The original mistake: planning what should be done, not what can be done
Plans are often built from manufacturer manuals, copying every recommendation. The result can be 1,400 preventive work orders a year for a team that can handle 900.
From there, everything is predictable. The least urgent ones get postponed, then the next ones, and by midyear hundreds of inspections are overdue. Nobody believes in the plan anymore. It did not fail in March: it was broken from the start.
Step 1. Work out your capacity before planning anything
Technicians × productive hours a day × working days = hours a year 6 technicians × 5.5 h × 240 days = 7,920 h
Be careful with productive hours. Out of an 8-hour day, travel, surprises and paperwork leave 5 or 6 hours of real work. Being optimistic here is what breaks the plan.
Now subtract corrective work. If 60% of your workload today is corrective, only 40% is left for preventive: about 3,200 hours. That is your real budget.
Step 2. Rank by criticality and split the budget
Not all equipment deserves the same attention. Give each asset a simple score: impact of a failure, how often it fails and how easy it is to detect. Then rank your assets and split the budget:
| Group | Share of assets | Share of budget | Strategy |
|---|---|---|---|
| Critical | 10–15% | 50% | Full preventive, with readings |
| Important | 25–30% | 35% | Basic preventive |
| Everything else | 55–65% | 15% | Run to failure or a yearly check |
It is a starting point, not a rule. What matters is that the plan fits in the hours you have. And it is the conversation to have with management: with the current team, you cannot maintain everything the same way.
Step 3. Write task lists a technician can follow
A useful task list has fewer than 20 items. Each step can be checked, carries its reference value and follows the order in which you walk around the equipment.
Check the general condition of the electrical panel.
Infrared scan of the power terminals. Record the maximum temperature. Flag it if it goes above 140 °F.
The second one can be done and verified, and it produces a data point you can use next year. To design them well, see how to build a checklist technicians actually fill in.
Step 4. Spread the calendar out
The classic mistake: every monthly task on the 1st and every annual one in January. Spread them by week and by area, so each technician has a steady workload and inspections in the same area fall together.
Leave 20% slack as well. A plan at 100% of capacity cannot absorb a single sick day or a major breakdown.
Step 5. Make postponing leave a trace
If postponing an inspection leaves no trace, it gets postponed until it disappears. At a minimum, a postponed inspection should be recorded with its reason, and the second time it should alert a manager.
It is not about blaming anyone. It is about making postponing a conscious decision, not a habit.
Step 6. Close the loop
Every finding from an inspection should become a scheduled corrective work order, with a date. If it stays as a note on the report, it is lost. And that worn belt will be the breakdown at seven o’clock on a Friday evening.
Also check that every plan actually generates its work orders. A plan assigned to an asset that produces no work order on its due date exists only on paper.
Step 7. Generate a sensible horizon, and change the plan without breaking work in progress
Do not generate a hundred work orders at once. Going up to the end of the contract, or a year at most, is enough, and the rest can be created as they come due. More future work orders do not give you more control: they give you more noise and more rework when the plan changes.
Because it will change. When you modify a running plan, regenerate only the future work orders. The ones already started get reviewed by hand, one by one, because someone has already worked on them.
How to know it is working
- Compliance: preventive work orders completed on time divided by those scheduled. If it stays low month after month, the plan is sized wrong, not executed badly.
- Corrective-to-preventive ratio, as a trend.
- Inspections postponed more than once. If that number grows, the plan is falling apart even if compliance does not show it yet.
A small plan that gets done beats a complete one that does not. Start with critical equipment, measure for three months and expand with whatever capacity is left. If there is none left, you know which conversation is overdue.
Last updated: