Start a preventive maintenance plan

Many teams know they should do more preventive maintenance, and many have tried. The usual failure is a plan that is too big: hundreds of tasks copied from manuals, scheduled all at once, and abandoned within months because the team cannot keep up with them and the breakdowns at the same time. Start small and grow from results.

1. Pick the machines that hurt most

Do not start with every asset. List the machines where a failure costs the most: those that stop a whole line, create a safety risk, or take long to repair because parts are slow to arrive. Your downtime records and recent work orders will point to them. If you have no records, ask the operators and technicians which machines they worry about.

Pick five to ten. That is enough to make a difference and small enough to do well.

2. Collect what you already know

For each machine, gather:

• the manufacturer's maintenance schedule from the manual,
• the failures of the last year or two, from work orders, logbooks, or people's memory,
• what the experienced technicians already do informally, such as "I always check the chain tension on Mondays".

The informal routines are valuable. They are usually the tasks that have proven themselves on your machines, in your conditions.

3. Write tasks someone can actually do

Turn each item into a task with a clear action and a clear result. Compare:

• "Check gearbox." The technician has to guess what to check and what counts as a problem.
• "Check gearbox oil level in sight glass. Top up with ISO VG 220 if below the mark. Record any visible leaks."

A good task says what to do, what tool or material to use, what normal looks like, and what to record. Add lockout/tagout steps where the task requires opening guards or panels.

Group tasks by machine and interval into checklists, so that one work order covers all the monthly tasks on one machine.

4. Set intervals you can defend

Start from the manufacturer's intervals. Adjust when you have a reason: a dusty environment may need more frequent filter changes; a machine that runs one shift instead of three may need fewer.

Use usage-based intervals, such as operating hours or cycles, where wear depends on use and you have a reliable counter. Use calendar intervals where you do not, or where the wear is time-related, such as rubber ageing.

5. Check the workload

Add up the estimated hours of all tasks per week. Compare that with the time your team really has for planned work after breakdowns. If the plan needs 40 hours a week and the team has 15 spare, the plan will fail. Cut tasks, extend intervals, or start with fewer machines.

Watch your maintenance backlog in the first months. If it grows every week, the plan is too big.

6. Schedule and protect the time

Put the work orders in the schedule, agreed with production. Planned stops are the best slots. The plan only works if PM time is protected; when every breakdown takes priority over every PM, the PMs stop happening, and more breakdowns follow.

7. Review the results

After two or three months, look at what the tasks found:

• Tasks that always find nothing: extend the interval or remove the task.
• Tasks that often find problems: shorten the interval, or consider measuring the condition instead. See condition-based maintenance.
• Failures between PMs: the task may be missing, the interval too long, or the failure not age-related at all.
• Failures right after PM: check how the work was done. Reassembly mistakes are a real source of faults.

This review only works if technicians record findings, not just "done". A reading, a photo of a worn part, or a short note is enough.

8. Expand

When the first machines run on a stable plan and the team keeps up, add the next group. Over a year or two, most critical equipment can be covered this way, with a plan that is based on what your machines need rather than on what a manual says in general.

Common mistakes

• Copying every manufacturer recommendation into the plan without checking workload.
• Tasks that are too vague to do or check.
• No time set aside, so PMs are always the first thing to slip.
• Never reviewing intervals, so the plan does not improve.

For how preventive work fits with other strategies, see Maintenance strategies compared.