Chapter 2: Commercial and Industrial Lighting Evaluation Protocol

Chapter 2: Commercial and
Industrial Lighting Evaluation

Protocol
The Uniform Methods Project: Methods for
Determining Energy Efficiency Savings for
Specific Measures
Created as part of subcontract with period of performance
September 2011 – September 2016
This version supersedes the version originally published in April

2013. The content in this version has been updated.
Dakers Gowans
Left Fork Energy
Harrison, New York
Chad Telarico
DNV GL
Mahwah, New Jersey
NREL Technical Monitor: Charles Kurnik
NREL is a national laboratory of the U.S. Department of Energy

Office of Energy Efficiency & Renewable Energy
Operated by the Alliance for Sustainable Energy, LLC
This report is available at no cost from the National Renewable Energy
Laboratory (NREL) at www.nrel.gov/publications.
Subcontract Report
NREL/SR-7A40-68558
October 2017
Contract No. DE-AC36-08GO28308

Chapter 2: Commercial and
Industrial Lighting Evaluation
Protocol
The Uniform Methods Project: Methods for
Determining Energy Efficiency Savings for
Specific Measures
Created as part of subcontract with period of performance
September 2011 – September 2016
This version supersedes the version originally published in

April 2013. The content in this version has been updated.
Dakers Gowans
Left Fork Energy
Harrison, New York
Chad Telarico
DNV GL
Mahwah, New Jersey
NREL Technical Monitor: Charles Kurnik
Prepared under Subcontract No. LGJ-1-11965-01
NREL is a national laboratory of the U.S. Department of Energy

Office of Energy Efficiency & Renewable Energy
Operated by the Alliance for Sustainable Energy, LLC
National Renewable Energy Laboratory Subcontract Report

15013 Denver West Parkway NREL/SR-7A40-68558
Golden, CO 80401 October 2017
303-275-3000 • www.nrel.gov
Contract No. DE-AC36-08GO28308
This publication was reproduced from the best available copy submitted by the subcontractor.
NOTICE
This report was prepared as an account of work sponsored by an agency of the United States government.
Neither the United States government nor any agency thereof, nor any of their employees, makes any warranty,
express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of
any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately
owned rights. Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation,
or favoring by the United States government or any agency thereof. The views and opinions of authors
expressed herein do not necessarily state or reflect those of the United States government or any agency thereof.

Available electronically at SciTech Connect http:/www.osti.gov/scitech
Available for a processing fee to U.S. Department of Energy

and its contractors, in paper, from:
U.S. Department of Energy

Office of Scientific and Technical Information
P.O. Box 62
Oak Ridge, TN 37831-0062
OSTI http://www.osti.gov
Phone: 865.576.8401
Fax: 865.576.5728
Email: reports@osti.gov
Available for sale to the public, in paper, from:
U.S. Department of Commerce

National Technical Information Service
5301 Shawnee Road
Alexandria, VA 22312
NTIS http://www.ntis.gov
Phone: 800.553.6847 or 703.605.6000
Fax: 703.605.6900
Email: orders@ntis.gov
Cover Photos by Dennis Schroeder: (left to right) NREL 26173, NREL 18302, NREL 19758, NREL 29642, NREL 19795.
NREL prints on paper that contains recycled content.

Disclaimer
These methods, processes, or best practices (“Practices”) are provided by the National
Renewable Energy Laboratory (“NREL”), which is operated by the Alliance for Sustainable
Energy LLC (“Alliance”) for the U.S. Department of Energy (the “DOE”).
It is recognized that disclosure of these Practices is provided under the following conditions and
warnings: (1) these Practices have been prepared for reference purposes only; (2) these Practices
consist of or are based on estimates or assumptions made on a best-efforts basis, based upon
present expectations; and (3) these Practices were prepared with existing information and are
subject to change without notice.
The user understands that DOE/NREL/ALLIANCE are not obligated to provide the user with
any support, consulting, training or assistance of any kind with regard to the use of the Practices
or to provide the user with any updates, revisions or new versions thereof. DOE, NREL, and
ALLIANCE do not guarantee or endorse any results generated by use of the Practices, and user
is entirely responsible for the results and any reliance on the results or the Practices in general.
USER AGREES TO INDEMNIFY DOE/NREL/ALLIANCE AND ITS SUBSIDIARIES,

AFFILIATES, OFFICERS, AGENTS, AND EMPLOYEES AGAINST ANY CLAIM OR
DEMAND, INCLUDING REASONABLE ATTORNEYS' FEES, RELATED TO USER’S USE
OF THE PRACTICES. THE PRACTICES ARE PROVIDED BY DOE/NREL/ALLIANCE "AS
IS," AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING BUT NOT LIMITED
TO THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL
DOE/NREL/ALLIANCE BE LIABLE FOR ANY SPECIAL, INDIRECT OR
CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER, INCLUDING BUT
NOT LIMITED TO CLAIMS ASSOCIATED WITH THE LOSS OF PROFITS, THAT MAY
RESULT FROM AN ACTION IN CONTRACT, NEGLIGENCE OR OTHER TORTIOUS
CLAIM THAT ARISES OUT OF OR IN CONNECTION WITH THE ACCESS, USE OR
PERFORMANCE OF THE PRACTICES.
iii
This report is available at no cost from the National Renewable Energy Laboratory (NREL) at www.nrel.gov/publications.
Preface
This document was developed for the U.S. Department of Energy Uniform Methods Project
(UMP). The UMP provides model protocols for determining energy and demand savings that
result from specific energy-efficiency measures implemented through state and utility programs.
In most cases, the measure protocols are based on a particular option identified by the
International Performance Verification and Measurement Protocol; however, this work provides
a more detailed approach to implementing that option. Each chapter is written by technical
experts in collaboration with their peers, reviewed by industry experts, and subject to public
review and comment. The protocols are updated on an as-needed basis.
The UMP protocols can be used by utilities, program administrators, public utility commissions,
evaluators, and other stakeholders for both program planning and evaluation.
To learn more about the UMP, visit the website, https://energy.gov/eere/about-us/ump-home, or

download the UMP introduction document at http://www.nrel.gov/docs/fy17osti/68557.pdf.
iv
Acknowledgments
The chapter authors wish to thank and acknowledge the following individuals for their
thoughtful comments and suggestions on drafts of this protocol:
• Jeff Cropp and M. Sami Khawaja of Cadmus
• Miriam Goldberg of DNV GL
• Dick Spellman of GDS
• Frank Stern of Navigant
• Kevin Warren of Warren Energy.
Suggested Citation
Gowans, D.; Telarico, C. (2017). Chapter 2: Commercial and Industrial Lighting Evaluation
Protocol, The Uniform Methods Project: Methods for Determining Energy-Efficiency Savings
for Specific Measures. Golden, CO; National Renewable Energy Laboratory. NREL/SR-7A40-
68558. http://www.nrel.gov/docs/fy17osti/68558.pdf
v
Acronyms
CF coincidence factor
COP coefficient of performance
CT current transformer
DOE U.S. Department of Energy
HGSF heat gain space fraction
HID high-intensity discharge
HOU hours of use
HVAC heating, ventilating, and air conditioning
ISR in-service rate
kW kilowatt
kWh kilowatt-hour
LED light-emitting diode
LPD lighting power density
M&V measurement and verification
PCF peak coincidence factor
UMP Uniform Methods Project
W watt
vi
Protocol Updates
The original version of this protocol was published in April 2013.
This chapter has been updated to incorporate the following revisions:
• Added new section for midstream programs.
• Provided additional detail on the recommended duration of metering.
• Changed the recommended minimum metering time from two weeks to four weeks.
• Added an alternative approach to estimating interactive effects through the use of an

engineering equation (in addition to the current approach that uses stipulated factors).
• Provided guidance for creating fixture codes for light-emitting diode fixtures not
currently found in most look-up tables.
• Updated the protocol on reporting uncertainty based on new material from the
International Performance Measurement and Verification Protocol.
• Applied the controls requirements of International Energy Conservation Code 2012 /

90.1-2010 to estimate of baseline hours of use for new construction projects.
vii
Table of Contents
1 Measure Description ............................................................................................................................ 1
2 Application Conditions of the Protocol .............................................................................................. 2
2.1 Common Program Types .............................................................................................................. 2
2.2 Program Target Markets................................................................................................................ 3
3 Savings Calculations ........................................................................................................................... 4
3.1 Algorithms ..................................................................................................................................... 4
4 Role of the Lighting Program Implementer ....................................................................................... 8
4.1 Program Implementer Data Requirements .................................................................................... 8
4.2 Implementation Data Collection Method ...................................................................................... 8
5 Role of the Evaluator .......................................................................................................................... 10
5.1 Evaluator Data Requirements ...................................................................................................... 10
5.2 Evaluator Data Collection Method .............................................................................................. 10
6 Measurement and Verification Plan .................................................................................................. 12
6.1 IPMVP Option............................................................................................................................. 12
6.2 Verification Process .................................................................................................................... 12
6.3 Measurement Process .................................................................................................................. 13
6.4 Report M&V and Program Gross Savings .................................................................................. 18
6.5 Data Requirements and Sources .................................................................................................. 19
7 Gross Impact Evaluation ................................................................................................................... 25
7.1 Sample Design............................................................................................................................. 25
8 Other Evaluation Issues..................................................................................................................... 27
8.1 Upstream/Midstream Delivery .................................................................................................... 27
8.2 New Construction ........................................................................................................................ 29
8.3 First Year Versus Lifetime Savings ............................................................................................ 30
8.4 Program Evaluation Elements ..................................................................................................... 31
9 Looking Forward................................................................................................................................. 32
10 References .......................................................................................................................................... 33
11 Resources ........................................................................................................................................... 34
12 Appendix ............................................................................................................................................. 36
viii
List of Figures
Figure 1. Dual baseline ............................................................................................................................... 31
List of Tables
Table 1. Required Lighting Data Form Fields .............................................................................................. 9
Table 2. Lighting Data Required by Evaluator ........................................................................................... 11
Table 3. Comparison of 12-Month and 3-Month HOU and Summer CF by Building Type ...................... 17
Table 4. Comparison of 1-, 2-, and 3-Month Metering to Actual 12-Month Energy Savings .................... 18
Table 5. LED Fixture/Lamp Categories ...................................................................................................... 20
Table 6. Lighting Data Required by Evaluator for Midstream Programs ................................................... 27
Table 7. Example Lighting Inventory Form ............................................................................................... 36
Table 8. New York Standard Approach for Estimating Energy Savings .................................................... 37
Table 9. New York Standard Approach for Estimating Energy Savings .................................................... 38
Table 10. Midstream Baseline Wattage, Linear Lamps, and Fixtures; HID Interior and Exterior Fixtures 39
ix
1 Measure Description
The Commercial and Industrial Lighting Evaluation Protocol (the protocol) describes methods to
account for gross energy savings resulting from the programmatic installation of efficient
lighting equipment in large populations of commercial, industrial, and other nonresidential
facilities. This protocol does not address savings resulting from changes in codes and standards,
or from education and training activities. A separate Uniform Methods Project (UMP) protocol,
Chapter 3: Commercial and Industrial Lighting Controls Evaluation Protocol, addresses
methods for evaluating savings resulting from lighting control measures such as adding time
clocks, tuning energy management system commands, and adding occupancy sensors.
Historically, lighting equipment has accounted for a significant portion of cost-effective, electric
energy efficiency resources in the United States, a trend likely to continue as old technologies
improve and new ones emerge. By following the methods presented here, the energy savings
from lighting efficiency programs in different jurisdictions or regions can be measured
uniformly, providing planners, policymakers, regulators, and others with sound, comparable data
for comprehensive energy planning. Also, the methods here can be scaled to match the
evaluation costs to the value of the resulting information. 1
An energy efficiency measure is defined as a set of actions and equipment changes that result in
reduced energy use—compared to standard or existing practices—while maintaining the same or
improved service levels for customers or processes. Energy-efficient lighting measures in
existing facilities deliver the light levels (illuminance and spatial distribution) required for
activities or processes at reduced energy use, compared to original or baseline conditions. In new
construction, “original or baseline condition” usually refers to the building codes and standards
in place at the time of construction.
Examples of energy-efficient lighting measures in commercial, industrial, and other

nonresidential facilities include:
• Retrofitting existing, linear, fluorescent fixtures with efficacious 2 lamps and ballasts, or
delamping over-lit spaces
• Replacing compact fluorescent lamps with screw-in light-emitting diodes (LED) lamps
• Replacing metal halide high-bay fixtures with efficacious LED high-bay equipment.
In practice, lighting retrofit projects and new construction projects commonly implement lighting
fixture and lighting controls measures concurrently. This protocol accommodates these mixed
measures.
1
As discussed in the “Considering Resource Constraints” section of the UMP Chapter 1: Introduction, small
utilities (as defined under U.S. Small Business Administration regulations) may face additional constraints in
undertaking this protocol. Therefore, alternative methodologies should be considered for such utilities.
2
Efficiency of lighting equipment is expressed as “efficacy,” in units of lumens per Watt, where lumens are a
measure of light output.
1
2 Application Conditions of the Protocol
Energy efficiency lighting programs result in the installation of commercial, industrial, and
nonresidential high-efficiency lighting measures in customer facilities. The programs can take
advantage of varying delivery mechanisms, depending on target markets and customer types.
Primarily, these mechanisms can be distinguished by the parties receiving incentive payments
from a program. Although the methods described in this protocol apply to all delivery
mechanisms, issues verifying customer and baseline equipment data vary.
2.1 Common Program Types

The following are descriptions of common program types used to acquire lighting energy and
demand savings.
2.1.1 Incentive and Rebate

Under this model, implementers pay program participants in target markets to install lighting
measures. This type of program is generally referred to as a downstream program. A participant
receives either an incentive payment, based on savings ($/kilowatt-hour [kWh]), or a rebate for
each fixture or lamp ($/fixture, $/lamp). The terms incentive and rebate sometimes are used
interchangeably, but generally, incentives are calculated based on project savings and rebates are
based on equipment installed. Examples of participants include contractors, building owners, and
property managers.
Savings can be estimated using simple engineering calculations. Some programs include a
measurement and verification (M&V) process, in which key parameters—such as hours of use
(HOU), baseline, and retrofit fixture wattages—are verified or measured, or both, as part of
project implementation.
Rebate programs typically pay for specific lighting equipment types (for example, a 4-foot, four-
lamp, T5 electronic ballast fixture), often after they have been installed, so assumptions must be
made about baseline or replaced equipment. The result is a tradeoff: increased administrative
efficiency for less certainty about baseline conditions (and therefore, savings).
Incentive programs often collect more detailed baseline data than do rebate programs. Typically,
these data include baseline and retrofit equipment wattages and HOUs, which facilitate
determination of savings impacts.
Although rebate programs typically track useful information about replacement lighting
equipment, they may not collect baseline data.
2.1.2 Upstream Buy-Down

In upstream buy-down scenarios, programs pay incentive dollars to one or more entities such as
retail outlets, distributors, or manufacturers in the lighting equipment market distribution chain.
The upstream approach has been widely used in the residential sector, particularly for compact
fluorescent lamp (CFL) commercial and industrial lighting programs.
Upstream programs do not interact with the end-use customers purchasing energy-efficient
equipment, making the determination of baseline conditions and installation rates more difficult
2
than for incentive and rebate programs. Program planners, implementers, and evaluators estimate
these parameters using regional bulk sales data, market research studies, assessments of current
product standards and practices, and experience with other programs.
A subset of upstream programs is the midstream model where incentives are paid to distributors
for sales of pre-approved qualified products. Purchasers are contractors and commercial,
institutional and governmental accounts. Programs can leverage the relationship between
distributor, purchaser, and end-user to require information about equipment sales and the end-use
installation. Midstream lighting programs are increasingly included in ratepayer-funded energy
efficiency portfolios.
2.1.3 Direct Install

Under this delivery approach, contractors, acting on a program’s behalf, install energy-efficient
lighting equipment in customer facilities. The programs pay contractors directly. Customers
receive a lighting retrofit at reduced cost. Direct-install programs often target hard-to-reach
customers—typically small businesses—that are overlooked by contractors working with
incentive and rebate programs.
Direct-install programs can usually collect precise information about baseline and replacement
equipment, and the program implementers may have reasonable estimates of annual operating
hours. Data, when collected, can be used directly by impact evaluation researchers.
2.2 Program Target Markets

In addition to being distinguished by their delivery mechanisms, commercial, industrial, and
non-residential lighting programs can be classified by targeting retrofits (serving existing
facilities) and new construction markets. Program delivery types described above apply to
retrofit programs. New construction programs also employ incentives and rebates (and customers
may benefit from upstream buy-downs) to improve lighting energy efficiency.
New construction programs present evaluators with a dilemma in establishing baselines for
buildings that have yet to be built. The problem is addressed by referring to new construction
energy codes for commercial, industrial, and nonresidential facilities (usually by referencing
International Energy Conservation Code or ASHRAE Standard 90.1). The codes define lighting
efficiency, primarily in terms of lighting power density (lighting watts/ft2), calculated using
simple spreadsheets. Other federal, state, and local standards may set additional baseline
constraints on lamps, ballasts, and fixture efficiency/efficacy.
3
3 Savings Calculations
Project and program savings for lighting and other technologies result from the difference
between the energy consumption that would have occurred had the measure not been
implemented (the baseline) and the consumption occurring after the retrofit. Energy calculations
use the following fundamental equation:
Equation 1. Energy Savings = (Baseline-Period Energy Use – Reporting-Period Energy

Use) ± Adjustments
The equation’s adjustment term calibrates baseline or reporting use and demand to the same set
of conditions. Common adjustments account for changes in schedules, occupancy rates, weather,
or other parameters that can change between baseline and reporting periods. Adjustments
commonly apply to heating, ventilating, and air conditioning (HVAC) measures, but less
commonly to lighting measures, or are inherent in algorithms for calculating savings.
Regulators and program administrators may require that lighting energy efficiency programs
report demand and energy savings. Demand calculations use the following fundamental
equation:
Equation 2. Demand Savings = (Baseline-Period Demand – Reporting-Period Demand)

± Adjustments
Demand savings, which is calculated for one or more time-of-use periods, is typically reported
for the peak period of the utility system serving the efficiency program customers.
3.1 Algorithms
The following equations calculate first-year energy and demand on-site savings for lighting
measures in commercial, industrial, nonresidential facilities:
3.1.1 Energy Savings

Equations in this section are used to calculate first-year energy savings for lighting measures.
Equation 3. Lighting Electric Energy Savings
𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡
𝑓𝑓𝑓𝑓𝑓𝑓 𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏,𝑖𝑖 ∙ 𝑞𝑞𝑞𝑞𝑞𝑞𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏,𝑖𝑖
= �� ∙ 𝐻𝐻𝐻𝐻𝐻𝐻𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 �
1000 𝑢𝑢
𝑢𝑢,𝑖𝑖
𝑓𝑓𝑓𝑓𝑓𝑓 𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒,𝑖𝑖 ∙ 𝑞𝑞𝑞𝑞𝑞𝑞𝑒𝑒𝑒𝑒,𝑖𝑖
− �� ∙ 𝐻𝐻𝐻𝐻𝐻𝐻𝑒𝑒𝑒𝑒 � � ∙ 𝐼𝐼𝐼𝐼𝐼𝐼
1000 𝑢𝑢
𝑢𝑢,𝑖𝑖
Where:
kWh Save light = Annual kWh savings resulting from the lighting efficiency project
fix watt base, energy efficient, i = Fixture wattage, baseline or energy-efficient, fixture type i
4
qty base, energy efficient, i = Fixture quantity, baseline or energy-efficient, fixture type i
u = Usage group, a collection of fixtures sharing the same operating hours and schedules,
for example all fixtures in office spaces or hallways
HOU base, energy efficient = Annual hours of use, baseline or energy-efficient, usually assumed
unchanged from baseline unless new controls are installed
ISR = In-service rate, the percentage of incentivized lamps or fixtures that are installed
and operating. Applies to upstream buy-down programs, normally not applicable for
incentive and rebate programs
Equation 4. Interactive Cooling Energy Savings for Interior Lighting

𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 ∙ 𝑘𝑘𝑘𝑘 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 ∙ 𝐻𝐻𝐻𝐻𝐻𝐻𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 =
𝐶𝐶𝐶𝐶𝐶𝐶𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
OR
𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 ∙ 𝐼𝐼𝐼𝐼𝑘𝑘𝑘𝑘ℎ,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
Equation 5. Interactive Heating Energy Penalty for Interior Lighting

−𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 ∙ 𝑘𝑘𝑘𝑘 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 ∙ 𝐻𝐻𝐻𝐻𝐻𝐻ℎ𝑒𝑒𝑒𝑒𝑒𝑒
𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−ℎ𝑒𝑒𝑒𝑒𝑒𝑒 =
𝐶𝐶𝐶𝐶𝐶𝐶ℎ𝑒𝑒𝑒𝑒𝑒𝑒
OR
𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−ℎ𝑒𝑒𝑒𝑒𝑒𝑒 = 𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 ∙ 𝐼𝐼𝐼𝐼𝑘𝑘𝑘𝑘ℎ,ℎ𝑒𝑒𝑎𝑎𝑎𝑎
Where:
kWh Save interact-cool = Interactive cooling energy impact due to a lighting efficiency
project
kWh Save interact-heat = Interactive heating energy impact from a lighting efficiency project
kW Save light = Connected kW savings (kWbase – kWefficient) due to a lighting efficiency

project
HOU cool = Hours of use of lighting equipment coincident with cooling system operation
HOU heat = Hours of use of lighting equipment coincident with heating system operation
IF kWh, cool = Interactive cooling factor: the ratio of cooling energy reduction per unit of
lighting energy reduction
5
IF kWh, heat = Interactive heating factor: the ratio of heating energy increase per unit of
lighting energy
COP cool = Cooling system coefficient of performance
COP heat = Heating system coefficient of performance
HGSF = Heat gain space fraction: the percent of lighting wattage that is transferred to the
conditioned space as thermal energy.
The protocol provides two options each for calculating cooling and heating interactive effects: an
engineering approach and a stipulated factor approach.
In the engineering approach, the HGSF represents the percentage of the lighting energy that is
thermal energy added to the conditioned space. According to a 2007 ASHRAE study
(Chantrasrisalai and Fisher 2007), the percentage of lighting energy transmitted to the space can
range from 12% to 100% depending on the type of light fixture at typical operating conditions. 3
The protocol recommends a default of 70% to 80% HGSF. Calculating building-specific HGSF
values is unusual due to the level of effort required.
Interactive effects apply only to interior lighting that operates in mechanically cooled or heated
spaces.
Equation 6. Total Annual Energy Savings Due to Lighting Project

𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡= 𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 + 𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 + 𝑘𝑘𝑘𝑘ℎ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−ℎ𝑒𝑒𝑒𝑒𝑒𝑒
3.1.2 Electric Peak Demand Savings

The equations in this section are used to calculate first-year electric peak demand savings for
lighting measures. Additional information is available in the Uniform Methods Project (UMP)
Chapter 10: Peak Demand and Time-Differentiated Energy Savings Cross-Cutting Protocol.
Equation 7. Lighting Electric Peak Demand Savings

𝑓𝑓𝑓𝑓𝑓𝑓 𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏,𝑖𝑖 ∙ 𝑞𝑞𝑞𝑞𝑞𝑞𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏,𝑖𝑖 𝑓𝑓𝑓𝑓𝑓𝑓 𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑒𝑒𝑒𝑒,𝑖𝑖 ∙ 𝑞𝑞𝑞𝑞𝑞𝑞𝑒𝑒𝑒𝑒,𝑖𝑖
𝑘𝑘𝑘𝑘 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 = 𝐶𝐶𝐶𝐶 ∙ � � − � ∙ 𝐼𝐼𝐼𝐼𝐼𝐼
1000 1000 𝑢𝑢
𝑢𝑢,𝑖𝑖
Where:
CF = coincidence factor, the fraction (0.0 to 1.0) of connected lighting load turned on
during a utility peak period
3
A value less than 100% means that a portion of the lighting energy is being transferred into the plenum rather than
the conditioned space.
6
Equation 8. Interactive Electric Cooling Demand Savings for Interior Lighting
𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻 ∙ 𝑘𝑘𝑘𝑘 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡
𝑘𝑘𝑘𝑘 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 =
𝐶𝐶𝐶𝐶𝐶𝐶𝑐𝑐𝑐𝑐𝑐𝑐𝑙𝑙
OR
𝑘𝑘𝑘𝑘 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 𝑘𝑘𝑘𝑘 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 ∙ 𝐼𝐼𝐼𝐼𝑘𝑘𝑘𝑘,𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
Where:
kW Peak Save interact-cool = Interactive electric cooling peak demand impact from a
lighting efficiency project
HGSF = Heat Gain Space Fraction: the percent of lighting wattage that is transferred to
the conditioned space as thermal energy
COP cool = Cooling system coefficient of performance
IF kW, cool = Interactive cooling factor, ratio of cooling demand reduction per unit of
lighting demand reduction during the peak period resulting from the reduction in lighting
waste heat removed by an HVAC system
Interactive effects apply only to interior lighting operating in mechanically cooled spaces.
Interactive heating effects are often ignored in North America because heating equipment
is typically nonelectric and heating demand may not coincide with utility system peaks.
Equation 9. Total Electric Peak Demand Savings Due to Lighting Project

𝑘𝑘𝑘𝑘 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑡𝑡𝑡𝑡𝑡𝑡𝑎𝑎𝑎𝑎= 𝑘𝑘𝑘𝑘 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑙𝑙𝑙𝑙𝑙𝑙ℎ𝑡𝑡 + 𝑘𝑘𝑘𝑘 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
7
4 Role of the Lighting Program Implementer
Successful application of this protocol requires collecting standard data in a prescribed format as
part of the implementation process. The protocol further requires tracking project and program
savings estimated on the basis of those standard data.
The implementer is responsible for ensuring necessary data are collected to track program
activity and to calculate savings at the project level. The implementer is also responsible for
maintaining a program activity record, including anticipated savings by project.
4.1 Program Implementer Data Requirements

The protocol recommends the program implementer collect and archive, for all projects, all data
needed to execute the savings algorithms. These data are:
• Baseline fixture inventory, including fixture wattage

• Baseline fixture quantities
• Baseline lighting HOU
• Efficient fixture inventory, including wattage
• Efficient fixture quantities
• Efficient lighting HOU
• Usage group assignments
• Heating and cooling equipment types
• Interactive factor for cooling, or cooling equipment COP and lighting HOU coincident
with cooling equipment operation (optional)
• Interactive factor for heating, or heating equipment COP and lighting HOU coincident
with heating equipment operation (optional).
Facilities—or spaces within facilities where the project is installed—are classified as
cooled/uncooled or heated/unheated. Recording information about heating and cooling
equipment and fuel types for each facility or space allows for more precise estimation of
interactive effects. Implementers may elect to use a program-level default values for the percent
of space that is heated or cooled. The values can be based on earlier studies or evaluation reports
for similar populations.
Note that some of the information will not be available for some program types (e.g., baseline
fixture information for new construction, upstream, or midstream programs). See Section 8 for
recommendations for midstream programs.
4.2 Implementation Data Collection Method

The protocol recommends participants collect and submit required data as a condition for
enrolling in the program. The protocol also recommends the implementer specify the data
reporting format, either by supplying a structured form (such as a spreadsheet) or by specifying
the data fields and types used when submitting material to the program.
8
The format of the data must be electronic, searchable, and sortable. It must also support
combining multiple files into single tables for analysis by the implementer. Microsoft Excel and
comma-separated text files are acceptable formats; faxes, PDFs, and JPEGs are not.
The data reporting format should be structured to allow verification of the project installation.
Each record or line in the report: (1) is a collection of identical fixture types, (2) is installed in an
easily located room, floor, or space, and (3) belongs to one usage group. Table 1 lists the fields
required in the data reporting format. All data are supplied by the participant or implementer.
Table 1. Required Lighting Data Form Fields

Field Notes
Location Floor number, room number, description
Usage group
Location heating Yes/no
Location heating type Boiler steam/hydronic, rooftop gas-fired, etc.
Location heating fuel Electric, natural gas, fuel oil, etc.
Location cooling Yes/no
Location cooling type Water cooled chiller, air cooled chiller, packaged DX, etc.
Location cooling fuel Electric, natural gas, etc.
Baseline fixture type From look-up table supplied by implementer, manufacturer cut sheet
Baseline fixture count
Baseline fixture watt From look-up table supplied by implementer, manufacturer cut sheet
Baseline HOU From look-up table supplied by implementer, estimated by customer, bulk
meter services or meter data
Efficient fixture type From look-up table supplied by implementer, manufacturer cut sheet
Efficient fixture count
Efficient fixture watt From look-up table supplied by implementer, manufacturer cut sheet
Efficient lighting HOU Same as baseline if no controls installed
IFcool, or COPcool and Interactive factor for cooling from look-up table, or site-specific COPcool and
HOUcool HOUcool (optional)
IFheat, or COPheat and Interactive factor for heating from look-up table, or site-specific COPheat and
HOUheat HOUheat (optional)
kWhsave Calculated using savings algorithms
kW-Peaksave Calculated using savings algorithms
The Appendix to this protocol contains an example of a lighting inventory form with the fields
listed in Table 1.
Information at the usage-group level will typically not be available for midstream or upstream
programs. Location-specific information such as cooling and heating type may also be
unavailable. In such cases, program-level or building type-level defaults will be used by the
implementer, and the evaluation may work to estimate and update these assumptions. See
Section 8 for recommendations for the evaluation of midstream programs.
9
5 Role of the Evaluator
This section describes the evaluator’s role in determining gross energy savings due to
participation in a lighting energy efficiency program. Gross savings result directly from
program-related actions taken by participants in an energy efficiency program. A simple way of
thinking about gross savings is that they can be observed at the customer utility meter, at least in
theory. In practice, it is difficult to isolate program-induced changes from other simultaneous
changes, or to attribute them solely to the program itself.
Gross savings, the focus of this protocol, are adjusted through a separate set of actions to report
net savings. Net savings are only those savings that can be attributed to the program. The concept
of net savings recognizes that some participants would have acted on their own to adopt energy
efficiency strategies, might have installed additional equipment as a result of increased
awareness of the value of energy efficiency through their participation, or improved their energy
efficiency operations due to market changes induced by a program’s operations. For more on net
savings, see UMP Chapter 21: Estimating Net Savings – Common Practices.
Steps taken by the evaluator under this protocol include:
1. Reviewing a statistically significant random sample of completed projects, including

conducting on-site M&V activities
2. Calculating a gross realization rate (the ratio of evaluator-to-implementer anticipated
gross savings)
3. Using the realization rate to adjust the implementer-estimated gross savings.
5.1 Evaluator Data Requirements
This protocol recommends the impact evaluator collect the same data as the implementer. As
described in Section 6, the evaluator must have access to the implementation lighting inventory
forms and participant application material for each project in the sample. For some program
types, specifically midstream and upstream, the evaluator will collect more data than the
implementer. This is the case when the evaluator conducts onsite verification of baseline
conditions for a midstream program.
5.2 Evaluator Data Collection Method

Under the protocol, the implementer provides the evaluator with a copy of the program and
project data tracking record for the evaluation review period. That record contains the fields
specified in Table 1. The implementer also provides all records for projects in the evaluation
review sample, including application materials and site contact information.
The protocol recommends the evaluator collect additional M&V data during site visits conducted
for the sample of evaluation review projects. Table 2 lists data required for each project in the
evaluation sample.
10
Table 2. Lighting Data Required by Evaluator
Field Note
Location From implementer
Usage group From implementer
Location heating From implementer, verified by evaluator
Location heating type From implementer, verified by evaluator
Location heating fuel From implementer, verified by evaluator
Location cooling From implementer, verified by evaluator
Location cooling type From implementer, verified by evaluator
Location cooling fuel From implementer, verified by evaluator.
Baseline fixture type From implementer, verified by evaluator
Baseline fixture count From implementer, verified by evaluator
Baseline fixture watt From implementer, verified by evaluator
Baseline HOU From implementer, verified by evaluator
Efficient fixture type From implementer, verified by evaluator
Efficient fixture count From implementer, verified by evaluator
Efficient fixture watt From implementer, verified by evaluator
Efficient lighting HOU Measured by evaluator
IFcool, or COPcool and Interactive factor for cooling from look-up table, or site-specific COPcool and
HOUcool HOUcool (optional)
IFheat, or COPheat and Interactive factor for heating from look-up table, or site-specific COPheat
HOUheat and HOUheat (optional)
ISR Measured by evaluator
kWhsave Calculated using savings algorithms
kW-Peaksave Calculated using savings algorithms
11
6 Measurement and Verification Plan
The M&V plan describes how evaluators determine actual energy savings in a facility where a
lighting efficiency project has been installed. Evaluators use M&V to establish energy savings
for a random sample of projects. The M&V results are applied to the population of all completed
projects to determine program gross savings. The sampling and application processes are
described in UMP Chapter 11: Sample Design Cross-Cutting Protocol. The sample size should
be determined following the recommendations in UMP Chapter 11.
All M&V activities in the protocol are conducted on a representative sample of completed
projects, drawn from a closed reporting period (for example, a program year).
6.1 IPMVP Option

The protocol recommends evaluators conduct M&V according to the International Performance
Measurement and Verification Protocol (IPMVP) Option A—Retrofit Isolation: Key Parameter
Measurement approach.
The key measured parameters are the HOU terms in Equation 1. The fixture quantity parameter
is verified through an inspection process. The fixture wattage parameter is verified through a
combination of on-site inspections and look-up tables of fixture demand (Watts).
Option A is recommended because the demand (Watts) values are known and published for
nearly all fixture types and configurations, and therefore need not be measured, whereas lighting
operating hours vary widely from building to building.
6.2 Verification Process

Verification involves visual inspections and engineering calculations to establish an energy
efficiency project’s potential to achieve savings. The verification process determines the fixture
wattage and fixture quantity parameters in Equation 1.
The process includes the following steps:
1. Select a representative sample of projects for review (see UMP Chapter 11: Sample
Design Cross-Cutting Protocol for guidance on sampling).
2. Schedule a site visit with a facility representative for each project in the sample.
3. Conduct an on-site review for each project. Inspect a representative sample of the energy
efficiency lighting fixtures reported by the implementer. The protocol recommends
selecting the sample from the implementer’s inventory records before going on site (see
UMP Chapter 11: Sample Design Cross-Cutting Protocol for guidance on sampling.)
4. Confirm or correct the reported energy-efficient fixture type and wattage for each fixture
in the sample.
5. Confirm or correct the reported quantity for all energy-efficient fixtures in the sample.
6. Confirm or correct the heating/cooling status and associated equipment for the spaces in
the sample.
12
7. Interview facility representatives to check baseline fixture types and quantities reported
for the sample. Confirmation or correction is based on the interviews. When available,
interviews are supplemented by physical evidence, such as: fixture types in areas not
changed by the project, replacement stock for lamps and ballasts, and/or stockpiles of
removed fixtures/lamps stored on-site for recycle or disposal.
8. Update lighting inventory form for the sample, based on findings from the on-site review.
The implementer has the primary responsibility for maintaining accurate project inventories that
support evaluation research, including locations of individual fixtures or lamps. An example of
an inventory form that meets the requirements of the protocol is provided in the Appendix.
Evaluators may have difficulty locating fixtures or lamps that contribute to a program’s reported
savings when project records are incomplete, information about fixture or lamp location and type
is imprecise, the facility representative guiding the evaluator during a site visit is unfamiliar with
the project, or the facility has undergone a change in ownership or retrofit since the project was
completed.
When faced with incomplete information, evaluators can use a mix of strategies to conduct
verification site visits. If a line in the inventory cannot be located, the verification sample can be
expanded to include an entire floor, wing, or other space, and all of the fixtures within the space
counted and identified. This approach works when room numbers are not provided in the
inventory, for example. Another strategy is to substitute a room or space for one that cannot be
located. This can work when there are large numbers of identical spaces such as classrooms,
offices, and restrooms. Another is to contact the lighting contractor who installed the efficient
lighting products. Contractors have to create their own inventories to manage construction and
order material for all lighting projects, and they may be willing to share their lists.
Evaluation field staff will need to exercise judgement when using these strategies as to whether
or not to count as verified any fixtures and lamps that cannot be located. The evidence can be
inconclusive but still support a reasonable inference.
At the completion of the verification process, the evaluator has confirmed or corrected the fixture
wattage and fixture quantity parameters in Equation 1. The process for determining the HOU
parameters is described in the following section.
6.3 Measurement Process

The measurement process involves using electronic metering equipment to collect the data for
determining the HOU parameters in Equation 1. Most often, the equipment is installed
temporarily during the measurement period. Energy management systems that monitor lighting
circuits can also be used to measure HOU.
Metering equipment used to measure lighting operating hours either records a change of state
(light on, light off) or continuously samples and records current in a lighting circuit or light
output of a fixture. All data must be time-stamped for application in the protocol.
13
6.3.1 Use of Data Loggers
Lighting operating hours are typically determined through the use of temporary equipment such
as data loggers.
Change-of-state lighting data loggers are small (matchbox size) integrated devices, which
include a photocell, a microprocessor, and memory. The data logger is mounted temporarily
inside a fixture (or in proximity to it) and is calibrated to the light output of the fixture. Each time
the lamp(s) in the fixture are turned on or off, the event is recorded and time-stamped.
Data loggers that continuously sample and record lighting operating hour information usually
require an external sensor such as a current transformer (CT) or photocell. Data loggers with CTs
can monitor amperage to a lighting circuit. Spot measurements of the circuit’s amperage with the
lights on and off establish the threshold amperage for the on condition. Similarly, a data logger
with an external photocell can record light levels in a space. Spot measurements of lumen levels
with the fixtures on and off establish the light level threshold for the on condition.
Although measuring amperage with data loggers is common, the continuous monitoring of
lumen levels to determine hours of operation is less common.
Data logger failure commonly occurs due to incorrect adjustments, locations, or software launch.
Thus, this protocol recommends following manufacturer recommendations carefully and
deploying extra loggers as a cushion against failure.
6.3.2 Metering
The measurement process involves metering lighting operating hours for the representative
sample of fixtures selected for the verification process. Meters are deployed or trends set up in an
existing energy management system during the verification site visit.
This process entails the following activities:
1. Meter operating hours for each circuit in the verification sample.

A. Develop a metering plan that includes the location of a random selection of
required metering points (the metering sample) by usage group. Guidance on
sampling is provided in UMP Chapter 11: Sample Design Cross-Cutting
Protocol. The plan should be developed before going on site. If the inventory is
missing good location information, the plan can be adjusted while in the field to
make sure the number of metering points by usage group is maintained and that
the selection remains random to the extent possible.
B. If using light loggers, deploy loggers in one or more fixtures controlled by the
circuit. Only one logger is required per circuit; additional loggers may be
deployed to offset logger failure or loss. A rule of thumb is to install the number
of loggers specified in the metering plan for each usage group plus an additional
10%.
C. If measuring amperage, install CT and data logger in a lighting panel for a
sampled circuit. The sampling interval should be 15 minutes or less. Spot-
14
measure amperage with lights on and off for the circuit leg with the CT. Record
the amperage threshold for the lights-on condition.
D. If using an energy management system, program it to sample and record lighting
on/off status for each circuit in the sample. The sampling interval should be 15
minutes or less. Check that the energy management system has sufficient capacity
to archive recorded data, and that the metering task will not adversely slow
system response times.
2. Check data logger operation. Before leaving the site, spot-check a few data loggers to
confirm they are recording data as expected. Correct any deficiencies and if the
deficiencies appear to be systemic, redeploy the loggers. If using energy management
system trends, spot-check recorded data.
3. Leave the metering equipment in place for the duration of the monitoring period. The
protocol recommends a monitoring period that captures the full range of facility operating
schedules. The following are some rules of thumb for specifying the length of the
monitoring period. More detailed guidance is provided Section 6.3.3.
A. For facilities with constant schedules (such as office buildings, grocery stores, and
retail shops), the protocol requires metering for a minimum of four weeks. The
weeks should not be abnormal (e.g. during the end-of-year holidays).
B. For facilities with variable or irregular schedules, additional metering time is
required. The protocol recommends a monitoring period long enough to capture
the average operation over the full range of variable schedules.
C. Facilities with seasonal schedules, such as schools, should be monitored during
active periods; additional monitoring can be done during the inactive periods, or if
the expected additional savings are small, the hours can be estimated as a percent
of active period hours.
4. Analyze metering data. Calculate the percent on-time for the metered lighting equipment
for each usage group. Percent on-time is the number of hours the lighting equipment is on
divided by the total number of hours in the metering period. Annual lighting hours are the
percent on-time times 8,760 hours per year less any closed hours such as for holidays.
Separate on-time factors can be developed for day-of-week, month-of-year, and seasonal
timeframes if the metered data capture the full range of operations for the more granular
reporting period.
A. For facilities with constant or variable schedules, the HOU parameter is calculated
as: 8,760 hours per year, less any hours when the facility is closed for holidays,
times the percent-on time.
B. For facilities with seasonal schedules, the HOU parameter is: the hours/year in the
active or operational period, times the percent-on time.
C. The data used in the analysis should represent a typical schedule cycle. For
example, 28 full days for an office space occupied Monday through Friday and
unoccupied on weekends. The hours/year in the active period may vary by usage
group; in schools, for example, office spaces may be active 8,760 hours/year,
while classrooms are only active 6,570 hours/year.
15
5. Evaluation timing usually requires the measurement of operating hours after the
efficiency project has been completed. This process assumes that the operating hours are
unchanged from the baseline period. Thus, HOU baseline and HOU energy-efficient in
Equation 1 have the same value. (Note that will not be the case if the project includes
lighting control measures.)
6. UMP Chapter 3: Commercial and Industrial Lighting Controls Evaluation Protocol
addresses lighting control measures, but Equation 1 can accommodate changes in lighting
operating hours, as would occur in combined lighting equipment and lighting controls
projects, provided measured hours of use data are available for the baseline period. For
example, these data may be available for a facility with an energy management system
with archived trends or if a lighting contractor conducted a metering study before
entering into a performance contract.
6.3.3 Duration of Metering Period
While a metering period of one year would provide the most accurate picture of a facility’s
lighting HOU, economic and customer considerations impose practical limits on the actual
duration. Regulators, program administrators, and customers have limits to their tolerance for
lengthy evaluation periods that delay studies and their results, and that require on-going facility
coordination. Evaluators are thus faced with the questions, “What is the optimal length of a
metering study to obtain acceptable estimates of annual lighting hours of use?” and “How
accurate is this optimal estimate?” A recent study conducted for a Massachusetts large
commercial/industrial program provides some answers (KEMA 2013).
This protocol recommends a one-month minimum metering period based, in part, on the results
of this long-term Massachusetts study.
The study included 12 months of continuous monitoring of lighting systems at 34 large

commercial and industrial sites. Evaluators estimated gross annual savings from each month of
data collected, as well as for each two- and three-month block of data collected. The three-month
results were later compared to the full 12-month study results to determine how well these
shorter metering periods results following completion of the full year of monitoring. The key
findings were that a three-month period of monitoring did a reasonable job of estimating full
year savings as compared to the 12 months of monitoring. Error! Reference source not found.
below presents the HOU and summer coincidence factors by building type from both monitoring
periods. A value greater than 100% means that the full year was higher than the three-month
estimate, or that the three-month data underestimated these parameters. Due to the seasonal
usage, school/university-type buildings were more difficult to annualize and estimate summer
coincidence factors from three months of data.
For most sites, the three-month period included winter/spring months. Fewer daylight hours in
the winter as compared to summer in the northern hemisphere explain why the three-month
results overestimated HOU and CF for this building type as compared to the full 12-month study.
16
Table 3. Comparison of 12-Month and 3-Month HOU and Summer CF by Building Type
Actual
Actual 12- 3-Month 12- 12-
12- Month/ Data - Month Month/
Count of 3-Month Data - Month 3- Estimated Data 3-
Building Annualized Data Month Summer Summer Month
Building Type Type HOU HOU HOU CF CF CF
Manufacturing
(n=6) 6 5,898 5,730 97% 88% 88% 100%
Office (n=5) 5 4,079 3,759 92% 89% 81% 91%
Retail (n=5) 5 5,727 5,473 96% 91% 91% 100%
School/University
(n=4) 4 3,114 2,839 91% 54% 39% 72%
Exercise Center
(n=2) 2 6,541 6,604 101% 89% 91% 102%
Library (n=2) 2 2,129 1,990 93% 58% 58% 101%
Other (n=10) 10 6,054 5,965 99% 81% 79% 98%
All Lighting
Systems (n=34) 34 5,140 4,963 97% 81% 77% 96%
The study also looked at the differences in annual energy savings when using blocks of one-,
two-, and three-month metering periods compared to the 12-month study results, as summarized
in Table 4. The percentages represent how close the annual energy savings would have been
compared to the full 12-month results had data from each specified that period been used to
estimate annual energy savings. Each of the monitoring periods were able to produce annual
energy savings estimates to within 10% of the full 12-month result, which is the basis for the
protocol’s recommendation for a one-month metering minimum. As more data were included,
the annual savings estimates improved. By including three months of data, evaluators could
estimate annual energy savings to within 5% of the full 12-month result regardless of the specific
three-month period.
17
Table 4. Comparison of 1-, 2-, and 3-Month Metering to Actual 12-Month Energy Savings
Percent of Percent of Percent of

Actual Actual Actual
Annual Annual Annual
Monitored One Energy Monitored Two Energy Monitored Three Energy
Month Savings Month Period Savings Month Period Savings
January 99% Jan-Feb 95% Jan-Mar 97%
February 90% Feb-Mar 95% Feb-Apr 96%
March 101% Mar-Apr 98% Mar-May 101%
April 96% Apr-May 102% Apr-Jun 102%
May 108% May-Jun 106% May-Jul 104%
June 103% Jun-Jul 103% Jun-Aug 103%
July 102% Jul-Aug 103% Jul-Sept 102%
August 104% Aug-Sept 103% Aug-Oct 103%
September 101% Sept-Oct 103% Sept-Nov 100%
October 105% Oct-Nov 99% Nov-Jan 97%
November 94% Nov-Dec 96% Oct-Dec 99%
December 98% Dec-Jan 98% Dec-Feb 96%
If either winter or summer peak demand savings are of concern, the protocol recommends
including at least one winter or one summer month in the monitoring period. If both winter and
summer peak demand savings are equally important, the protocol recommends monitoring
during both seasons. Note that monitoring for both seasons extends the study timeline to at least
nine months and increases the overall cost.
6.4 Report M&V and Program Gross Savings

Information collected during the M&V processes is used to calculate M&V project savings, as
follows:
1. Using the results from the last step in verification process, update the inventory HOU
parameters and calculate M&V savings for the sample of projects.
2. Calculate the program gross realization rate, the verified project savings divided by the
reported project savings for the sampled projects.
Equation 10. Program Gross Realization Rate
∑ 𝑘𝑘𝑘𝑘ℎ, 𝑘𝑘𝑘𝑘𝑀𝑀&𝑉𝑉
𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑘𝑘𝑘𝑘ℎ,𝑘𝑘𝑘𝑘 =
∑ 𝑘𝑘𝑘𝑘ℎ, 𝑘𝑘𝑘𝑘𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅
3. Calculate the evaluated program savings, the product of the program realization rate and
the program reported savings.
18
Equation 11. Evaluated Program Savings
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑘𝑘𝑘𝑘ℎ,𝑘𝑘𝑘𝑘 = 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑎𝑎𝑡𝑡𝑡𝑡𝑘𝑘𝑘𝑘ℎ,𝑘𝑘𝑘𝑘 ∙ 𝑘𝑘𝑘𝑘ℎ, 𝑘𝑘𝑘𝑘𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅

The uncertainty and, therefore, the reliability of the program realization rate depend on the
sample size and variance in the findings (described in Chapter 11: Sample Design Cross-Cutting
Protocol). These are usually a function of the confidence and precision targets stipulated by
regulators or administrators, and evaluation budgets. The sample sizes for homogeneous lighting
efficiency programs can range from as few as 12 for an 80/20 confidence/precision target to as
many as 68 (or more) for a 90/10 target, assuming an average coefficient of variation of 0.5.
Higher coefficient of variations will result in larger samples.
The confidence level and its associated precision of the evaluated savings in
Equation 11 should be included when reporting results; for example, 732 MWh/year ±7%
(relative), or 732 MWh/year ±51.2 MWh/year (absolute) at 90% confidence. UMP Chapter 11:
Sample Design Cross-Cutting Protocol describes the calculation of precision for reported
savings. A worked example showing the precision calculations for reported savings from a
lighting project is also available as part of the IPMVP. 4
Precision can only be calculated for the metering period. Care should be taken to ensure that the
metering period is representative of the entire year as described in Section 6.3.3.
6.5 Data Requirements and Sources

This section contains information on the fixture wattage, annual HOU, interactive cooling, and
interactive heating factor parameters found in the algorithm equations. Data requirements are
described in Section 4 Role of the Lighting Program Implementer and Section 5 Role of the
Evaluator, with additional detail in Section 6 Measurement and Verification Plan.
6.5.1 Fixture Wattage

The protocol recommends use of fixture wattage tables, developed and maintained by existing
energy efficiency programs and associated regulatory agencies. The tables list all common
fixture types. Most tables are updated as new fixtures and lighting technologies become
available.
The wattage values are measured according to ANSI standards 5 by research facilities working on
behalf of manufacturers and academic laboratories (CEC 1993).
In the wattage table, each fixture and screw-in bulb is fully described and assigned a unique
identifier. The implementer enters a fixture code into a lighting inventory form, which, if
programmed, can search by a look-up function to show the associated demand. The evaluator
4
IPMVP, Uncertainty Assessment. Anticipated to be available to the public winter of 2018. http://evo-
world.org/en/
5
The ANSI 82.2-2002 test protocol specifies ambient conditions for ballast/lamp combinations in luminaires.
The test is conducted on an open, suspended fixture. Actual fixture wattage will vary, depending on the
installation (suspended, recessed) and housing type. Differences are small—less than 5%.
19
then verifies or corrects the fixture type for the evaluation sample, updating the lighting wattage
values if needed.
Fixture wattage tables do not include records for many LED fixtures and lamps in part because
the tables lag behind this developing technology, but also because LEDs do not lend themselves
to the same clear-cut classifications used for older technologies such as fluorescent or metal
halide. LED fixture codes are needed to classify them by application and cost so they support
market trending and cost-effectiveness analysis.
A solution is to allow users to create LED fixture codes that capture type and wattage using a set
nomenclature. Following is an example of one scheme:
LEDnXXXXww
Where:
n = number of lamps
XXXX = fixture category from Table 5
ww = fixture wattage from manufacturer cut sheets
Table 5. LED Fixture/Lamp Categories
48" Linear Fluorescent Tube Replacement LT

24" Linear Fluorescent Tube Replacement LT
High-Bay Luminaires HBR
Outdoor Pole/Arm-Mounted Luminaires OP/A
Outdoor Wall-Mounted Luminaires OW
Refrigerated Display Case Luminaires RDL
Street Lamp Luminaires ST
Custom C
Thus the fixture code LED1OP/A50 for a 10-lamp, 50-watt outdoor pole/arm-mounted
luminaire.
The protocol recommends adopting a fixture wattage table, used by an established and
recognized lighting efficiency program. As of August 2017, the following sources provide
examples (many others are available in most U.S. regions):
• Massachusetts Program Administrators. (October 2011). Massachusetts Technical

Reference Manual for Estimating Savings from Energy Efficiency Measures (October
2015). http://ma-eeac.org/wordpress/wp-content/uploads/2016-2018-Plan-1.pdf.
• TecMarket Works. (October 2010). New York Standard Approach for Estimating Energy
Savings from Energy Efficiency Programs—Residential, Multi-Family and
Commercial/Industrial Measures. (Version 5) (July 2017). Prepared for the New York
20
Public Service Commission.
http://www3.dps.ny.gov/W/PSCWeb.nsf/96f0fec0b45a3c6485257688006a701a/72c23de
cff52920a85257f1100671bdd/$FILE/TRM%20Version%205%20-
%20January%202018.pdf.
• Database for Energy Efficiency Resources (DEER). Available from the California Public
Utilities Commission at: www.deeresources.com. An exhaustive list of all parameters
driving energy use and savings for a lengthy list of measures. References California
codes and weather zones.
Wattage tables are used by both the implementer and the evaluator. An excerpt from the New
York Standard Approach for Estimating Energy Savings from Energy Efficiency Programs is
included in the Appendix to this protocol as an example.
6.5.2 Hours of Use

The protocol requires the evaluator to measure operating hours for a sample of buildings and
fixtures, as described in Section 6.3 Measurement Process.
This section describes data sources and methods used by the program implementer for estimating
HOU values for individual projects. Accurate estimates of the HOU parameter are needed for the
implementer to reliably estimate and report project and program savings. Accurate reporting by
the implementer also results in more accurate evaluated savings for a given sample size.
The protocol requires program participants to provide estimates of HOU values by usage group
in their lighting inventory forms. The estimate should not be based on the building schedule
alone, although this may inform the estimate. Instead, the protocol recommends participants
develop the HOU values using one of the following sources, with guidance from the program
implementer:
• Lighting schedules in buildings with energy management systems or time clocks

controlling lighting equipment. The project participant should interview the building
manager to verify that the schedules are not overridden. Control schedules (or trend data)
are reliable estimates of true lighting operating hours, but they are normally available
only for larger, newer facilities.
• Interviews with building managers. Building managers are usually familiar with lighting
schedules, and can describe when lights are turned on and off for typical weekdays and
weekends. They may not know about abnormalities such as newly vacant spaces, how
cleaning crews operate lights, or whether lights are actually turned off after hours. The
protocol recommends interviewing two or more people familiar with a facility’s
operation to verify scheduling assumptions.
• Tables of HOU values by building type provided by the program implementer. HOU
values have been developed from impact evaluation and M&V studies for many
commercial and nonresidential buildings. Like wattage tables, HOU tables are maintained
by energy efficiency programs and associated regulatory agencies; sources can be found
using the same references provided for wattage tables. An excerpt from the New York
Standard Approach for Estimating Energy Savings from Energy Efficiency Programs is
included in the Appendix to this protocol as an example of a table of HOU values.
21
Actual operating schedules vary widely for any given building type, and tabulated average values
contain larger variations than values for fixture wattages. Also, tabulated HOU values are given
for entire buildings, not by usage groups within buildings. The protocol requires HOU estimates
be entered into the inventory by usage group, which will vary from the building average. For
these reasons, the protocol recommends use of building-specific lighting operating hours when
these are available, supplemented if necessary by tables of HOU values.
6.5.3 Interactive Cooling and Heating

Energy-efficient lighting equipment contributes less waste heat to building conditioned spaces,
compared to baseline equipment. This results in a reduced cooling and increased heating loads.
This protocol provides two options for calculating interactive cooling and heating effects: an
engineering approach and a stipulated factor approach. A third approach, simulation modeling, is
also an option; however, it tends to be labor-intensive and is usually reserved for large-scale
studies used to quantify stipulated factors. It is unusual to model interactive savings on a project-
by-project basis, and it is not required by the protocol.
The engineering approach requires site-specific estimates of the COP for the cooling and heating
equipment, and the lighting HOU coincident with the cooling or heating equipment operation.
These values can be developed from information gathered during site visits conducted as part of
the verification process.
6.5.3.1 Interactive Cooling and Heating – Stipulated Approach

The stipulated factor approach uses interactive factors—terms IFcool and IFheat in Section 3.1
Algorithms—to account for the additional changes in cooling or heating energy use. Values are
dependent on the type of facility, regional climatic conditions, and cooling and heating
equipment. Guidance is provided below for several common situations.
Interactive cooling effects are generally small for spaces conditioned for human comfort (2% to
6% for cooling in offices in New York City, for example) (TecMarket Works 2010). They are
also highly dependent on HVAC system types and efficiencies. For example, in a large office
building in New York City, the IFcool varies with the equipment: (1) with gas heat and no
economizer, the IFcool is 3.3%, (2) with an economizer, the IFcool is 1.9%, and (3) with
economizer and a variable air volume system, the IFcool is 6.5%. In regions with hot climates
where cooling loads are higher than in New York City, IFcool values will be larger than these
examples. In cooler climates, the values will be lower.
Interactive heating effects are also small for conditioned spaces and will vary with HVAC
system types and efficiencies. For example, in a large office building in New York City, the
IFheat ranges from -2.2% to -1.3% (TecMarket Works 2010). The negative value indicates that
decrease in waste thermal energy from the efficient lighting equipment must be replaced by the
heating system.
Electric efficiency programs often ignore interactive heating effects when territory’s heating
systems are primarily nonelectric; e.g., natural gas or oil. For comprehensive programs with an
all-fuels reporting responsibility, or where electric heating is significant, the increased heating
energy can be included.
22
Interactive factors are usually too small to be measured accurately; instead, they are developed
using computer simulations and the interactive impacts are stipulated. Interactive effects are
available from the same sources as fixture wattages and HOU.
Interactive effects can be significant in cold-temperature conditioned spaces, such as freezers or

refrigerated warehouses. For example, in Pennsylvania, the default interactive cooling factors are
defined by space temperature ranges as follows (Pennsylvania Public Utility Commission 2016):
• Freezer spaces (-20 °F–27 °F) = 50%

• Medium-temperature refrigerated spaces (28 °F–40 °F) = 29%
• High-temperature refrigerated spaces (47 °F–60 °F) = 18%
• Uncooled space (e.g. warehouse with no mechanical cooling) = 0%.
Not all programs estimate, report, and evaluate interactive effects, and the decision is often a
policy choice. Further, because programs are often energy specific (electricity or gas), the effect
on other fuels is sometimes ignored. For example, electric energy efficiency programs might
report interactive electric cooling savings, but omit interactive increases in gas heating energy.
A sample of IFs can be found in the documents listed in Resources.
6.5.3.2 Interactive Cooling and Heating – Engineering Approach

A complete description of the engineering approach to estimating interactive cooling and heating
effects is provided in Section 3.1.1 Energy Savings.
6.5.4 Coincidence Factors (CF)

CFs adjust the change in connected electric load from lighting efficiency projects for electric
peak demand savings. Electric demand savings that occur during utility system peak periods help
lower utility capacity requirements, reduce the load on peak generation equipment that is usually
the costliest to operate, and improve system reliability. The value of peak demand generation is
reflected in rate structures that charge customers for their demand during peak time-of-use
periods.
CFs can range from a high of 1.0 down to 0.0, where 1.0 indicates that 100% of a lighting
project’s change in connected load occurs during the utility peak period. An example is the CF of
1.0 for commercial lighting efficiency projects in New York State (TecMarket Works 2010).
Dawn-to-dusk exterior lighting has a CF of 0.0 when system peaks occur during daylight hours,
which is normal for most utilities. Some programs or utilities may have very specific targets for
the timing of demand reductions. For example, the Con Edison Brooklyn Queens Demand
Management Program targets savings from 9 p.m. to 10 p.m. on weekdays. Use of typical
commercial CF for such a program is not advised.
CFs can be developed from lighting HOU meter data. The CF is the peak period energized
lighting kW as measured by the meter data, divided by the total connected kW for the energy
efficiency lighting project.
23
This protocol recommends using tables of CFs (including any interactive effects from reduced
cooling loads) to report system peak coincident electric demand savings. If regulators or program
administrators require greater reliability for evaluated demand reductions (as would occur for a
program designed to increase capacity reserves), CFs should be developed from metered data.
Like IFs, unique CFs can also be adapted from programs with similar customer and utility
profiles.
A sample of CFs can be found in the documents listed in Resources. CFs are also discussed in
UMP Chapter 10: Peak Demand and Time-Differentiated Energy Savings Cross-Cutting
Protocol.
24
7 Gross Impact Evaluation
Gross impact evaluations entail a detailed review of a random sample of completed projects,
concluding with an independent assessment of their gross savings. The ratio of program-claimed
savings and gross evaluated savings for the projects (the gross realization rate) is used to adjust
claimed savings for all completed projects (the program).
Gross impact evaluations are coordinated in conjunction with program milestones, usually at the
end of a program year or cycle. The evaluation’s subject is the population of all projects
completed up to the milestone.
It is preferable to begin evaluation activity before the program cycle ends, because difficulties
and inaccuracies often occur when collecting data retroactively, particularly in attempts to
backfill missing data, determine baseline data, or deal with poor customer recall of project
details. This may require drawing a preliminary sample before the milestone date and then
adjusting (adding to) the sample after the milestone date.
The evaluator uses the same algorithms and data as the program implementer (subject to review
and site inspections), except that HOU values are based on measurements of actual lighting
operating hours for all projects in the evaluation sample, and lighting inventories (including
baseline and energy efficiency fixture types and counts) are corrected as needed based on on-site
reviews of the sample projects.
The ratio of evaluator savings to program reported savings for the projects in the M&V sample is
the program realization rate. Total reported program savings for the reporting period are then
multiplied by the program realization rate to determine program evaluated savings for the period.
Realization rates can also be developed for facility and customer types if the implementer is
interested in the savings performance of these sub-populations.
7.1 Sample Design

The protocol requires sampling to select:
• Projects from a program database for an impact study

• Inventory lines for deploying light loggers.
Regulators normally prescribe the confidence and precision levels for the sample, or the
implementer may impose them. UMP Chapter 11: Sample Design Cross-Cutting Protocol
describes general sampling procedures and should be consulted when developing evaluation
plans for lighting efficiency programs. The following details pertain specifically to lighting.
The protocol recommends stratified sampling when selecting projects for an impact study
because it usually results in smaller sample sizes as compared to simple random sampling. The
idea behind stratified sampling is to select subpopulations of relatively homogeneous projects
such that the variance within each stratum is smaller than for the population as a whole, as
explained in UMP Chapter 11: Sample Design Cross-Cutting Protocol.
25
A simplified stratified strategy is to rank all projects in the population to be studied by their
reported savings (ranked from largest to smallest) and to define three strata. The top stratum
contains large projects that cumulatively account for 50% of reported savings, and the remaining
projects are grouped into medium strata contributing 30% and small strata contributing 20%.
A more rigorous method is to use a stratified strategy with a customized stratum threshold where
techniques are employed to define strata that minimize the expected variance in their realization
rates, and thereby minimize the sample size. The stratification thresholds are designed to
minimize the variance of a stratified ratio estimator. Stratified ratio estimation is fully explained
in UMP Chapter 11: Sample Design Cross-Cutting Protocol, which should be referenced when
developing sampling plans. Projects may also be stratified by technology types, or by other
characteristics, if known, such as business type or primary space type of the installations.
Light-logger studies also use stratified sampling within projects selected for M&V by selecting
samples of fixtures for metering, with strata defined by usage groups. The desired confidence
and precision interval (typically prescribed with an assumed coefficient of variation of 0.5)
determines the sample size. The Federal Energy Management Program M&V Guidelines
(Federal Energy Management Program 2008) describe a detailed routine for selecting logging
lines.
Oversampling of projects by 30% and of loggers within projects by 10% is recommended to

replace participants that cannot be scheduled for a site visit, and to provide a cushion against lost
or failed loggers in HOU studies.
26
8 Other Evaluation Issues
8.1 Upstream/Midstream Delivery
As upstream programs do not interact with individual customers, they lack the lighting inventory
forms (with associated data) used to estimate savings. Implementers and evaluators can use sales
data, surveys, saturation studies, and other indirect methods to estimate baseline fixture wattages
and facility HOUs. Implementers and evaluators can also draw on incentive and rebate program
data by analyzing baseline fixtures and operating hours associated with fixtures promoted in the
upstream buy-down program, thereby developing savings factors for upstream buy-down
equipment.
Midstream programs are a subset of the upstream family where incentives for qualified lighting
products are paid to distributors selling to contractors and facility managers. Implementers can
leverage the distributor-purchaser relationship to collect key information needed for evaluation.
This information includes the purchased equipment and the site where it will be installed. Many
of the details such as baseline equipment, scheduling, and lighting HOU for the installation site
facility must be collected after the sale, by the evaluator, on a random sampling basis. The
implementer must make assumptions for these and deem them to report savings. Table 6 lists
data required for each project in an evaluation sample, and shows the source of each element.
Table 6. Lighting Data Required by Evaluator for Midstream Programs

Data Source
Field
Implementer Evaluator
Facility Distributor invoice From implementer
Facility type Distributor or utility account Evaluator data gathering
Usage group Not reported Evaluator data gathering
Facility heating (yes/no) Deemed Evaluator data gathering
Facility heating type Deemed Evaluator data gathering
Facility heating fuel Deemed Evaluator data gathering
Facility cooling Deemed Evaluator data gathering
Facility cooling type Deemed Evaluator data gathering
Facility cooling fuel Deemed Evaluator data gathering
Baseline fixture type Deemed based on efficient fixture type Evaluator data gathering
Baseline fixture count Deemed based on efficient fixture count Evaluator data gathering
Baseline fixture watt Deemed based on efficient fixture type Evaluator data gathering
Baseline HOU Deemed based on facility type Evaluator data gathering
Usage group Not reported Evaluator data gathering
Efficient fixture type Distributor invoice From implementer
Efficient fixture count Distributor invoice From implementer
Efficient fixture watt Distributor invoice, qualified products list From implementer
Efficient lighting HOU Deemed, look-up table by facility type Measured by evaluator
IFcool, or COPcool and Deemed, look-up table by facility type Interactive factor for cooling,
HOUcool from look-up table or evaluator
data gathering, optional
IFheat, or COPheat and Deemed, look-up table by facility type Interactive factor for heating,
HOUheat from look-up table or evaluator
data gathering, optional
27
Data Source
Field
Implementer Evaluator
ISR Deemed based on previous studies Evaluator data gathering
kWhsave Calculated using savings algorithms Calculated using savings
algorithms
kW-Peaksave Calculated using savings algorithms Calculated using savings
algorithms
8.1.1 Role of the Implementer in Lighting Midstream Programs

Successful application of this protocol to midstream lighting programs requires collecting
standard data in a prescribed format as part of the implementation process. The protocol further
requires tracking project and program savings estimated on the basis of those standard data.
Distributors are required to submit sales invoices to the implementer. Invoices capture the
efficient lighting product type, specifications, and quantity for each purchase.
Because the implementer does not have contact with end-users who purchase efficient lighting
products through midstream programs, savings estimates must make assumptions for five
baseline variables used in Section 3.1 equations. Following are standard approaches to
determining the baseline assumptions for the five variables used to report program savings. The
assumptions and savings are subject to revision by an evaluation review.
1. Baseline fixture/lamp wattage. Most programs will use a replace-on-burnout baseline

where existing equipment would fail and likely be replaced by a minimally code- or
standard-compliant product, or the most commonly installed product if not regulated.
Thus, the implementer will match each efficient product with a baseline specification
using codes and standards, or market practice. For example, a four-foot LED tube is
assumed to replace a T8 lamp. Another example, high bay LED fixtures can be mapped
to high-intensity discharge (HID) fixtures using lumen bins; a 15,500 to 20,100 high bay
LED replaces a 462-watt (400-watt lamp) metal halide fixture. An example of table-
mapping high bay LED fixtures to baseline HID fixtures is provided in the Appendix. In
midstream programs, the evaluator can determine, based on site visits, if an early
replacement baseline should be used rather than replace-on-burnout.
2. Baseline fixture/lamp quantity. Assume a one-for-one replacement. Baseline quantity is

equal to the efficient product quantity.
3. Annual HOU. Identify the building type where the efficient product is installed and use
look-up table to select HOU values by building type. The building type can be identified
by using business name, address, and utility account number for each sale, or by
requiring distributors to collect it at the point of purchase.
4. Interactive cooling factor, or cooling equipment COP. Use look-up table of deemed
interactive factor or equipment COP by building type.
5. Interactive heating factor, or heating equipment COP. Use look-up table of deemed
interactive factor or equipment COP by building type.
28
The implementer uses the invoice and assumed baseline data to report savings using the
equations in Section 3.1.
The implementer is responsible for ensuring all necessary data are collected to track program
activity and to calculate savings at the project level. The implementer is responsible for
maintaining a program activity record, including anticipated savings by project.
8.1.2 Role of the Evaluator in Lighting Midstream Programs

As described in Section 5, the evaluator’s role in midstream programs is to determine energy
savings resulting from the operation of lighting efficiency programs. The unique feature for
midstream programs is the need to collect more of the baseline and facility data for each project
in the evaluation sample as indicated in Table 6. The steps in this procedure include:
1. Reviewing a sample of completed projects, including conducting on-site M&V activities

2. Calculating a gross savings realization rate (the ratio of evaluator-to-implementer
anticipated savings)
3. Using the realization rate to adjust the implementer-estimated savings.
8.1.2.1 Evaluator Data Requirements
The protocol recommends that the program evaluator develop the same data as the implementer.
However, for midstream programs, the sources for most of the data points will be different for
each party; the implementer is forced to make assumptions for the baseline and facility while the
evaluator contacts each facility in the evaluation sample to verify the actual conditions. The
evaluator must have access to the distributor sales data for the sample of incentivized lighting
products, including information about the facility where they are installed.
8.1.2.2 Evaluator Data Collection Method

Under the protocol, the implementer provides the evaluator with a copy of the program and
project data tracking record for the evaluation review period. That record contains the fields
specified in Table 6. The implementer also provides all records for projects in the evaluation
review sample, including application materials and site contact information.
This protocol recommends the evaluator collect M&V data during site visits conducted for the
sample of evaluation review projects. The data include information about the baseline equipment
using the same techniques as for rebate and incentive programs. The data are used to update
assumptions and values made by the implementer. Table 6 lists data required for each project in
the evaluation sample.
8.2 New Construction

Installed power (kW) savings for new construction projects are calculated by subtracting as-built
building lighting power from the lighting power of a code-compliant alternative, or common
practice. The code-compliant alternative or common practice is the baseline. For jurisdictions
where common practice is more efficient than code, a common practice baseline should be used.
This is occurring in regions where LED lighting is specified in new construction (as opposed to
T8/CFL/Metal Halide technologies). Code defines compliance in terms of lighting power density
(LPD, lighting watts/ft2). Lighting power equals LPD times building area.
29
New construction codes require controls with automatic lighting shutoff, with some exceptions
for safety. The controls reduce the lighting HOU compared to existing facilities. Implementers
can use look-up tables of new construction lighting hours of use that account for controls. These
tables are available from some of the references in Section 11. Evaluators measure HOU using
meters or bulk meter services.
8.3 First Year Versus Lifetime Savings

This protocol provides planners and implementers with a framework for reliable accounting of
energy and demand savings resulting from lighting efficiency programs during the first year of
measure installation.
Savings over the life of a measure may be less than the product of first-year savings and measure
life. The discount results from replacement, degradation, or failure of the efficient equipment.
Lifetime savings are covered further in UMP Chapter 13: Assessing Persistence and Other
Evaluation Issues Cross-Cutting Protocol. However, because lifetime savings for lighting
projects are strongly driven by federal standards and changes in the market, they are discussed
here.
Most T12 lamps do not meet federal efficacy (lumens/watt) standards that went into effect in
July 2012, accelerating a long-term trend toward T8 and T5 lamps and electronic ballasts, or
LED tubes or panels. The effect is that first-year savings for T12 to T8, T5, and LED
replacements can be assumed only for the remaining useful life of T12 equipment, at which point
customers have no choice but to install equipment meeting the new standard.
For retrofit lighting programs, at the time when old equipment would be replaced, there is
effectively a step up in the baseline and a step down in the annual savings for the replacement
equipment. This leads to a dual baseline:
• An initial baseline with full first-year savings for the remaining useful life of the replaced
equipment
• An efficient baseline with reduced savings for the remaining effective useful life of the
efficient equipment.
The protocol methodologies, which specify tracking data for each installation, support the
calculation of lifetime savings (including the use of a dual baseline).
The remaining useful life can be estimated from research studies. It can also be assumed to be a
third of the effective useful life of the baseline equipment.
30
15
Figure 1. Dual baseline 6

Figure courtesy of Regional Technical Forum
8.4 Program Evaluation Elements

Setting the foundation for a successful evaluation of a commercial, industrial, non-residential
lighting program begins early in the program design phase. Implementers support future
evaluations by ensuring data required to conduct an impact study are collected, stored, and
checked for quality. These data include measured and estimated values available from past
studies or equipment tests. Implementers must set data requirements before a program’s launch
to ensure the information required to conduct the research will be available.
6
“Current practice” in the “Savings Period 2” time frame could be codes and standards, or current market
practice for products not covered by codes or standards.
31
9 Looking Forward
Market baseline studies can support gross impact evaluation research of upstream programs by
identifying associations between incentivized products and categories of baseline equipment,
including their demand and energy patterns. Longitudinal market effects studies can supplement
traditional site visit data gathering by characterizing changes over time in lighting equipment
installations.
There is a need to develop hybrid approaches for lighting programs that include both market
baseline and market effects studies in addition to the sampling and site visit model described in
this protocol. As the delivery of lighting energy efficiency changes to include upstream and
midstream models along with traditional downstream (rebate) models, as appears to be occurring
now, there will be greater need for these market data to 1) establish baselines and 2) quantify
gross (and net) savings impacts.
32
10 References
Chantrasrisalai, C.; Fisher, D.E. (2007). Lighting heat gain parameters: Experimental results.
HVAC&R Research 13(2):305-324.
California Public Utilities Commission (CPUC). Database for Energy Efficient Resources
(DEER). Available online: www.deeresources.com
Federal Energy Management Program (FEMP). (2008). M&V Guidelines: Measurement and
Verification for Federal Energy Projects Version 4.0.
https://energy.gov/sites/prod/files/2016/01/f28/mv_guide_4_0.pdf.
KEMA, Inc. (June 21, 2013). Impact Evaluation of 2010 Prescriptive Lighting Installations.
Prepared for the Massachusetts Program Administrators and Massachusetts Energy Efficiency
Advisory Council. http://ma-eeac.org/wordpress/wp-content/uploads/Impact-Evaluation-of-
2010-Prescriptive-Lighting-Installations-Final-Report-6-21-13.pdf
Pennsylvania Public Utility Commission. (2016). Technical Reference Manual, Appendix C.

http://www.puc.state.pa.us/filing_resources/issues_laws_regulations/act_129_information/techni
cal_reference_manual.aspx.
TecMarket Works. (July 2017). New York Standard Approach for Estimating Energy Savings
from Energy Efficiency Programs—Residential, Multi-Family and Commercial/Industrial
Measures. (Version 5). Prepared for the New York Public Service Commission.
http://www3.dps.ny.gov/W/PSCWeb.nsf/96f0fec0b45a3c6485257688006a701a/72c23decff5292
0a85257f1100671bdd/$FILE/TRM%20Version%205%20-%20January%202018.pdf
33
11 Resources
This protocol depends heavily on reliable estimates of fixture wattages and HOU, CF, and IF
values. A rich body of publicly available research provides these data, which can be found in the
resources listed below. Although this is not an exhaustive list, it is representative. Users should
select the references that best match their markets and program needs.
The documents cited below have been produced through regulatory and administrative processes,
and, as they were developed with considerable oversight and review, they are considered reliable
by each sponsoring jurisdiction for their intended applications. HOU, CF, and IF values have
been developed from primary data collected during project M&V reviews or evaluation studies,
or they are based on engineering analysis. Some of these references provide source
documentation.
Fixture wattages are generally based on manufacturers’ ratings, obtained during tests conducted
according to ANSI standards, although this is not well documented in these sources. Fixture
wattages are independent of geographic location. Also, HOU values also tend to be consistent for
non-residential building types regardless of location. The sources cited here can be used for these
parameters in any service territory.
IF and CF parameters, on the other hand, are dependent on local conditions (weather and system
load shape) and users should select carefully so that the referenced values reflect local
conditions. Alternatively, local IF and CF parameters can be developed using computer
simulations and system load shapes for the service territory where they will be used.
The following documents have informed the development of this protocol. Users will find them a
useful starting point in locating the data required to implement the protocol’s savings algorithms
and procedures.
California Energy Commission (CEC).
DOE Advanced Lighting Guidelines.1993
“Database for Energy Efficient Resources (DEER).” California Public Utilities Commission
(CPUC).www.deeresources.com.
Efficiency Valuation Organization (EVO). IPMVP, Uncertainty Assessment. Anticipated to be

available to the public winter of 2018. Free registration required to download: http://evo-
world.org/en/
Massachusetts Program Administrators. (October 2011). Massachusetts Technical Reference

Manual for Estimating Savings from Energy Efficiency Measures (October 2015). http://ma-
eeac.org/wordpress/wp-content/uploads/2016-2018-Plan-1.pdf.
New York Department of Public Service. (2010). New York Standard Approach for Estimating
Energy Savings from Energy Efficiency Programs. Prepared for the New York Department of
Public Service. Albany, New York: New York Department of Public Service, pp. 109-270.
34
Vermont Energy Investment Corporation. (2010). State of Ohio Energy Efficiency Technical
Reference Manual. Prepared for the Public Utilities Commission of Ohio.
35
12 Appendix
Table 7. Example Lighting Inventory Form
Project Name: Example Building Lighting Project #3 Facility Type: Office
Site Name: Example Building Facility Location: NYC
Utility Acct Number(s): XXX-XXXXXXXXXX Facility Square Feet 40,000
Type of Heating Equipment: Gas fired boiler
Type of Cooling Equipment: Rooftop DX Date(s) Survey completed:
Survey completed by (name):
INSTRUCTIONS
PRE-INSTALLATION POST-INSTALLATION
Heating Cooling Baseline Proposed
Area Usage Pre Fixt. Pre Fixt. Pre Pre Existing Post Fixt Post Fixt Post Watts/ Post Proposed kW InterActive InterActive Annual Annual Annual kWh
Description Group ID Heat? Cool? No. Code Watts/Fixt kW/Space Control No. Code Fixt kW/Space Control Saved Factor Factor Hours Hours Saved
Unique Descriptive name Yes or Yes or Number of Code from Value from (Pre Pre- Number of Code from Value from Table (Post Post- (Pre Change in Change in Existing Propsed [(Pre kW/Space *
description of the for the usage No No fixtures Table of Table of Watts/Fixt) * installation fixtures after Table of of Standard Watts/Fixt) * installation kW/Space) - heating energy cooling energy annual annual hours Baseline Annual Hours)
location that group before the Standard Standard (Pre Fixt No.) control the retrofit Standard Fixture Wattages (Post Fixt control device (Post due to lighting due to lighting hours for for the usage - (Post kW/Space *
matches the site retrofit Fixture Fixture device Fixture No.) kW/Space) project project the usage group Proposed Annual
map Wattages Wattages Wattages group Hours)] * (1+Heat-IF)
Room 343 Office Yes Yes 8 2F40SEM 70 0.56 Switch 8 2F25EEE 43 0.34 Switch 0.22 - 0.03 2,500 2,500 558
Room 344 Office Yes Yes 3 2F40SEM 70 0.21 Switch 3 2F25EEE 43 0.13 0.08 - 0.03 2,500 2,500 209
Corridor Floor 3 Hallway Yes Yes 17 1F40SEE 38 0.65 Switch 17 1F25EEE 30 0.51 0.14 - 0.03 3,700 3,700 520
Women RR Flr 3 Restroom Yes Yes 4 110060 60 0.24 Switch 4 1C00185 20 0.08 0.16 - 0.03 3,700 3,700 612
Men RR Flr 3 Restroom Yes Yes 4 110060 60 0.24 Switch 4 1C00185 20 0.08 0.16 - 0.03 3,700 3,700 612
TOTAL 36.00 298.00 1.90 36.00 1.14 156.00 1.14 0.75 2,510
36
Table 8. New York Standard Approach for Estimating Energy Savings
FIXTURE LAMP DESCRIPTION BALLAST Lamp/ WATT/ WATT/
CODE CODE fix LAMP FIXT
F42SSILL F28T8 Fluorescent, (2) 48", Super T-8 lamp, Electronic 2 28 48
Instant Start Ballast, NLO (BF: .85-.95)
F41SSILL/T4 F28T8 Fluorescent, (2) 48", Super T-8 lamp, Electronic 2 28 47
Instant Start Ballast, NLO (BF: .85-.95),
Tandem 4 Lamp Ballast
F42SSILL-R F28T8 Fluorescent, (2) 48", Super T-8 lamp, Electronic 2 28 45
Instant Start Ballast, RLO (BF<0.85)
F41SSILL/T4- F28T8 Fluorescent, (2) 48", Super T-8 lamp, IS Electronic 2 28 44
R Ballast, RLO (BF<0.85), Tandem 4
Lamp Ballast
F42SSILL-H F28T8 Fluorescent, (2) 48", Super T-8 lamp, Electronic 2 28 67
Instant Start Ballast, HLO (BF:.96-2.2)
F42ILL/T4 F32T8 Fluorescent, (2) 48", T-8 lamp, Instant Electronic 2 32 56
Start Ballast, NLO (BF: .85-.95),
F42ILL/T4-R F32T8 Fluorescent, (2) 48", T-8 lamp, Instant Electronic 2 32 51
Start Ballast, RLO (BF<0.85), Tandem
4 Lamp Ballast
F42ILL-H F32T8 Fluorescent, (2) 48", T-8 lamp, Instant Electronic 2 32 65
Start Ballast, HLO (BF:.96-1.1)
F42ILL-R F32T8 Fluorescent, (2) 48", T-8 lamp, Instant Electronic 2 32 52
Start Ballast, RLO (BF<0.85)
F42ILL-V F32T8 Fluorescent, (2) 48", T-8 lamp, Instant Electronic 2 32 79
Start Ballast, VHLO (BF>1.1)
F42LE F32T8 Fluorescent, (2) 48", T-8 lamp Mag-ES 2 32 71
F42LL F32T8 Fluorescent, (2) 48", T-8 lamp, Rapid Electronic 2 32 60
Start Ballast, NLO (BF: .85-.95)
F42LL/T4 F32T8 Fluorescent, (2) 48", T-8 lamp, Rapid Electronic 2 32 59
Start Ballast, NLO (BF: .85-.95),
F42LL/T4-R F32T8 Fluorescent, (2) 48", T-8 lamp, Rapid Electronic 2 32 53
Start Ballast, RLO (BF<0.85), Tandem
4 Lamp Ballast
F42LL-H F32T8 Fluorescent, (2) 48", T-8 lamp, Rapid Electronic 2 32 70
Start Ballast, HLO (BF:.96-1.1)
F42LL-R F32T8 Fluorescent, (2) 48", T-8 lamp, Rapid Electronic 2 32 54
Start Ballast, RLO (BF<0.85)
F42LL-V F32T8 Fluorescent, (2) 48", T-8 lamp, Rapid Electronic 2 32 85
Start Ballast, VHLO (BF>1.1)
F42SE F40T12 Fluorescent, (2) 48", STD lamp Mag-ES 2 40 86
F42GHL F48T5/HO Fluorescent, (2) 48", STD HO T5 lamp Electronic 2 54 117
F42SHS F48T12/HO Fluorescent, (2) 48", STD HO lamp Mag-STD 2 60 145
F42SIL F48T12 Fluorescent, (2) 48", STD IS lamp, Electronic 2 39 74
Electronic ballast
F42SIS F48T12 Fluorescent, (2) 48", STD IS lamp Mag-STD 2 39 103
(New York Department of Public Service 2010)
37
Table 9. New York Standard Approach for Estimating Energy Savings
Facility Type Lighting Facility Type Lighting
Hours Hours
Auto Related 4,056 Manufacturing Facility 2,857
Bakery 2,854 Medical Offices 3,748
Banks 3,748 Motion Picture Theatre 1,954
Church 1,955 Multi-Family (Common Areas) 7,665
College – Cafeteria (1) 2,713 Museum 3,748
College - 2,586 Nursing Homes 5,840
Classes/Administrative
College - Dormitory 3,066 Office (General Office Types) (1) 3,100
Commercial Condos (2) 3,100 Office/Retail 3,748
Convenience Stores 6,376 Parking Garages 4,368
Convention Center 1,954 Parking Lots 4,100
Court House 3,748 Penitentiary 5,477
Dining: Bar Lounge/Leisure 4,182 Performing Arts Theatre 2,586
Dining: Cafeteria / Fast Food 6,456 Police / Fire Stations (24 Hr) 7,665
Dining: Family 4,182 Post Office 3,748
Entertainment 1,952 Pump Stations 1,949
Exercise Center 5,836 Refrigerated Warehouse 2,602
Fast Food Restaurants 6,376 Religious Building 1,955
Fire Station (Unmanned) 1,953 Restaurants 4,182
Food Stores 4,055 Retail 4,057
Gymnasium 2,586 School / University 2,187
Hospitals 7,674 Schools (Jr./Sr. High) 2,187
Hospitals / Health Care 7,666 Schools (Preschool/Elementary) 2,187
Industrial - 1 Shift 2,857 Schools (Technical/Vocational) 2,187
Industrial - 2 Shift 4,730 Small Services 3,750
Industrial - 3 Shift 6,631 Sports Arena 1,954
Laundromats 4,056 Town Hall 3,748
Library 3,748 Transportation 6,456
Light Manufacturers (1) 2,613 Warehouse (Not Refrigerated) 2,602
Lodging (Hotels/Motels) 3,064 Waste Water Treatment Plant 6,631
Mall Concourse 4,833 Workshop 3,750
(New York Department of Public Service 2010)
38
Table 10. Midstream Baseline Wattage, Linear Lamps, and Fixtures; HID Interior and Exterior Fixtures (Pennsylvania Public Utility
Commission 2016)
Efficient Lamp or Fixture Minimum Lumen Maximum Lumen Watts Base Note
3850 6550 189 150-watt HID lamp
6551 9300 215 175-watt HID lamp
9301 11150 241 200-watt HID lamp
11151 12200 295 250-watt HID lamp
Highbay & Lowbay LED Fixture 12201 15550 365 320-watt HID lamp
15551 20100 462 400-watt HID lamp
20101 34700 843 750-watt HID lamp
34701 57250 1090 1000-watt HID lamp
250 4650 133 100-watt HID lamp
4651 7900 215 175-watt HID lamp
Exterior Fixture (Pole, Wall Pack or 7901 11050 295 250-watt HID lamp
Parking Garage)
11051 24700 462 400-watt HID lamp
24701 40750 843 750-watt HID lamp

40751 54650 1090 1,000-watt HID lamp
39

Chapter 2: Commercial and Industrial Lighting Evaluation Protocol

Uploaded by

Copyright:

Available Formats

Chapter 2: Commercial and Industrial Lighting Evaluation Protocol

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 2: Commercial and Industrial Lighting Evaluation Protocol

Uploaded by

Copyright:

Available Formats

Chapter 2: Commercial and

Industrial Lighting Evaluation

This version supersedes the version originally published in April

NREL is a national laboratory of the U.S. Department of Energy

Contract No. DE-AC36-08GO28308

This version supersedes the version originally published in

Prepared under Subcontract No. LGJ-1-11965-01

NREL is a national laboratory of the U.S. Department of Energy

National Renewable Energy Laboratory Subcontract Report

This report is available at no cost from the National Renewable Energy

Available electronically at SciTech Connect http:/www.osti.gov/scitech

Available for a processing fee to U.S. Department of Energy

U.S. Department of Energy

Available for sale to the public, in paper, from:

U.S. Department of Commerce

NREL prints on paper that contains recycled content.

USER AGREES TO INDEMNIFY DOE/NREL/ALLIANCE AND ITS SUBSIDIARIES,

To learn more about the UMP, visit the website, https://energy.gov/eere/about-us/ump-home, or

• Jeff Cropp and M. Sami Khawaja of Cadmus

• Miriam Goldberg of DNV GL

• Dick Spellman of GDS

• Frank Stern of Navigant

• Kevin Warren of Warren Energy.

COP coefficient of performance

DOE U.S. Department of Energy

HGSF heat gain space fraction

HID high-intensity discharge

HOU hours of use

HVAC heating, ventilating, and air conditioning

ISR in-service rate

LED light-emitting diode

LPD lighting power density

M&V measurement and verification

PCF peak coincidence factor

UMP Uniform Methods Project

This chapter has been updated to incorporate the following revisions:

• Added new section for midstream programs.

• Provided additional detail on the recommended duration of metering.

• Added an alternative approach to estimating interactive effects through the use of an

• Applied the controls requirements of International Energy Conservation Code 2012 /

Examples of energy-efficient lighting measures in commercial, industrial, and other

2.1 Common Program Types

2.1.1 Incentive and Rebate

2.1.2 Upstream Buy-Down

2.1.3 Direct Install

2.2 Program Target Markets

Equation 1. Energy Savings = (Baseline-Period Energy Use – Reporting-Period Energy

Equation 2. Demand Savings = (Baseline-Period Demand – Reporting-Period Demand)

3.1.1 Energy Savings

Equation 3. Lighting Electric Energy Savings

Equation 4. Interactive Cooling Energy Savings for Interior Lighting

Equation 5. Interactive Heating Energy Penalty for Interior Lighting

kW Save light = Connected kW savings (kWbase – kWefficient) due to a lighting efficiency

COP cool = Cooling system coefficient of performance

COP heat = Heating system coefficient of performance

Equation 6. Total Annual Energy Savings Due to Lighting Project

3.1.2 Electric Peak Demand Savings

Equation 7. Lighting Electric Peak Demand Savings