What are MTBF and MTTR?
MTBF (mean time between failures) measures how often an asset breaks down; MTTR (mean time to repair) measures how long it takes to get it running again. Together they explain availability, and they point to different solutions: a low MTBF calls for reliability work such as root cause analysis, better maintenance or redesign, while a high MTTR calls for maintainability work such as spare parts, skills, diagnostics and standard repair procedures.
Formulas
| Measure | Formula |
|---|---|
| MTBF | (Scheduled time − downtime) ÷ number of failures |
| MTTR | Total downtime ÷ number of failures |
| Availability | (Scheduled time − downtime) ÷ scheduled time ≈ MTBF ÷ (MTBF + MTTR) |
| Failure rate | 1 ÷ MTBF |
MTBF/MTTR example
A packaging line logs 35 breakdowns on six machines over six months, 73.8 hours of downtime in 24,000 scheduled machine-hours. Across all assets, MTBF is 684 h, MTTR 2.1 h and availability 99.69%. Against targets of 800 h MTBF and 2 h MTTR, two machines are in the priority quadrant: the filler (MTBF 442 h, MTTR 2.4 h, nine failures) and the case packer (MTBF 793 h, but 7 hours per repair). The labeler fails most often (12 times) but is fixed in half an hour, so it needs reliability work rather than faster repair. Servo drive faults and gearbox failures on the case packer cause the most downtime, 15.5 and 15 hours.
Tips
- Count operating time, not calendar time: use the scheduled production hours of each asset.
- Record downtime from failure to restart, including waiting for parts and technicians, and keep the cause.
- Small numbers are noisy. With two or three failures, MTBF can change a lot from one period to the next.
- Analyse failure ages of replaced parts with Weibull analysis, dig into the top causes with a fishbone diagram, and see the effect on OEE.
How to use this MTBF/MTTR Analysis tool
- Set the period. Name the line and the period, and optionally enter MTBF and MTTR targets.
- List the assets. Add each asset with its scheduled operating time in the period.
- Enter the failure log. Enter or paste one row per breakdown: asset, date, downtime and cause.
- Read the quadrants. Assets in the lower right fail often and take long to repair: start there. Lower left needs better reliability, upper right faster repair.
- Find the causes. The downtime by cause table shows where to focus root cause analysis.
- Export. Export the chart and tables to PDF, PNG or Excel.
Frequently asked questions
What is MTBF?
Mean time between failures: the average operating time between breakdowns of a repairable asset. It is calculated as total operating time (scheduled time minus downtime) divided by the number of failures. A higher MTBF means a more reliable asset.
What is MTTR?
Mean time to repair: the average time to restore an asset after a failure, from breakdown to running again. It is total downtime divided by the number of failures. A lower MTTR means a more maintainable asset and a better-organised maintenance response.
How are MTBF, MTTR and availability related?
Inherent availability = MTBF ÷ (MTBF + MTTR), the share of scheduled time the asset is able to run. You can improve it by failing less often (raising MTBF) or by repairing faster (lowering MTTR). The quadrant chart shows which of the two each asset needs.
What is the difference between MTBF and MTTF?
MTBF applies to repairable items that are restored after each failure. MTTF (mean time to failure) applies to non-repairable items that are replaced, such as bearings or lamps; for those, a Weibull analysis of the failure times is more informative.
Does MTBF mean an asset will run that long?
No. MTBF is an average. With random failures, about 63% of assets fail before reaching the MTBF. Use it to compare assets and track trends, not to predict when one particular asset will fail.
What counts as a failure?
Define it before you collect data, for example any unplanned stop longer than 5 minutes that needs maintenance. Short stops cleared by the operator are better tracked as minor stops in OEE. Consistent definitions matter more than the exact threshold.