NMEC Pre-Qualification Analysis

This Project is aimed at a set of customers that do not have the appropriate SCE EE Offering to accomplish their goals. Specifically, the Project will allow chain-type commercial customers with similar buildings to implement the same set of measures at each site. Identical projects at near-identical sites, with identical NMEC performance calculation procedures, will allow for SCE and CPUC review to be streamlined, as it is applied to “batches” of projects at the program approval stage. After approval, each site will be put on its own individual meter-based, pay-for-performance EE project structure, as opposed to calculating performance across aggregated sets of sites, or “rubber-stamping” savings estimates from one site onto another without measuring.

Technology Description The normalized metered energy consumption (NMEC) program design enables program implementers to offer unique bundles of measures to individual customers, where the savings claimed (and incentives paid) depend on the savings observed at the meter. The California Public Utilities Commission requires that NMEC programs use the custom program and accompanying project review process, where each customer and project is required to submit a unique application, determine an appropriate baseline, and receive approval before proceeding with the project. The proposed pilot program would target chain businesses that are interested in installing the same set of measures at a series of locations that have similar characteristics affecting energy consumption (e.g., building size, existing equipment, operating hours, and business practices). A single application and project plan would cover all branches, with some unique pre-screening criteria to confirm that the branches are sufficiently similar and a good fit for an NMEC program. Project Goal In this initial proof-of-concept study, we will select a modeling approach and pre-screening algorithm to develop baseline models of energy consumption for each chain business (at all eligible branches) that will meet or exceed the NMEC requirements of model fit. We will also test whether a matched comparison group can be extracted from the remaining branches (i.e., non-participants) to estimate net savings in the post-period. Since the pilot has not begun implementation, this study does not cover performance payment calculations or savings claim estimates. Project Findings We looked at three commercial chains, including two grocery chains and one retail chain, with 39 proposed participants, as well as a larger sample of non-participant branches from the same chain businesses. The report provides a summary of the results from the baseline modeling phase of the feasibility study. The model fit tests for each of these chains demonstrated that pooled and segmented baseline models are feasible but may not be a good fit for all types of chains. The individual baseline models met all of the NMEC model fit criteria for the vast majority of participant sites (n=38/39).[1] The one site with a failed individual model had a significant change in its energy consumption during the baseline period, which was identified during our pre-screening for non-routine events. These events will require a follow-up discussion with the customer to explain the event, and then adjust the baseline model prior to program intervention. [1]These NMEC model fit criteria are based on the current SCE site-level NMEC procedures manual and CPUC draft rulebook for population-level NMEC: CV(RMSE)