MTTR — Mean Time To Repair
Definition
Average time to restore a failed repairable asset to operating condition. Maintainability metric — lower MTTR = faster recovery from failure.
Formula
MTTR = Total Repair Time / Number of Repairs
Where “repair time” = elapsed time from failure detection to restored production, including wait-for-tech, parts-fetch, diagnostic, fix, validation. Some shops separate “active repair time” (technician hands on) from total MTTR — be explicit about which.
Required data
- Repair start timestamp (failure-detection time)
- Repair end timestamp (restored-to-production time)
- Repair count over the window
- Consistent failure/repair definitions ↔ pairs with MTBF
Possible data sources
- CMMS work-order timestamps
- PLC fault → ack → reset event chain
- Operator-logged shift-handover notes
Factory example
Same auto stamping press, monthly:
- 4 failures
- Total repair time: 8 hours (avg 2h each, but one was 5h waiting for a part)
MTTR = 8 / 4 = 2.0 hours
The 5-hour outlier is the real signal — spare-parts policy gap.
Economic impact
- Lost production minutes × value-per-minute
- Operator idle time
- Drives spare-parts inventory decisions
- With MTBF → asset availability:
Availability = MTBF / (MTBF + MTTR)
Common mistakes
- Stopping the clock at “fix complete” instead of “production restored”
- Excluding wait-for-tech time → MTTR looks artificially low, maintenance staffing decision misled
- Reporting MTTR plant-wide instead of per asset/failure-mode → top contributor hidden
- Comparing MTTR across very different asset types → noise dominates
See also
- MTBF · OEE · First Pass Yield
- NOTQIN Industrial Intelligence Operating Thesis 2026 §5 Stage 1, Stage 7
- ISO 22400-2