MUNI-LCC: An Open Toolkit Linking Measured Energy Savings, Life-Cycle Cost, and Federal Performance Reporting for Municipal Energy Projects

Abstract

Energy upgrades and retrofits in small and mid-sized municipalities are funded through federal and state awards. Recipients of federal awards must report project accomplishments against award objectives, provide reasons for objectives not met, and account for any cost overruns or high unit costs pursuant to 2 CFR 200.329. The existing methods for evaluating savings over the projects lifespan are standardized and mostly prepared through the support of a consultant or a licensed software. This financial implication creates a burden for municipalities with budget constraints. MUNI-LCC addresses this challenge: It is an open-source python toolkit that implements Options A and C of the International Performance Measurement and Verification Protocol as defined in the U.S. Department of Energy FEMP M&V Guidelines Version 5.0. It implements life-cycle cost analysis consistent with ASTM E917 through the NIST Handbook 135 framework and integrate the content elements of 2 CFR 200.329 to ensure accurate reporting. MUNI-LCC introduces a distinguishing feature whereby verified quantities directly feed the life-cycle cost study to ensure the net savings, savings-to-investment ratio, and unit cost reported to the sponsor is computed from the metered savings.

Validated using a synthetic worked example with a known answer, a 3PH change-point baseline (CV(RMSE) 7.9%, R2 = 0.988) yields 6,879 therms of avoided gas use against a true 6,629 therms, inside the 90% uncertainty of 2,352 therms. In 2,000 Monte Carlo replications of the same data-generating process, the ASHRAE Guideline 14 uncertainty approximation applied for Option C attains empirical coverage of 90.0% at a nominal 90%, and mean estimation error is 17 therms. Outcome of these tests spurred two changes in version 0.1.1: information-criterion model selection and an uncertainty reported for any sign of savings, without altering the published result of the worked example. Sensitivity scans demonstrates how measured-savings uncertainty and the discount rate propagate to net savings.