What Is MTTR (Mean Time To Repair)?

MTTR (Mean Time To Repair) Explained

What the clock includes is the single most contested definition in maintenance metrics. Strict MTTR covers only hands-on repair time, from technician arrival to function restored. Mean time to restore service covers the full outage: detection, notification, technician travel, diagnosis, parts acquisition, repair, and verification. The second is what production actually loses, and it is often three to five times the first. Plants that report the narrow definition while production experiences the wide one generate persistent distrust of maintenance data.

MTBF and MTTR combine into inherent availability as MTBF divided by the sum of MTBF and MTTR. An asset with 400 hours MTBF and 6 hours MTTR has availability of 400/406, or 98.5 percent. That formula makes the trade-off explicit: an asset that fails often but is restored in minutes can outperform one that fails rarely but takes two days to fix, which is why designing for maintainability often beats chasing reliability alone.

MTTR reduction levers are usually logistical rather than technical. In most plants, diagnosis and parts wait dominate the outage clock while actual wrench time is a minority of it. The high-yield countermeasures are stocking critical spares locally, standardizing components across assets to shrink the spare-parts matrix, documenting fault trees and known failure signatures, providing machine-side access to manuals and drawings, and pre-positioning special tooling and lifting gear.

The metric also distorts easily. A single catastrophic rebuild can dominate the mean, so median repair time and a distribution view are usually more informative for planning. Splitting MTTR by failure mode is more actionable still, since the response to a chronic ten-minute sensor fault is completely different from the response to a twelve-hour spindle replacement even though both feed the same average.