Meval Methodology

The Meval method for estimating and verifying energy savings from HVAC system upgrades.

Energy savings are not just a prediction problem

Traditional M&V often estimates savings by predicting what energy consumption would have been after the retrofit. That approach can become fragile when the building is used differently after the upgrade.

Meval instead treats M&V as a comparison between matching states and conditions. The method uses both pre- and post-retrofit data to infer the activity levels that act as context for HVAC operation.

Given observed weather and inferred activity, Meval estimates the energy that would have occurred under pre-retrofit efficiency and post-retrofit operating conditions. This helps isolate retrofit-attributable savings from unrelated changes in how the building is used.

From an IPMVP perspective, this extends Option B and Option C style analysis with an additional conditioning variable for operational intensity.

The Meval method enables M&V practitioners to deal with:

Weather effects

Normalize consumption against temperature and weather so a mild season is not mistaken for a successful retrofit.

Occupancy changes

Account for shifts in how a building is used so occupancy swings do not inflate or hide true savings.

Operational shifts

Detect changes in schedules, setpoints, and control strategies that affect energy use independently of the upgrade.

Non-routine events

Identify structural change events and adjust for them without a full, costly re-baseline.

A four-step process for defensible savings verification

The workflow is designed to explain why energy consumption changed, not just predict what the next meter reading should be.

Step 01

Model pre-retrofit behavior

We fit a physics-informed baseline model that captures how energy consumption responds to weather, operating conditions, thermal inertia, free cooling, baseload variation, and capacity effects.

Step 02

Estimate operating activity

We infer hidden activity levels that explain variation around the physical model, using calendar structure to distinguish meaningful operational intensity from noise.

Step 03

Transform the model after the upgrade

We use post-retrofit data to identify which parts of the energy response changed, while requiring stronger evidence before updating parameters that should remain stable.

Step 04

Compare matching conditions

Savings are estimated by comparing pre- and post-retrofit performance under aligned weather and activity conditions, separating true efficiency gains from unrelated operational changes.