Pull your preventive maintenance list for one critical system and ask one question about every task on it: what failure does this prevent or find? You’ll be surprised how often nobody in the room can answer.
I spent years doing reliability-centered maintenance, and that question is the heart of it. It’s a simple question. Most PM programs have never been asked it.
How PM programs grow
PM programs get built one task at a time. The OEM manual recommended it. Something failed in 2014, so we added an inspection. A new supervisor liked monthly checks. An auditor asked for it. Tasks get added all the time. Almost nobody ever takes one away.
What you end up with is a PM program full of work that has nothing to do with how the equipment fails. Your crews spend hours on tasks that don’t prevent anything, and the failures that matter get no attention at all. Then the backlog grows, PM compliance slips, and everyone decides the answer is more people. That’s the wrong answer.
Ask what each task prevents
Take every PM task on a critical asset and make it answer these questions:
- What failure mode does it address? Name it. Bearing wear, seal leakage, insulation breakdown. If you can’t name one, that’s your first finding.
- Does the task prevent that failure or detect it early? Both are fine. Doing neither is not.
- Is the interval based on anything? Failure history, condition data or an OEM basis you still believe. “We’ve always done it monthly” is not a basis.
- Would condition monitoring do it better? Vibration, thermography or oil analysis often catch the problem earlier for less labor.
- What happens if we stop? For low-consequence assets, running to failure can be the right call. Just make sure somebody decided it on purpose.
Each task ends up in one of five places: keep it, change the interval, switch it to condition-based, run to failure by decision, or delete it. Whatever you decide, write down why. The next person will want to know.
A PM task that doesn’t prevent or find a failure is labor you’re spending to feel safe.
Start with criticality
Don’t try to review everything. You’ll run out of steam. Start with the assets where failure hurts the most: safety, service, compliance, cost. A careful review of your top systems will do more for you than a quick pass over the whole plant. Put the labor you free up on the critical work that’s been sitting in your backlog.
What to do this quarter
- Pick one critical asset class. Pumps, chillers, switchgear, whatever keeps you up at night.
- Pull every PM task and its interval along with two years of corrective work order history for those assets.
- Spend one morning with a planner and your most experienced technician and put every task into one of the five buckets.
- Update the job plans in the EAM and track what the labor savings go to.
If you had to cut your PM hours by a fifth tomorrow, would you know which tasks to cut? Or would you be guessing?
