increased lighting efficiency. Lighting systems
convert a large portion of their input energy to useful light output, but a substantial portion also
converts to heat. Any reduction in lighting input energy also reduces waste heat. Reducin8 waste heat
lowers the site's required cooling load, but increases the site's heating load.
Cadmus noted that Avista tracked and recorded these gas consumption effects for many projects to
determine electric program cost-effectiveness. Most tracked interactive effects involved prescriptive or
site-specific lighting projects, although some therm penalties resulted from the Energy Smart Grocer (in
Avista's electric portfolio) and site-specific HVAC program projects.
ln addition, Avista did not factor interactive effects into its portfolio energy-savings goals (which would
have reduced goals).
2.4 Conclusions
Cadmus evaluated 77 of 77 measures installed through the ldaho program, representing 39% of
reported savings.
The evaluation determined that Avista generally implemented the programs well. Cadmus identified the
following key issues that reduced evaluated energy savings below the reported values:
. Programs sometimes provided incentives for measures that may not have been appropriate,
such as installing night-time temperature setbacks for a laboratory with consistent temperature
requirements.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 33 of 45
Table 22. PY zOtZ Gas Gross Program Realization Rates - ldaho
. Post-installation inspection process may not have always identified operational issues. An
example is the Site-Specific HVAC census project, for which Avista staff verified the lighting
measure but performed only cursory review of the HVAC measure.
2.5 Recommendotions
Cadmus offers the following recommendations, based on evaluation results:
o Review whether reported HVAC measures are appropriate for facilities with consistent space
conditioning requirements, such as laboratories.
o Consider focusing post-installation inspections on the projects with the highest tracked
energy savings.
Exhibit No.3
Case Nos. AW-E-13 AVU-C-I3
L. Hermanson, Avista
Schedule 4, Page 34 of45
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 35 of 45
3 2012 Low lncome Gas lmpact Report
3.T lntroduction
ln 2010, Cadmus conducted a statistical billing analysis, determining adjusted gross savings and
realization rates for energy-efficient measures installed through Avista's Low lncome Weatherization
Program. We performed analysis and calculated savings at the household or participant level, rather
than the measure level.
This report:
. Applies these 2010 billing analysis savings estimates to the 2012 participant population; and
o Reports total gas impacts associated with the 2072 programyear.
Cadmus anticipates collecting a full year of post-period consumption data to perform a billing analysis of
the 2012 participant population. ln the interim, this evaluation report extrapolated results from the
recent 2010 gas impact analysis to 2Ot2 participants, The new billing analyses will take place in the first
quarter of 2OL4.
To estimate 2010 energy savings resulting from the program, Cadmus used a pre- and post-installation,
combined CSA and PRISM approach that utilized monthly billing data. We analyzed savings estimates for
ldaho and Washington, and ran a series of diagnostics (such as a review of savings by pre-consumption
usage quartile), and outlier analysis. Avista's 2010 Gos lmpoct Report presents a detailed discussion of
the regression model and methodology used for this analysis.
3.1.1 Program Description
Five programs, listed in Table 23, make up Avista's Low lncome Weatherization Program. Local
Community Action Partners (CAPs), within Avista's ldaho and Washington service territories, implement
these low income programs. CAPs holistically evaluate homes for energy-efficiency measure
applicability, combining funding from different programs to apply appropriate measures to a home,
based on results of a home energy audit.
Table 23 also describes the measures installed under each program component, along with counts of gas
measures installed in PY 2012 and included in our gas impact analysis (a separate report contains
findings on evaluated electric measures).
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 36 of 45
Table 23. 2012 Gas Efficiency lnstallations by Program Component
Shell/Weatherization
HVAC Efficiency
Hot Water Efficiency
Fuel Conversion*
lnsulation, window/door installation, air infi ltration,
programmable thermostat .
High-efficiency gas furnace replacement
High-efficiency water heater replacement
Electric furnace, heat pump, or water heater
replacement with gas units
240
L7
0
N/A
N/A
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 37 ol 45
ENERGYSTARAppliance High-efficiencyrefrigeratorreplacement
*The Avista portfolio considers (and reports) fuel conversion measures as electric-saving measures.
3.L.2 Data Collection
Cadmus primarily drew impact evaluation data from the program participant database. Avista provided
information regarding program participants and installed measures for ldaho. Specifically, these data
included:
. Lists of measures installed per home; and
o Expected savings from each completed measure installation.
The data, however, did not include the quantity of measures installed (such as the square footage of
installed insulation) or per-unit savings estimates.
Starting in 2012, Avista incorporated TRM savings estimates that Cadmus developed specific to Avista's
low income customer segment. These measure-specific savings estimates incorporated data from
regional and secondary research (e.9., RTF, U.S. Department of Energy [DOE]) as well as input
assumptions derived from analysis of low income weatherization program participant consumption
(e.g., pre-period heating consumption).
3.2 Methodology
3.2.1 Sampling
ln applying the 2010 gas billing analysis results, we used a census of program participants, comprised of
81 gas accounts, but excluding the 16 gas participants receiving conversion measures.
3.2.2 Data Collection Activities
Documentotion Review/Databose Review
Cadmus used the 2012 ldaho and Washington program participant database, provided by Avista, to
develop a complete population for applying the 2010 billing analysis results. Participant data included:
. Customerinformation;
. Account numbers;
. Types of measure installed;
o Rebate amounts;
o Measure installation costs;
. Measure installation dates; and
o TRM savings per measure.
Billing Analysis-CSA Modeling Approoch
To estimate energy savings from this program, we used a pre-post CSA fixed-effects modeling method,
which utilized pooled monthly time-series (panel) billing data.
The fixed-effects modeling approach corrected for differences between pre- and post-installation
weather conditions as well as for differences in usage consumption between participants, and included
a separate intercept for each participant. Our modeling approach ensured model savings estimates
would not be skewed by unusually high-usage or low-usage participants. Monthly consumption was also
paired between pre- and post-months to maintain the same time frame for evaluating unique
participants.
Additional details regarding the 2010 billing analysis can be found in the Avr,3to 2070 Gos lmpoct Report.
3.2.3 Estimating Conversion Participant Savings
The evaluation team used a similar approach for calculating gas savings for ldaho conversion
participants as used in the 2011 evaluation report. This approach assigned savings to conversion
participants (n = 15), based on three distinct customer categories:
1, Full model savints (123 therms), assigned to participants (n = 1) receiving three or more distinct
gas-saving measures (including a high efficiency furnace).
2. Partial model savings (61 therms), specific to participants that installed a high-efficiency gas
furnace in place of a standard efficiency electric furnace. to These participants received the high-
efficiency furnace replacement and no more than one additional gas-saving measure (n = 13).
For participants in this group with one additional gas-savings measure, we passed through the
TRM savings associated with the non-furnace measures.
3. No mode! savings, for customers receiving at most one gas-saving measure (n = 2) and not a
high-efficiency furnace. For these customers, we passed through TRM savings if they received a
gas-savings measure.
'o The program participant database did not indicate that water heater conversions were replaced with
efficient units; therefore, no additional gas savings were applied.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 38 of 45
81
To account for gas savings experienced through high-efficiency furnace replacements, we used savings
calculated through the 2010 evaluation of Avista's residential furnace replacement program (84 therms),
scaling this value to reflect low income participant home square footage, which resulted in 61 therms.ls
3,3 Results and Findings
3.3.1 Overall Program Results
N o n -Conve rsion Po rticipa nt Resu fts
Applying savings estimates from the billing analysis to the gas-saving participant program population
produced total savings of 123 therms per participant. we applied these modeled savings to gas-savings
participants not receiving conversion measures, and we calculated average reported TRM savings by
summing measure savings at each household, then taking the mean household savings across individual
participants. Table 24 provides a comparison between average participant savings TRM and modeled
savings for non-conversion customers.
129%9,963
Table 25 shows the count of 2012 gas-saving measure installations (including both non-conversion and
conversion participants). Air infiltration has the highest distribution of installations, followed by attic and
duct insulation.
Attic insulation
Doors
Duct insulation
Floor insulation
H igh-efficiency furnace replacement
H igh-effi ciency water heater replacement
lnfiltration controls
Thermostat (AC)
Thermostat (No AC)
Wall insulation
Windows
Low income participants averaged 1,250 square feet per home, while single-family participants averaged
1,728 square feet per home.
26
8
57
40
103
N/A
31
t4
t4
35
22
42
16
31
73
L7
N/A
72
1
7
2r
27
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 39 of 45
Table 24. Non-Conversion Gas Savings
Table 25. Averate Reported Savings and lnstallation Count by Measure
To highlight some distinctions in Avista's reported savings, we compared average expected measure
savings from 2011 to the 2012 TRM estimates. Figure 4 highlights differences between average savings.
Figure 4. Comparison of 2011 and 2012 Average Reported Savings by Measure
r 2011 r 2012 (TRM)
Attic lnsulation
Doors
Duct lnsulation
Floor lnsulation
lnfiltration controls
Wall lnsulation
windows
0 50 100 150 200 250 300
Reported Therm Savings
Savings reported in 2012 using TRM estimates were lower for a number of measures than 2011 average
savings, most notably for infiltration controls, doors, and insulation measures, Generally, the two years
offered a relatively similar mix of measure installations, with infiltration controls and insulation the most
frequently installed measures for gas-saving participants.
Co nve rsio n Po rticipa nt Resu lts
Of the 97 total ldaho gas-savinSs participants, 16 received electric-to-gas conversion measures, including
electric-to-gas furnace and water heater replacements. This analysis considered these participants
separately, as the methodology for estimating evaluated savings differed slightly from the non-
conversion participant group. Table 25 provides a distribution ofall Avista-funded measure installations
for conversion participants.
Table 26. Measure lnstallations for Conversion Participants
Electric-Saving Conversion Measures Electric-to-gas furnace replacement
Electric-to-gas water heater replacement
Doors
Duct insulation
High-effi ciency furnace replacement
lnfiltration controls
Thermostat (No AC)
Thermostat (AC)
Windows
L4
L4
1
1
t4
L
7
1
1
Gas-Saving Measures
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 40 of 45
Table 27, Conversion Particlpant Gas Savings - ldaho
Ofthe 14 participants receiving a gas furnace conversion, all had high-efficiency gas furnaces installed,
and none ofthe 14 water heater conversion participants received high-efficiency gas water heaters.
ln total, we estimated an additional 1,096 therms savings for gas conversion participants, as shown in
Table 27.
Full model savings
Partial savings (high-
efficiency furnace)'
No model savingst
Total
1
13
2
16
L23
61
N/A
L23
906
67
1,(Xr5
Exhibit No.3
Case Nos. AVU-E-I3 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 41 of 45
*Total evaluated savings may include instances of pass-through TRM measurelevel savings.
A net increase in therm usage occurred for all conversion customers. However, based on Avista's
approach to correcting for these impacts through its cost-effectiveness analysis, this report calculated
therm savings associated with the following:
1. lnstallation of gas-savings weatherization measure bundles.
2. Furnace conversion replacements, using high-efficiency gas equipment, compared to standard
gas equipment.16
Overoll Porticipant Results
Table 28 provides overall gas savings, including savings attributed to fuel conversion participants
receiving gas-savi ng measures.
3.3.2 Goals Comparison
Cadmus compared evaluated savings for the 97 ldaho gas participants against Avista's IRP goals. Table
29 summarizes: overall evaluated savings, IRP savings goals, and achievement rates. ln all, the low
income weatherization program achieved approximately 44% of its gas savings goals.
lt Electric savings associated with conversion measure installations will be addressed in the 207r2071 Avisto
Electic lmpoct Report.
Table 28. Overall Gas Savings
Table 29. lRP Protram Goals Comparison
25,212 11,059
*lncludes 81 participants receiving model savings and 16 conversion customers.
3.4 Conclusions
Upon comparing 2011 and2OL2 results, changes in Avista's expected savings calculations led to
differences in realization rates. Average reported gas savings per (non-conversion) participant decreased
by 62% between the years, falling from 305 therms in 20!L to 115 therms in 2012 (based on the TRM).
This appears to primarily drive shifting realization rates, from 4t% for ldaho in 2011 to tt8% in 20L2,
As shown in Figure 4 (above), except windows, all measure-level estimates observed significant changes
in therm savings between 2011 reporting and the 2012 TRM estimates, with these decreases in average
savings ranging between 3 to 10 times the previously reported estimates, most notably for infiltration
and insulation measures.
3.5 Recommendotions
The following section outlines our suggestions for enhancements to help improve program
impact results.
. Use a contro! or comparison Sroup in future billing analyses. For upcoming impact evaluations
that employ billing analysis, we suggest using 2013-2014 participants as a control group to
analyze the treatment group of 2012 participants. For such analysis, 2011 and 2013 annual
participant consumption histories would be used as the pre- and post-periods, Using a control or
comparison group of nonparticipants allows analysis to control for exogenous factors (e.g.,
macroeconomic, rate changes, technological trends) that may result in trends affecting
consumption. Controlling for these trends using a control/comparison group reflects a more
robust experimental design and defensible methodology for estimating accurate energy-savings
impacts.
o lnclude high-use customens in program targetint. While prioritization guidelines for targeting
low income weatherization participants are set at the federal level, some utilities, for targeting
purposes, actively track customer usage and provide agencies with lists of customers that
experience particularly high energy consumption. ln fact, DOE protocols list high-energy
consumption as a factor allowed in participant prioritization. ln such cases, along with other
targeting criteria (e.9., families with children, senior citizens), agencies may incorporate energy-
consumption characteristics into their program participant prioritization. Not only would
weatherizing high-use customers likely result in higher energy savings, but some customers may
be overly burdened with energy bills due to their housings' characteristics, and the program
could provide some financial relief.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule4, Page42ot45
Methods exist for identifying high-usage customers while controlling for factors contributing to
consumption (e.g., square footage, income, number of people per household). Using such an
approach would allow Avista to identify high-use customers.
Given reductions in federal funding for weatherization and associated reduced agency capacities
resulting in more limited leveraging opportunities, Avista has an opportunity to lead new efforts
for continued delivery of energy-savings resources to low income residential customers. By
considering high-usage targetinB, potential exists to secure cost-effective energy savings
through one segment of this population, while continuing to support weatherization for income-
qualified customers, which may result in lower savings and prove less cost-effective. Efficient
targeting can aid in balancing these efforts to provide whole-house weatherization, while
continuing to leverage the agency network as a resource for outreach and delivery.
Track and compile additional data from atency audits. These data include information on
primary and secondary heating and cooling and on the size of a home. As an inexpensive
alternative to gas heat, gas customers may turn to electric room heaters and wood stoves,
thereby reducing impacts of weather-sensitive measures installed through weatherization (e.9.,
insulation). Collecting information on customers' primary heating usage at the time of
weatherization will provide more reasonable savings estimates.
We recommend working with agencies to develop explicit, on-site tracking protocols for
collecting information on participant heating sources. Agencies should collect the following
information to better inform heating (and cooling) sources:
. Visual inspections of all heating equipment found on site;
r Participant-reported primary and supplemental heating sources used;
. Quantities of secondary heating, if applicable (e.g., numbers of electric room
heaters); and
. Any indicators suggesting discrepancies between actual and reported primary heating.
Consider performing quantatative, non-energy benefit analyses. With respect to ongoing
Advisory Group discussions surrounding quantifying non-energy benefits, we recommend Avista
consider pursuing additional analyses, aimed at quantifying non-energy benefits associated with
low income weatherization and applicable to the TRC test. ln particular, analyses of economic
impacts and payment pattern improvements (including reduced arrearages and collections
costs) can provide program stakeholders with monetized values of benefits. Other utilities have
used such analyses in reporting low income weatherization cost-effectiveness in the northwest
(e.9., ldaho, Washington). Standard cost-effectiveness testing, using TRC test accounts for all
program costs (only including energy savings as program benefits), clearly omits some genuine
non-energy benefits experienced by participants (as discussed in greater detail in the 2070
Process Evoluotionl.
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-13
L. Hermanson, Avista
Schedule 4, Page 43 of45
Appendlx 1A: Residential ENERGY STAR Home Model lnputs
The following table summarizes the inputs used to simulate homes in ldaho.
Table 1A. ENERGY STAR, and ldaho Construction Standards for New Homes
Insulation
Ceiling R-38 R-38
Wall R-19 R-19
Floors Orer
Unconditioned Space R-30 R-30
Slab Floorc R-10 R-10
Windovw & Doors
Windows 0.3s 0.35
Max GlazinE Area 0.21 Set to EIERGY STAR standards
Doors R-5 set to ENERGY STAR standards
Ducts
lnsulation R-8 R-8
Sealing Mastic onlv Taoes allowed
Max Leakage <0.06 CFM/sq. ft. or 75 CFM
total @sOPa Set to ENERGY STAR standards
Ventilation & Alr
Sealinc
Ventilation System Exhaust ventilation Exhaust ventilatlon
Envelope Tightness 0.35 normal ACH 0.35 normal ACH
Heating & Cooling
Eouioment
Gas Furnace 90AFUE SOAFUE
Air Condttioner SEER 13 SEER 13
Extrlblt No. 3
Case Nos. AW-E-13 AVU-G'I 3
L. Hermanson, Avista
Scfiedule4, Pago44of 45
Appendix 18: Electricity Savings Achieved by Residentia! Gas Programs
The following table shows the electricity saved in kWh by the 2012 gas energy efficiency programs. The
believed high penetration of electric dryers in homes with gas domestic hot water heating is the reason
for the significant savings achieved. The electricity saved through the installation of an efficient
dishwasher is associated with the machine operation, not water savings. The 2010 gas furnace billing
analysis showed that a portion of participants are choosing to install an air source heat pump at the
same time they install a new high efficiency furnace. This switch from all gas heating to dual fuel heating
results in an electric penalty.
The values shown in the table are for all measure installations in ldaho, both inside and outside Avista's
electric service territory,
G Clothes Washer-Nat Gas H20
G Dishwasher-Nat Gas H20
G Nat Gas Furnace
Total
383
L49
\,025
L,557
85,409
4,035
-169,258
-79,8L4
Exhibit No. 3
Case Nos. AVU-E-13 AVU-G-I3
L. Hermanson, Avista
Schedule 4, Page 45 ot 45
Table 18. Electricity Savings for Gas Program in tdaho