Meter-Based vs. Calendar-Based PM Scheduling: Which to Use for Each Asset

The two triggers, and what each assumes

Calendar-based scheduling triggers work at fixed time intervals: inspect the compressor monthly, change the filter quarterly, service the boiler annually. It assumes wear accumulates with time — or that time is a close-enough proxy for usage.

Meter-based scheduling triggers work at usage intervals: every 250 run-hours, every 5,000 miles, every 100,000 press strokes, every 500,000 shots on a molding machine. It assumes wear accumulates with use, and it tracks the actual number instead of guessing at it.

Both assumptions are correct — for different failure modes. Rubber hoses crack with age whether the machine runs or not. Spindle bearings wear with revolutions, not birthdays. The mistake isn't choosing one philosophy; it's applying one philosophy to every asset.

When calendar-based is the right call

• Wear driven by time, not use. Seals dry out, coolant degrades, batteries self-discharge, calibrations drift. A backup generator's coolant needs attention on a schedule even if the generator never ran.
• Steady, predictable usage. If a machine runs one shift, five days a week, all year, then 90 days is a usage interval — just measured in a convenient unit. No meter needed.
• Inspections and regulatory intervals. Fire extinguisher checks, hoist inspections, boiler certificates — external requirements are written in calendar time, so schedule them that way.
• Anything where reading a meter costs more than the precision is worth. A $150 exhaust fan doesn't justify a usage-tracking routine. Quarterly glance, done.

Calendar-based is also simply easier to run, which counts for a lot on a small team. It should be your default — with deliberate exceptions.

When meter-based earns its keep

Meter-based scheduling pays off when usage swings, because that's when the calendar lies to you in one of two expensive directions:

• Over-maintaining the lightly used. The backup forklift gets the same quarterly service as the workhorse, despite running a tenth of the hours. That's money and wrench time spent on oil that's still clean.
• Under-maintaining the heavily used. Busy season doubles the line's hours, but the PM still fires every 90 days — so the machine quietly runs twice as far between services exactly when you can least afford a failure.

Classic meter-based candidates: vehicles and mobile equipment on miles or engine hours (forklift PMs at 250 hours are standard practice — see our forklift maintenance guide), compressors on run-hours, injection molding machines on shot counts, stamping presses on stroke counts, and CNC spindles on cutting hours. If your machines already have hour meters, cycle counters, or odometers, most of the tracking hardware exists — the question is only whether anyone reads it. This applies across manufacturing generally: wherever duty cycles vary, the meter knows something the calendar doesn't.

A simple decision rule

For each PM (not each asset — one machine can carry both types), ask two questions:

1. Does this wear come from use or from time? Time → calendar. Use → question two.
2. Does usage vary enough to matter? If the asset's busiest month is more than roughly double its slowest, usage-driven PMs belong on a meter. If usage is steady, the calendar approximates it fine.

Run your asset list through that filter and a typical small plant lands at 70–80% calendar, 20–30% meter — with the meter-based PMs concentrated on the most expensive, hardest-working machines. That's the right shape. If you haven't ranked your assets yet, an ABC criticality pass tells you which machines deserve the extra precision.

Vehicle fleets deserve one extra note: the standard practice is “whichever comes first” — 5,000 miles or six months — because both clocks genuinely apply. Oil oxidizes with time even in a truck that barely moves. Dual triggers give you the protection of both.

Making meter-based actually work

Meter-based scheduling has one operational dependency calendar scheduling doesn't: someone has to record the readings. This is where it fails in practice, and the fix is workflow design, not discipline campaigns.

• Piggyback on something that already happens. Add “record hour meter” as the last line of the daily forklift inspection or the weekly walk-through. A reading captured inside an existing routine costs nothing; a separate meter-reading round dies within a month.
• Weekly is usually enough. You don't need real-time telemetry to run 250-hour PMs. A weekly reading puts you within a few percent of the true interval — dramatically better than a calendar guess.
• Make entry take seconds. A QR code on the machine that opens the asset with a reading field beats transcribing numbers onto a clipboard for later re-entry.
• Watch for readings that stop. A meter-based PM with no fresh readings never triggers — which looks exactly like a machine that doesn't need service. Stale readings deserve the same attention as an overdue work order.

Where to start

Don't convert your whole program at once. Pick the two or three machines where the calendar is clearly lying — the seasonal workhorse, the near-idle backup, the press whose cycle count nobody looks at — and move just their usage-driven PMs to meters. Run that for a quarter, compare service intervals against what the calendar would have done, and let the results argue for the next batch. The goal isn't a philosophically pure program; it's a schedule where every trigger matches the way the asset actually wears.