AI agent for energy engineers
Efficiency Measure Screening Agent
A ranked, interaction-aware shortlist of measures that meets the client's payback and savings targets.
What it does
After an audit, there are 40 possible measures and a client who wants the best five. Simple paybacks ignore that a lighting upgrade reduces cooling load and that two measures together save less than the sum. This agent estimates savings and cost for each measure with consistent methods and models the interactions, such as lighting with cooling and envelope with heating. It then reranks the list and checks each measure's payback against the client's limit. When the shortlist misses a client goal like total savings, it tries alternative combinations. You get a ranked shortlist with the assumptions. You approve it. Edge case: measures with long life and big non-energy benefits are shown separately from the payback ranking.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- Audit measure list ready
- Load audit data, costs and energy prices
- Estimate savings and cost for each measure with standard methods
- Model interactions between related measures
- Rank measures by net savings and payback
- Does each measure on the shortlist meet the payback limit?If not: remove or rework the measure and re-estimate the interaction effects. Back to step 3.
- Do the combined savings meet the client's goal?If not: try alternative combinations and rerun the interactions. Back to step 4.
- Write assumptions and show non-energy benefits separately
- Engineer approves the shortlistThe agent waits here for your OK.
- Ranked measure shortlist
How it decides
It ranks by net savings after interactions and applies the payback limit. A combination is accepted if total savings meet the goal within the limit.
- Measures that overlap: count savings once and reduce the second by the interaction
- Payback above the client limit: move to a second list
- Measure with life over 20 years: show lifecycle cost beside payback
- Combined savings under the goal: test the next best measure
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Payback limit (default 7 years)
- Energy price assumptions
- Savings goal
- Interaction models used
- Report format
What keeps you in control
It always asks you first
- The final shortlist before it is shared with the client
Hard limits
- Never state savings without listing assumptions
- Never count overlapping savings twice
It stops when
- Done: shortlist approved
- Stop: goal cannot be reached with any combination
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide
