A preventive maintenance schedule is a plan that lists every PM task across your assets: what needs to be done, on which machine, at what interval, and by whom, so the work happens on time instead of depending on someone remembering it. It's the document that turns a PM program from an intention into an operation.
Most plants have some kind of PM schedule in someone's head. The maintenance manager knows the compressor needs a service every six months. The line technician knows the hydraulic press filter gets changed every quarter. But "someone knows" isn't a system. It's a dependency on one person's memory, and when that person is on leave, out sick, or gone, the PM stops.
A written schedule makes that knowledge belong to the plant. This guide covers the three ways to schedule PM, how to build a schedule from scratch in six steps, a sample layout you can adapt, and the four reasons most PM schedules fail before the machines do.
Types of PM scheduling
There are three different ways to decide when a PM is due, and most factories use at least two of them.
Time-based (calendar)
The simplest and most common approach. PMs run at fixed calendar intervals, every 30 days, every quarter, every six months, regardless of how much the machine actually ran in that period.
Calendar scheduling works well for parts that age with time rather than use: seals that dry out, lubricants that oxidize, filters that collect ambient dust even when the machine is idle. It's also the easiest to manage. The schedule is predictable, anyone can read it, and it needs no instrumentation.
The limitation is obvious. A machine that ran 2,000 hours this month and 100 hours last month gets the same PM interval on a calendar schedule. For assets with very uneven utilization, that means either over-maintaining or under-maintaining them.
Usage-based (meter-based)
Usage-based PMs trigger when the asset crosses a threshold: every 500 operating hours, every 10,000 cycles, every so many miles on a vehicle. The interval follows actual wear, not the calendar.
This is more accurate where wear is driven by use: cutting tools, bearings under continuous load, hydraulic pumps, engines. It also avoids pointless work. A PM that's "due" on the calendar when the machine has run 40 hours since its last service usually isn't due in any meaningful sense.
The catch is data. Someone has to record the meter reading at each PM, or the machine has to report its run hours to your maintenance system. Without that input, usage-based scheduling can't work.
Condition-based
Condition-based PMs trigger when a measurement crosses a limit: vibration on a rotating machine above its alarm level, oil contamination above a set particle count, insulation resistance below a minimum.
You do the work when the machine actually needs it. Machines running cleanly aren't disturbed, and machines showing early signs of trouble get attention before they fail.
It needs monitoring in place: sensors or routine readings, plus someone (or something) reviewing them against limits. That's a real investment in setup and upkeep, so most plants apply condition-based maintenance to their most critical or most expensive-to-fail assets and keep calendar scheduling for everything else.
| Scheduling type | Trigger | Suits | Complexity |
|---|---|---|---|
| Time-based (calendar) | Fixed date or interval | Parts that age with time (seals, lubricants, filters) | Low |
| Usage-based (meter) | Operating hours, cycles or distance | Parts that wear with use (bearings, cutting tools, engines) | Medium |
| Condition-based | A measured value crosses a limit | Critical or expensive-to-fail assets | High |
For how these fit alongside breakdown and corrective work, see types of maintenance.
How to build a PM schedule, step by step
Step 1: List your assets and rank criticality
Before you can schedule PMs, you need to know which assets get them. Start by ranking criticality.
Ask three questions about each asset. If it stops, does the line stop? Does it cause quality defects as it degrades? Is it safety-critical?
A yes to any of those makes it an A-class (critical) asset. Assets that slow you down but don't stop the line are B-class. Everything else is C-class.
Start the schedule with A-class assets only. B and C assets matter, but a program that tries to cover 150 assets on day one collapses under its own weight. Get the A-class schedule running reliably, then add the rest once the habit is there.
Step 2: Identify how each asset fails
For each A-class asset, list the two or three ways it actually fails. Not in theory, in practice: a seized bearing because lubrication was missed, a clogged filter that starved the hydraulic circuit, a worn seal that let contamination in.
If you have breakdown history, pull it. Failure modes that show up more than once belong on the schedule, with a task aimed at each one. If you don't have history yet, the OEM manual is the place to start. The people who wrote it know how the machine fails.
This step is what makes the difference. A schedule built without failure modes in mind is a list of activities. A schedule built around them is a prevention program.
Step 3: Define the PM tasks
For each failure mode, define the work that prevents it. Lubrication prevents bearing seizure, so the task is lubricating specific points, with a specific grade, at a set interval. Filter replacement prevents flow restriction, so the task is inspecting and replacing the filter on a set schedule.
Every task should state the action, the interval, the expected duration, and the parts or consumables needed. A vague task like "check lubrication," with no points, grade or quantity, gets done differently by every technician who touches it.
The task content itself lives on the checklist. Our preventive maintenance checklist guide covers how to write one.
Step 4: Assign an owner to every task
Every task on the schedule needs a named owner: a technician, a team, or an outside provider. A task without an owner slips. Nobody is being careless. "Someone will do it" just means everyone assumes it's someone else.
If a PM needs a specialist, such as OEM service, a vendor under an annual maintenance contract, or a calibration contractor, it goes on the schedule too, with the contract reference and vendor contact. Vendor PMs are the ones most likely to fall through the gap between "our team" and "their service team."
Operator checks belong on the schedule as well. Daily and weekly checks done by the machine operator (autonomous maintenance, in TPM terms) should be listed against the operator role, so the whole PM picture is visible in one place.
Step 5: Set up the schedule
For each asset and task, set the first due date, the interval, and the assignee.
In a CMMS, the system creates the work order on the due date, assigns it, and attaches the checklist. Nobody has to remember anything, because the system dispatches the job.
In a spreadsheet, all of that falls on a person. Someone has to open the sheet every week, see what's due, create the jobs, and chase them to completion. It can work, but it's a system that runs on memory, which is exactly what a PM schedule is supposed to replace. Our guide to work order management covers that trade-off in more detail.
Step 6: Build in a review
A schedule written in January and never touched will be wrong by June. Machines get modified, failure modes change, and some intervals turn out too short while others turn out too long.
Put a quarterly review in place from the start, and ask four questions. Which PMs are being done on time? Which keep slipping, and why? Which assets still break down despite their PMs? And when the same failure recurs after three completed PMs in a row, is the task actually addressing that failure? If it isn't, fix the checklist, not just the interval.
That review loop is what makes a schedule improve over time instead of just repeating itself.
A sample PM schedule layout
The table below shows the structure, not a real schedule for these machines. Your intervals, tasks and owners will depend on your operating conditions, OEM specs and site environment, so build the content from your manuals and breakdown history.
| Asset | PM task | Interval | Duration | Owner |
|---|---|---|---|---|
| Hydraulic press P-3 | Check hydraulic fluid level and condition | Weekly | 15 min | Operator |
| Hydraulic press P-3 | Replace hydraulic filter | Every 3 months | 45 min | Maintenance team |
| Air compressor C-1 | Clean air intake filter | Monthly | 20 min | Maintenance team |
| Air compressor C-1 | Full service (valves, belts, oil) | Every 6 months | 4 hrs | OEM service contract |
| CNC lathe L-7 | Lubricate spindle, check coolant | Weekly | 10 min | Operator |
| CNC lathe L-7 | Inspect chuck and collets | Monthly | 30 min | Maintenance team |
| CNC lathe L-7 | Full geometry check and calibration | Annual | 1 day | OEM service |
| Conveyor CB-2 | Check belt tension and tracking | Monthly | 20 min | Maintenance team |
Add a next due column when you use it. In a spreadsheet that's a date you update after each completion. In a CMMS it's calculated from the recurrence rule.
Three things worth noticing in that layout:
Operators own the short, frequent checks. Fluid levels, visual inspections and routine lubrication take 10 to 15 minutes and catch problems early. Keeping them on the same schedule as the maintenance team's tasks shows who's doing what, and when.
Outside service is listed explicitly. The compressor service and the lathe calibration are on the schedule, so it's visible that a vendor has to be booked, and the due date is tracked instead of left until the machine needs it.
Big tasks sit next to small ones. A one-day calibration and a ten-minute lube check are on the same page. That's what makes it possible to see whether the week's workload actually fits the team.
Why PM schedules fail
A well-designed schedule can still fail once it's running. These are the four most common reasons.
Too many tasks for the team. Two hundred PM tasks across 50 machines, for a team of five. Completion falls below half, and the schedule becomes a wish list. It looks like the program failed, but the real problem is the schedule was never sized to the team. Add up the planned maintenance hours per week and compare them with the hours your team actually has. If they don't balance, you aren't behind on PMs. You have too many PMs. Start with A-class assets and a credible task list, and grow it as capacity allows.
Tasks that aren't specific. "Lubricate," with no points, grade or quantity. "Inspect belts," with no pass or fail criterion. The record says the PM was done and the machine says it wasn't. Being specific isn't pedantry. It's the difference between a PM that works and one that only looks done.
No dispatch. A schedule in a spreadsheet only fires when someone opens it. By Monday's meeting three PMs are overdue, and one of those machines has already broken down. The schedule was right, but nothing pushed the work out. Dispatch, through a CMMS or a disciplined weekly routine, is what makes a schedule a system instead of a document.
No feedback from breakdowns. PMs are done on time and the same machine keeps failing. Either the tasks don't address the real failure mode or the interval is too long. The answer is to look at the cause recorded on the breakdown work orders and revise the PM, including the content, not just the interval.
One more thing: a PM program doesn't need to be big to count. Five critical machines, three well-written tasks each, dispatched every week without fail, is a real program. It beats a 200-line schedule nobody follows.
Where a PM schedule connects to maintenance software
A schedule only works if the work actually reaches the people doing it, on time, with the right steps.
In MachDatum, you set up a preventive maintenance plan once (the asset, the task, the recurrence and the assignee) and the work order is created on the due date with the plan's checklist already attached. The assigned technician is notified, overdue jobs are flagged, and every completed PM is recorded against the asset's history, so next quarter's review starts from real data instead of a spreadsheet nobody updated.
If your PM schedule lives in a spreadsheet and a few people's heads, see how MachDatum CMMS handles preventive maintenance.
