Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for maintenance managers

Preventive Maintenance Interval Tuning Agent

PM intervals adjusted to match real failure and inspection data

Preventive Maintenance Interval Tuning Agent: what goes in, what the agent does and what you get

What it does

Preventive maintenance intervals usually come from the manual and are never revisited, so some parts fail early while others get serviced for nothing. Each quarter this agent pulls failure history, PM findings and run hours for each asset. It removes failures caused by operating errors, since those are not wear. When parts fail before their PM, it proposes a shorter interval. When PMs keep finding no wear, it proposes a longer one. It checks that each change has enough history behind it, and extends the look-back period if not. It also checks that a proposed change does not conflict with warranty or legal inspection intervals. It drafts a short case for each change with the evidence and expected effect on downtime. The maintenance engineer approves before the maintenance system is changed.

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.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Quarter closes 2 USES A TOOL Pull failures, PM findings and run hours per asset 3 DOES Remove failures caused by operating errors 4 CHECKS THE RESULT Is there enough history to support a change? If not: extend the look-back period by six months andpull again. Back to step 2. 5 DOES Propose shorter or longer intervals with evidence 6 CHECKS THE RESULT Does each change respect warranty and legalinspection limits? If not: cap the interval at the limit and revise thecase. Back to step 5. 7 DOES Draft change cases with expected downtime effect 8 YOU APPROVE Engineer approves changes to the PM plan 9 RESULT Updated PM interval list
Read the steps as a list
  1. Quarter closes
  2. Pull failures, PM findings and run hours per asset
  3. Remove failures caused by operating errors
  4. Is there enough history to support a change?If not: extend the look-back period by six months and pull again. Back to step 2.
  5. Propose shorter or longer intervals with evidence
  6. Does each change respect warranty and legal inspection limits?If not: cap the interval at the limit and revise the case. Back to step 5.
  7. Draft change cases with expected downtime effect
  8. Engineer approves changes to the PM planThe agent waits here for your OK.
  9. Updated PM interval list

How it decides

It shortens an interval when failures happen before PM and lengthens it when the set number of PMs in a row find no wear, but only with enough history.

  • Shorten when two or more failures happen before PM
  • Lengthen when four PMs in a row find no wear
  • Never lengthen intervals on safety-critical assets

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Assets included
  • Minimum history required
  • Safety-critical asset list
  • How many clean PMs justify a longer interval

What keeps you in control

It always asks you first

  • Changing PM intervals in the CMMS

Hard limits

  • Never edits the CMMS directly
  • Never extends safety-critical intervals

It stops when

  • Done: changes approved
  • Stop: failure coding too poor to use; recommend coding fixes

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.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • 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
Get access to this agent

An example run

What happensConveyor gearbox bearings failed twice at about 1,700 hours, with PM set at 2,000. The history check failed because only two failures existed in 12 months. The agent extended the look-back to 18 months, found a third failure at 1,650 hours, and proposed 1,500 hours. Six fan PMs in a row found no wear, so it proposed doubling that interval. The engineer approved both.

More agents for maintenance managers