Maintenance KPIs: PM compliance, planned work and backlog
Define PM compliance, planned work percentage and backlog weeks, then see how the PM planner and breakdown log dashboards report related figures.
What these KPIs tell you
Three figures give a quick picture of how maintenance work is being managed. PM compliance shows whether scheduled preventive tasks are done on time. Planned work percentage shows how much effort goes into work that was scheduled in advance rather than reactive jobs. Backlog in weeks shows how long the approved, ready-to-schedule work would take at the planned labour capacity. The maintenance KPI calculator computes all three, plus the reactive hours.
They are management indicators, not a measure of maintenance quality. A site can have high compliance on poorly chosen tasks.
The formulas and their definitions
- PM compliance (%) = PM tasks completed on time ÷ PM tasks scheduled × 100
- Planned work (%) = planned maintenance hours ÷ total maintenance hours × 100
- Backlog (weeks) = open backlog hours ÷ weekly planned labour capacity
- Reactive hours = total hours − planned hours
The calculator rejects tasks on time above tasks scheduled, and planned hours above total hours. Definitions matter more than arithmetic: write down what “on time” means (a window around the due date), how rescheduled or cancelled tasks are counted, and what the backlog includes. Backlog normally counts approved work that is ready to schedule and excludes emergency work.
Worked examples: two months compared
Month 1. 120 PM tasks scheduled, 104 completed on time; 310 planned hours out of 520 total; backlog of 640 hours; weekly planned capacity 160 hours.
- PM compliance = 104 ÷ 120 × 100 = 86.6667 %
- Planned work = 310 ÷ 520 × 100 = 59.6154 %
- Backlog = 640 ÷ 160 = 4 weeks
- Reactive hours = 520 − 310 = 210 h
Month 2. 120 tasks scheduled, 111 on time; 360 planned hours of 480; backlog 480 hours at the same 160 hours per week.
| KPI | Month 1 | Month 2 |
|---|---|---|
| PM compliance | 86.6667 % | 92.5 % |
| Planned work | 59.6154 % | 75 % |
| Backlog | 4 weeks | 3 weeks |
| Reactive hours | 210 h | 120 h |
All three moved in the expected direction, but one pair of months is only a snapshot. Compare trends over several periods and ask what changed: crew size, a shutdown, a change in how tasks are counted or a new work order rule. The OEE calculator works the same way for production: 92 %, 88 % and 97 % for availability, performance and quality gives 78.5312 %.
What the PM planner dashboard shows
The PM planner workbook has four tiles at the top of its Dashboard: Active tasks (all rows except cancelled and blank), Overdue (next due date before today), Due soon (inside the due-soon window, 7 days by default) and Completed on time. Panels below show tasks by schedule status, by frequency and by task type, with a ranked list of the ten most overdue tasks.
The Completed on time tile is not the same definition as the calculator’s compliance. It is on-time completions divided by completed tasks that have both a planned and an actual date, where on time means within the grace days (3 by default) of the planned date. Tasks not yet done do not count against it. With the built-in example data, 3 of 4 completed tasks are within the grace window, so the tile shows 75 %. The calculator divides by all scheduled tasks. If you compare numbers from the two, align the definitions first.
What the breakdown log dashboard shows
The breakdown log has eight tiles: Breakdowns logged, Total downtime (h), MTTR (h), Still open, MTBF (h), Availability, Parts cost and Total cost of failures. Panels show events and hours by failure type, downtime and cost by area, and the ten longest breakdowns.
The formulas: MTTR is the average downtime of closed events; MTBF = (period operating hours − total downtime) ÷ number of closed events; availability = 1 − downtime ÷ period operating hours. In the example data, 7 closed events total 30 h downtime and one event is still open. With the period hours at 720: MTTR = 30 ÷ 7 = 4.2857 h, MTBF = 690 ÷ 7 = 98.5714 h and availability = 95.8333 %. The MTBF, MTTR and availability calculator agrees when given 690 operating hours, 7 failures and 30 repair hours. This is inherent availability only, since planned downtime and other delays are not included. See the MTBF and MTTR guide for details.
Common mistakes
- Comparing PM compliance between sites that define “on time” differently.
- Counting deferred or cancelled tasks as complete.
- Using different hours bases for planned and total hours, for example including overtime in one but not the other.
- Including emergency jobs in backlog, which inflates weeks of work.
- Dividing backlog by total headcount hours rather than planned labour capacity.
- Chasing 100 % compliance on tasks that add no value.
- Using a breakdown log period-hours setting that does not match the assets and dates logged.
Checks before you trust the result
- Check that scheduled and on-time counts use the same period and the same task list.
- Check that planned and total hours come from the same work order system and period.
- Check that backlog hours are estimates for approved, ready-to-schedule work.
- Check that open work orders have not been left unclosed, which inflates the backlog and lowers apparent compliance.
- Compare against several periods and look at causes, not a single number.
- Record the definitions used (on-time window, treatment of cancelled tasks, backlog scope) next to the figures, so a later reader can reproduce them.
Frequently asked questions
What is a good PM compliance figure?
This site does not set a target. Your own policy and risk should define it.
Is a high planned work percentage always good?
Not automatically. It shows planning effort, not that the right jobs were chosen.
How many weeks of backlog is right?
It depends on site policy. A value that keeps growing across periods is a signal to review capacity or work approval.
Why does the tile differ from the calculator?
The workbook tile uses completed tasks with a grace window; the calculator uses scheduled tasks. They answer slightly different questions.
Recording data for maintenance KPIs
KPIs are only as reliable as the underlying records. Record the due date, planned date, actual completion date, status and owner for each PM task, and keep cancelled and deferred tasks visible with a reason. Record hours against each work order and tag it planned or reactive. Keep breakdown start and restore times to the minute. Review monthly in the same meeting, using the same definitions, and write the definitions on the dashboard sheet or in the procedure. See also preventive maintenance in Excel.