How do I track clash report exports round to round as new, persisting and resolved with AI?

AI tie-in

Tool: Copilot in Excel or Gemini in Google Sheets, plus Claude or ChatGPT

Claude or ChatGPT turns XML or HTML clash exports into clean rows and writes the formulas; Copilot or Gemini appends each round to a history tab, marks clashes NEW, PERSISTING, REAPPEARED or RESOLVED-CHECK by a stable key, and summarizes open items by test and assignee.

Here is the first part of my clash report export [paste XML or HTML table]. Write a short Python script or Power Query steps that turn the full file into one row per clash with: test name, clash ID, status, assignee, level, grid, element ID 1, element ID 2 and clash point. Explain how to run it on my own file. Don't invent columns that aren't in the export.

Here are the headers and a few sample rows of my History tab [paste]. Write Excel formulas (or a step-by-step Copilot plan) that key each clash on test name plus both element IDs sorted, mark the latest round NEW, PERSISTING or REAPPEARED, list keys from the previous round missing now as RESOLVED-CHECK, and summarize open clashes by test, assignee and rounds open. Do not change statuses.

Human check: Every RESOLVED-CHECK item in the model, since tests and selection sets change; which clashes matter and how they're resolved stay with the trades.

The problem

Track clash report exports round to round as new, persisting and resolved with AI. This is a common, time-consuming task for a bim coordinator.

What this workflow does

Each coordination round produces a fresh clash export, and the meeting question is always the same: which clashes are new, which are still open from last time, which came back, and who is sitting on the oldest ones. Keep one history workbook, append each round's export with its date, and have Copilot in Excel or Gemini in Google Sheets classify every clash against the previous round by a stable key and summarize counts by test, level and assignee. Claude or ChatGPT can turn a messy XML or HTML export into clean rows and write the formulas. Whether a clash is real, and how it is resolved, stays with the trades in the meeting.

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

• Claude (Anthropic)
• ChatGPT
• Microsoft Excel
• Microsoft Copilot
• Google Sheets
• Autodesk Navisworks

Follow the steps

1. Export each round's clash results from your coordination tool in the same format every time (XML or tabular HTML), with the test name, clash name or ID, status, assignee, level or grid, both element IDs and the clash point.
2. If the export doesn't open cleanly in a spreadsheet, have Claude or ChatGPT write a short script or Power Query steps that turn it into one row per clash with those columns. You run it on your own file.
3. Append the rows to a History tab with a Round date column. Agree on a match key, for example test name plus the two element IDs sorted, so a rerun test with renumbered clashes still matches.
4. Have Copilot in Excel or Gemini in Google Sheets classify each clash in the latest round against the previous one as NEW, PERSISTING or REAPPEARED, and list keys from the previous round that are missing now as RESOLVED for review.
5. Build the meeting summary: counts by test and status, open clashes by assignee with rounds open, and the oldest persisting items first.
6. Check every RESOLVED item in the model before reporting it; a clash can vanish because a test or selection set changed, not because it was fixed.

Try a small example

Replace the bracketed fields with your own project details. No tested sample run is published for this workflow yet.

Fill in the bracketed [fields] in the text below with your project data.

Here are the headers and a few sample rows of my History tab [paste: round date, test, clash ID, status, assignee, level, grid, element ID 1, element ID 2]. Write Excel formulas (or a step-by-step Copilot plan) that build a match key from test name plus the two element IDs in sorted order, then for the latest round mark each clash NEW, PERSISTING or REAPPEARED against earlier rounds, list keys present in the previous round but missing now as RESOLVED-CHECK, and summarize open clashes by test, assignee and number of rounds open. Use only my data and do not change any statuses.

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 tracks what the exports say from round to round. It does not decide whether a clash matters, which trade moves, or whether a missing clash was actually fixed. Every RESOLVED-CHECK item is verified in the model, and resolutions stay with the trades and the design team.
• 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

Where to go next

• How do I run clash and issue coordination in one BIM platform instead of email?
• How do I renumber Revit viewport detail numbers by their position on the sheet?
• How do I triage Revit warnings into a prioritized cleanup list before model coordination with AI?
• All BIM Coordinator workflows
• All Model QA and coordination workflows

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.

• Claude (Anthropic) (vendor site)
• ChatGPT (vendor site)
• Microsoft Excel (vendor site)
• Microsoft Copilot (vendor site)
• Google Sheets (vendor site)
• Autodesk Navisworks (vendor site)