A maintenance KPI is only useful if someone changes a decision because of it. Most maintenance dashboards track 20+ metrics, and most of them sit there unmoving while the ones that matter go un-examined. This guide separates the 8 KPIs that drive actual operational change from the 12 that create the illusion of measurement.

You will find the formula, the target range, and the specific decision each KPI should trigger.

Availability KPIs (Top 3)

These three should be on every maintenance leader's dashboard:

• 1. MTBF (Mean Time Between Failures). Formula: operating hours / failure count. Target: top-quartile for equipment class. Decision: when MTBF trends down on a specific asset, schedule RCA.
• 2. MTTR (Mean Time To Repair). Formula: total repair time / repair count. Target: industry median or better. Decision: when MTTR is above median, invest in diagnostic tools or training.
• 3. Availability. Formula: MTBF / (MTBF + MTTR). Target: 95%+ for most manufacturing. Decision: availability < 90% means you are in firefighting mode; pause improvement projects and focus on the top failures.

Work Execution KPIs (Next 3)

• 4. PM Compliance. Formula: PMs completed on time / PMs scheduled. Target: 90%+. Decision: below 80%, add resources or reduce PM scope.
• 5. Schedule Attainment. Formula: scheduled work completed / scheduled work planned. Target: 85%+. Decision: below 70%, planning is broken.
• 6. Wrench Time. Formula: actual repair time / technician shift time. Target: 50%+. Decision: below 35%, rework the workflow (kitting, mobile, parts availability).

Financial KPIs (Next 2)

• 7. Maintenance Cost as % of RAV. Formula: annual maintenance cost / replacement asset value. Target: 2-4% for most manufacturing. Decision: > 5% means aging fleet or reactive culture.
• 8. Backlog Week-Supply. Formula: total backlog hours / weekly maintenance capacity. Target: 4-6 weeks. Decision: > 8 weeks means work is not closing; < 2 weeks means no improvement runway.

The 12 KPIs You Can Skip

These often appear on dashboards but rarely drive decisions:

• Number of work orders closed. Easy to game by closing small work.
• Total labor hours. Without context on output, meaningless.
• Number of PMs scheduled. Measures schedule size, not effectiveness.
• Uptime. Duplicates availability but less precise.
• Emergency work order count. Depends entirely on how you define emergency.
• Average work order age. Skewed by a few long-open work orders.
• Parts used count. Proxy, and a bad one.
• Overtime hours. Symptom, not cause.
• MRO inventory value. Matters for working capital, not maintenance operations.
• Training hours completed. Input measure; measure training effectiveness instead.
• Safety incident rate in maintenance. Important but belongs on the safety dashboard, not maintenance.
• Work order response time. Only useful when combined with MTTR.

Frequently Asked Questions