MTTR vs MTBF: What They Measure and How to Calculate Them

Two different questions

MTTR and MTBF often get grouped together as the reliability metrics, but they answer two different questions. Mean Time Between Failures (MTBF) asks how often a piece of equipment fails. Mean Time To Repair (MTTR) asks how long it takes to get that equipment back into service once it has failed. A maintenance team that only tracks one of the two is missing half of the reliability picture.

How MTBF is calculated

MTBF is calculated as total operating time divided by the number of failures recorded over that period, counting only the failures that have a recorded operating time. If a piece of equipment ran for 1,000 hours across a month and failed 4 times during that period, its MTBF for the month is 1,000 divided by 4, or 250 hours between failures. A higher MTBF means the equipment fails less often.

How MTTR is calculated

MTTR is calculated as total repair time divided by the number of failures that have a recorded repair time. If those same 4 failures took a combined 8 hours to repair, the MTTR is 8 divided by 4, or 2 hours per repair. A lower MTTR means the team gets equipment back into service faster once something goes wrong.

Availability ties the two together

Availability combines both metrics into a single number: MTBF divided by the sum of MTBF and MTTR. Using the figures above, availability would be 250 divided by 252, or about 99.2 percent. Two pieces of equipment can have very different failure and repair patterns and still land on a similar availability number, which is exactly why it helps to look at MTBF and MTTR separately rather than relying on availability alone.

Deciding what counts as a failure

Not every stoppage should count as a failure in these calculations. A nuisance trip that resets itself with no repair action, or a false alarm from a sensor, is usually excluded from MTBF and MTTR, since including it would understate how reliable the equipment actually is. What should count is any event where the equipment stopped performing its function and needed a corrective action, however small, before it could run again. Applying that rule consistently across every entry in the failure log matters more than getting the exact definition perfect on the first try.

Why tracking both changes what you fix

A low MTBF with a low MTTR points to equipment that fails often but is quick and easy to fix, which usually means the priority is reducing the failure rate itself: better preventive maintenance, addressing a known wear pattern, or replacing an undersized component. A high MTBF with a high MTTR points to the opposite problem: failures are rare, but when they happen they are expensive in downtime, which usually means the priority is spare parts availability, clearer repair procedures, or faster diagnosis. Reading the two numbers together, instead of only watching an overall downtime total, is what turns a failure log into a decision about where to spend maintenance effort next.

Getting clean data to calculate from

Both formulas depend on having a consistent failure log: a timestamp, the equipment or asset ID, the failure mode, the operating time since the last failure, and the repair time for that specific event. Planned maintenance downtime is generally excluded from both calculations, since MTBF and MTTR are meant to describe unplanned failures, not scheduled work. Failures with missing or unclear repair times should be flagged rather than guessed at, since a few bad data points can swing an MTTR average noticeably in a small dataset.

As a quick reference, the three formulas covered above are:

• MTBF = total operating time / number of failures
• MTTR = total repair time / number of failures with a recorded repair time
• Availability = MTBF / (MTBF + MTTR)

Putting it into practice

Calculating MTBF, MTTR and availability by hand in a spreadsheet works, but it gets slower and more error prone as the failure log grows. The MTTR and MTBF Calculator turns a failure log into these KPIs automatically, with a dashboard and failure analysis by mode, equipment and trend. For a one-off analysis of your own failure and repair records instead, the MTTR / MTBF and Reliability Analysis service applies the same calculations directly to your data.