MTTR
Mean Time To Repair: average time from fault detection to service restoration, a critical operational KPI for NOC performance.
MTTR is the stopwatch on how long it takes to get service back after something breaks — averaged across incidents, from the moment a fault is detected to the moment service is restored. It's one of the numbers a NOC lives and dies by, and it frequently appears in SLAs because restoration speed is part of what customers are paying for. Lower is obviously better, but the value of the metric is in tracking the trend and breaking down where the time actually goes.
Two things trip people up. First, the "R" is ambiguous in the wild — MTTR is variously read as Mean Time To Repair, Restore, Resolve or Respond, and they measure different spans. Pin down which one a given report means before comparing numbers across teams or vendors. Second, MTTR pairs with MTBF (Mean Time Between Failures): MTBF tells you how often things break, MTTR how fast you recover, and availability is essentially a function of both. A network with rare faults but slow repairs can end up with worse availability than one that fails more often but bounces back fast.
Frequently asked questions
- What is the difference between MTTR and MTBF?
- MTBF — Mean Time Between Failures — measures reliability: how long, on average, the system runs before it fails. MTTR measures recoverability: how long it takes to restore service once it has failed. They work together to determine availability, roughly MTBF / (MTBF + MTTR). Improving either lifts availability, but they're different levers — MTBF is about building or maintaining things that fail less, MTTR is about detecting and fixing faster.
Related terms
Want to truly understand MTTR? Learn it in context.
A free account opens all 13 courses — 404 video lessons, labs and full practice exams — for 7 days, no card. Prime Membership keeps them and adds the certification exam and a verifiable certificate.
- 7 days of every course
- Then the first module, free
- No card needed
- TELCOMA since 2009
Get one 5G/LTE engineering deep-dive a fortnight
One thing explained properly every other Tuesday — a real failure, and the lab or trace where you work it yourself. No spam, no marketing, just engineering. Unsubscribe in one click.