What Is MTBF (Mean Time Between Failures)?

MTBF (mean time between failures) is the average operating time a repairable asset runs between one unplanned breakdown and the next. It is the standard headline metric for reliability: a higher MTBF means the asset fails less often. It is almost always read alongside MTTR (mean time to repair), which measures how quickly a failed asset comes back - together the two describe how much of the time a machine is actually available to work.

The MTBF formula

MTBF = total operating time ÷ number of failures.

A worked example: a tractor runs 1,200 operating hours over a season and suffers three unplanned breakdowns. Its MTBF for that season is 1,200 ÷ 3 = 400 hours - on average, a failure every 400 hours of work.

Two definitions matter when applying the formula:

• Operating time only. Hours the machine actually ran. Idle weekends and planned downtime do not count, or the metric flatters equipment that simply sits still a lot.
• Failures means unplanned breakdowns. A scheduled oil change is not a failure. A repair raised because something stopped working is.

MTBF applies to repairable assets - things you fix and put back to work. For non-repairable items (a bulb, a fuse, a disposable battery) the equivalent metric is MTTF, mean time to failure.

Calculating MTBF from asset history

The formula is trivial; the inputs are the hard part. You need a dated record of every failure and a measure of operating time. Failure dates come from fault reports - if every breakdown is logged as a service ticket against the specific machine, the timestamps are already there. Operating time comes from an hours meter where one exists; for equipment in continuous or steady use, calendar time is a workable proxy.

The classic trap is under-reporting. Every failure that gets fixed informally, off the record, silently inflates MTBF and makes an unreliable machine look healthy. In practice this is where tooling earns its keep: AMPthilly keeps a permanent, timestamped service history on each asset record, so the failure dates the calculation needs are captured as a side effect of reporting issues, not as a separate chore.

MTBF vs MTTR

MTBF tells you how often an asset breaks; MTTR tells you how long each fix takes. They fail independently - a machine can break rarely but take a week to repair, or break weekly but be fixed in minutes - and they combine into availability: MTBF ÷ (MTBF + MTTR). The 400-hour tractor above, with a 4-hour average repair, is available 400 ÷ 404, roughly 99% of its scheduled time. When uptime disappoints, this split tells you whether the problem is reliability (raise MTBF) or repair speed (cut MTTR).

What counts as a good MTBF

There is no universal benchmark, and published figures rarely match your duty cycle. A skid-steer on demolition work and a tractor doing light yard duty cannot be compared on the same scale. What works instead:

• Compare an asset against itself. A falling MTBF on one machine is the earliest honest signal that it is wearing out - and the cue to start end-of-life planning before the failures get expensive.
• Compare identical units. If one of three identical machines fails twice as often, the problem is that machine (or its operator, or its site), not the model.
• Act on the outliers. A low-MTBF asset is where an inspection work order, an operator conversation, or a replacement decision pays off fastest.

Related terms

• MTTR (Mean Time to Repair) - the companion metric: how long each failure takes to fix
• Work Order - the job record raised when a failure needs repair
• Service Ticket - the fault reports that supply MTBF’s failure dates
• Warranty Tracking - whether a failure is the supplier’s cost or yours
• End of Life (EOL) - the planning a falling MTBF should trigger