MTBF (Mean Time Between Failures) is the average run time between two breakdowns of a machine. It answers the question: “How often does it break down?” The higher it is, the more reliable the machine.
Formula
MTBF = run time ÷ number of failures
Only the time the machine was actually running is counted — not weekends, not planned stops. And only failures are counted (the machine cannot run without a repair), not tool changes or waiting for material.
Example
Lathe CNC-02 ran for 320 hours in September. In that period it had 8 failures: 3 × hydraulics, 3 × electrical / sensor, 2 × broken tool.
MTBF = 320 ÷ 8 = 40 hours. With two 8-hour shifts, that is a breakdown every two and a half days.
If in October, after replacing the hydraulic hoses, there are 4 failures in 330 hours — MTBF = 82.5 hours. Reliability is twice as good, and that is a number you can show to management.
How it is used
- Comparing machines. The same machine in two halls with a very different MTBF — something is being done differently (lubrication, operator, material).
- Preventive maintenance. If the hydraulics fail at around 120 hours, the check should come before that — at 100.
- Replacement decisions. An MTBF that falls month after month is a sign that the machine is ageing faster than it is being repaired.
- Together with MTTR. MTBF says how often, MTTR says how long. A machine with a high MTBF and a huge MTTR can be a bigger problem than one that fails often but is fixed in 5 minutes.
Common mistakes
- Calendar time is counted instead of run time — a machine that stood idle without an order looks “more reliable”.
- Failures and other stops are mixed — a tool change is not a failure.
- Too little data — two failures are not a trend. MTBF makes sense after a few months of history.
In MachineTakt
Every stop coded as “Machine breakdown” goes into the machine’s history, together with the real cause written by the technician and the parts used. MTBF and MTTR are calculated automatically by machine and period in the Maintenance module.