MTTR (mean time to repair)

MTTR, most often short for mean time to repair, is the average time it takes to fix a failure and bring the asset or service back into use. It measures how quickly a maintenance team recovers, not how often things break.

Formula

MTTR = total repair time / number of repairs

Total repair time is the sum of the downtime for every repair in the period. Decide up front what the clock covers and keep it consistent. A common choice is from the moment the failure is reported to the moment the asset is back in service, which includes travel, diagnosis, waiting for parts, the fix and testing. Some teams count only hands-on repair time, which gives a lower number.

Worked example

A building’s booster pumps fail three times in a quarter. The repairs take 2 hours, 5 hours and 2 hours.

MTTR = (2 + 5 + 2) / 3 = 3 hours.

The other MTTRs

The acronym is overloaded. Mean time to respond, recover, restore and resolve are all in use, especially in IT. Always state which one a report means. In maintenance, “repair” is the usual reading.

How it differs from neighbouring terms

• MTBF (mean time between failures): the average running time between failures. MTBF measures reliability; MTTR measures maintainability. Together they give availability: MTBF / (MTBF + MTTR).
• First response time: the time until someone acknowledges a request. It is one part of the repair clock, not a substitute for it.
• SLA resolution targets are usually set per job; MTTR is an average across jobs, so a good average can hide individual breaches.

Where TaskLoop fits

TaskLoop records a completion timestamp on every task, and its reports show response and completion times by staff member and department. For corrective work, that gives you the raw data for MTTR without keeping a manual log. See analytics and reports.