How do I cross-check mechanical equipment power data against the electrical panel schedules by tag with AI?
A lot of preventable change orders on MEP jobs start the same way: the mechanical schedule changes a unit's voltage, phase, minimum circuit ampacity (MCA) or maximum overcurrent protection (MOCP) after the electrical side already circuited it, or a tag on the mechanical plans never shows up on a panel schedule at all. Export or extract both sides into one workbook keyed on the equipment tag, and have Copilot in Excel or Gemini in Google Sheets line them up and flag every tag that is missing on one side or whose power data disagree. Claude or ChatGPT can read schedules off PDF sheets into rows when there is no model export. Sizing the conductors and breakers, and deciding which side is right, stay with the engineers.
MEP Engineer · MEP · Claude · ChatGPT · Microsoft Excel · Microsoft Copilot · Google Sheets
Content updated · Tool links listed below; vendor features and pricing change, so confirm with the vendor
Short answer
Use Copilot in Excel or Gemini in Google Sheets, plus Claude or ChatGPT: Claude or ChatGPT reads mechanical equipment schedules and electrical panel schedules off PDF sheets into rows by tag; Copilot or Gemini joins the two on the tag and flags tags missing on one side and voltage, phase, circuit count or breaker-versus-MOCP disagreements for the engineers to resolve. A lot of preventable change orders on MEP jobs start the same way: the mechanical schedule changes a unit's voltage, phase, minimum circuit ampacity (MCA) or maximum overcurrent protection (MOCP) after the electrical side already circuited it, or a tag on the mechanical plans never shows up on a panel schedule at all. Export or extract both sides into one workbook keyed on the equipment tag, and have Copilot in Excel or Gemini in Google Sheets line them up and flag every tag that is missing on one side or whose power data disagree. Claude or ChatGPT can read schedules off PDF sheets into rows when there is no model export. Sizing the conductors and breakers, and deciding which side is right, stay with the engineers. A person still checks: every flag and every proposed tag match with the mechanical and electrical engineers, and the submitted equipment data against both sets once it arrives.
AI tie-in
Tool: Copilot in Excel or Gemini in Google Sheets, plus Claude or ChatGPT
Claude or ChatGPT reads mechanical equipment schedules and electrical panel schedules off PDF sheets into rows by tag; Copilot or Gemini joins the two on the tag and flags tags missing on one side and voltage, phase, circuit count or breaker-versus-MOCP disagreements for the engineers to resolve.
Prompt 1
Read these mechanical equipment schedule sheets [upload]. Build a table with one row per tag and circuit: tag exactly as drawn, description, voltage, phase, MCA, MOCP, number of circuits, disconnect, furnished by and installed by. Copy values exactly. Write UNREADABLE for anything you cannot read and NOT SHOWN for anything missing. Do not convert units or fill gaps.
Prompt 2
Here are the headers and a few sample rows of my Mech tab [paste] and my Elec tab [paste]. Write Excel formulas (or a step-by-step Copilot plan) that join them on tag, list tags found on only one tab, and flag voltage, phase or circuit count mismatches and breakers that differ from the MOCP shown. Treat near-identical tags as possible matches for me to confirm. Do not size conductors or breakers or say which value is right.
Human check: Every flag and every proposed tag match with the mechanical and electrical engineers, and the submitted equipment data against both sets once it arrives.
The problem
Cross-check mechanical equipment power data against the electrical panel schedules by tag with AI. This is a common, time-consuming task for a mep engineer.
What this workflow does
A lot of preventable change orders on MEP jobs start the same way: the mechanical schedule changes a unit's voltage, phase, minimum circuit ampacity (MCA) or maximum overcurrent protection (MOCP) after the electrical side already circuited it, or a tag on the mechanical plans never shows up on a panel schedule at all. Export or extract both sides into one workbook keyed on the equipment tag, and have Copilot in Excel or Gemini in Google Sheets line them up and flag every tag that is missing on one side or whose power data disagree. Claude or ChatGPT can read schedules off PDF sheets into rows when there is no model export. Sizing the conductors and breakers, and deciding which side is right, stay with the engineers.
How it works: AI drafts or analyzes. A person operates the tools, checks the work, and takes action.
Your checkpoint: The workbook only shows where the two sets of documents disagree on paper. It does not size conductors or overcurrent protection, check the manufacturer's nameplate data, or decide which discipline's value governs. Every flag goes to the responsible mechanical and electrical engineers, and the submitted equipment data is checked against both once it arrives.
Before you start
- Access to the named tool; confirm license, version and cost with the vendor.
- A small, non-confidential sample to try the workflow on first.
- A qualified person to review the output before it is used.
Tools and access
Follow the steps
- Pick one current issue set for both disciplines and note its date on a Settings tab, so every comparison is against the same drawings.
- Build a Mech tab with one row per tagged unit and circuit: tag exactly as drawn, description, voltage, phase, MCA, MOCP, number of circuits, disconnect type and who furnishes and installs it per the documents. Use a model schedule export if you have one; otherwise have Claude or ChatGPT read the schedule sheets into the columns, copying values exactly and writing UNREADABLE instead of guessing.
- Build an Elec tab the same way from the panel schedules and equipment connection schedule: tag or load name, panel, circuit numbers, voltage, phase, breaker size, conductor callout and disconnect noted.
- Have Claude or ChatGPT propose matches where a panel schedule uses a load name instead of the tag (for example a unit described by area), marking every proposed match REVIEW. You confirm or reject each one.
- Have Copilot in Excel or Gemini in Google Sheets join the tabs on the tag and flag: tags on one side only, voltage or phase mismatch, a breaker that differs from the MOCP shown, circuit count mismatch, and disconnect scope that differs between the two sets.
- Send the exceptions list to the mechanical and electrical designers, record each answer and the revised value, and rerun the check at every issue set.
Try a small example
Replace the bracketed fields with your own project details. No tested sample run is published for this workflow yet.
Sample input
Fill in the bracketed [fields] in the text below with your project data.
Starter prompt
Here are the headers and a few sample rows of my Mech tab [paste: tag, description, voltage, phase, MCA, MOCP, circuits, disconnect, furnished by, installed by] and my Elec tab [paste: tag or load name, panel, circuits, voltage, phase, breaker, conductor, disconnect]. Write Excel formulas (or a step-by-step Copilot plan) that join the two on tag, list tags that appear on only one tab, and flag rows where voltage, phase or circuit count differ, or where the breaker does not equal the MOCP shown on the Mech tab. Treat AHU-1, AHU-01 and AHU 1 as different tags and list them as possible matches for me to confirm. Use only my data; do not size conductors or breakers and do not say which value is correct.
Expected output
Not published yet. Treat any output as a draft and check it against the source documents.
Check the result
- Every value, citation or quantity traces back to a source document.
- Missing information is marked as missing, not filled in.
Limits and recovery
- The workbook only shows where the two sets of documents disagree on paper. It does not size conductors or overcurrent protection, check the manufacturer's nameplate data, or decide which discipline's value governs. Every flag goes to the responsible mechanical and electrical engineers, and the submitted equipment data is checked against both once it arrives.
- Check your organization’s data policy before sharing project files with third-party tools.
Stop or switch when: The source cannot be verified, permissions are missing, or the result contains unsupported facts.
Frequently asked questions
How does AI help with this?
Copilot in Excel or Gemini in Google Sheets, plus Claude or ChatGPT: Claude or ChatGPT reads mechanical equipment schedules and electrical panel schedules off PDF sheets into rows by tag; Copilot or Gemini joins the two on the tag and flags tags missing on one side and voltage, phase, circuit count or breaker-versus-MOCP disagreements for the engineers to resolve. A person still checks: Every flag and every proposed tag match with the mechanical and electrical engineers, and the submitted equipment data against both sets once it arrives.
How do I how do I cross-check mechanical equipment power data against the electrical panel schedules by tag with AI??
Pick one current issue set for both disciplines and note its date on a Settings tab, so every comparison is against the same drawings. Build a Mech tab with one row per tagged unit and circuit: tag exactly as drawn, description, voltage, phase, MCA, MOCP, number of circuits, disconnect type and who furnishes and installs it per the documents. Use a model schedule export if you have one; otherwise have Claude or ChatGPT read the schedule sheets into the columns, copying values exactly and writing UNREADABLE instead of guessing. Build an Elec tab the same way from the panel schedules and equipment connection schedule: tag or load name, panel, circuit numbers, voltage, phase, breaker size, conductor callout and disconnect noted. The full step list is above.
Does this use an AI agent or regular software?
It is labeled AI assists: AI drafts or analyzes. A person operates the tools, checks the work, and takes action.
What tools do I need?
Claude, ChatGPT, Microsoft Excel, Microsoft Copilot, Google Sheets. Confirm license, version and cost with the vendor, and try it on a small non-confidential sample first.
Has this workflow been tested?
Workflow guide · not independently tested. Written from vendor documentation and practitioner use. Labels apply to the steps described here, not every feature of a product. Links were checked when the page was written.
Evidence and maintenance
Written from vendor documentation and practitioner use. Labels apply to the steps described here, not every feature of a product. Links were checked when the page was written.