Key takeaways
Short answer: Failure rate and MTBF are two ways of expressing the same thing — reliability — from opposite directions. Failure rate is how frequently an asset fails per unit of time (failures per hour, per year). MTBF — Mean Time Between Failures — is the average operating time between failures. For a constant failure rate, they are reciprocals: MTBF is roughly one divided by the failure rate. Failure rate emphasizes frequency; MTBF emphasizes the interval between failures. They are two views of the same reliability. For the repair-time counterpart, see MTBF vs MTTR.
Failure rate is the frequency at which an asset fails per unit of time — how often failures occur. It is expressed as failures per unit of time (failures per hour, per million hours, per year) and answers the question: how frequently does this thing break? A higher failure rate means more frequent failures; a lower failure rate means a more reliable asset that fails less often. Failure rate is a natural way to think about reliability when you care about frequency — how many failures to expect over a period, how often to plan for a breakdown. It is also the basis for reliability engineering and the famous bathtub curve, which describes how an asset's failure rate changes over its life (high early from infant mortality, low and steady through useful life, rising at the end through wear-out). Failure rate is the frequency view of reliability: how often, per unit of time, does the asset fail.
MTBF — Mean Time Between Failures — is the average operating time between consecutive failures of a repairable asset. It answers a related but differently-framed question: how long, on average, does the asset run between failures? Calculated as total operating time divided by the number of failures, MTBF is expressed in time (hours between failures), and a higher MTBF means a more reliable asset that runs longer between breakdowns. MTBF is a natural way to think about reliability when you care about the interval — how long you can expect a run before the next failure. It is widely used because the time-between framing is intuitive for maintenance planning: an MTBF of 1,000 hours tells you to expect a failure roughly every 1,000 operating hours. MTBF is the interval view of reliability: how much time, on average, between failures.
The clean distinction is that failure rate measures frequency (failures per unit time) while MTBF measures the interval (time between failures) — two views of the same underlying reliability, from opposite directions. They are mathematically linked: for an asset with a constant failure rate (the steady, useful-life portion of the bathtub curve), MTBF is the reciprocal of the failure rate — MTBF is about 1 divided by the failure rate. A failure rate of one failure per 1,000 hours corresponds to an MTBF of 1,000 hours; double the failure rate and the MTBF halves. So they are not really competing metrics but two expressions of the same thing: one emphasizes how often (frequency), the other how long between (interval). The choice between them is often just which framing is more intuitive for the audience or the task — failure rate for reliability-engineering and statistical work, MTBF for maintenance planning and communication. Importantly, the reciprocal relationship holds cleanly only for a constant failure rate; when the failure rate changes over the asset's life, the simple relationship is an approximation.
An asset accumulates 10,000 operating hours and suffers 10 failures over that time. Its failure rate is 10 failures divided by 10,000 hours, which is 0.001 failures per hour (or one failure per 1,000 hours). Its MTBF is the reciprocal: 10,000 hours divided by 10 failures, which is 1,000 hours between failures. The same reliability, two numbers: a failure rate of 0.001 per hour, or an MTBF of 1,000 hours — and indeed MTBF equals 1 divided by 0.001, which is 1,000, the reciprocal relationship holding because the failure rate is roughly constant. Now improve the asset's reliability so it fails half as often: the failure rate drops to 0.0005 per hour, and the MTBF doubles to 2,000 hours. Whether you say the failure rate halved or the MTBF doubled, you are describing the same improvement from two directions. The frequency view and the interval view are reciprocal expressions of one reliability.
Because they are two views of the same thing, the choice is largely about framing and context. Failure rate is favoured in reliability engineering and statistical reliability work, where the frequency view fits naturally with probability models, the bathtub curve, and combining the reliability of components into systems (failure rates add in certain configurations). MTBF is favoured in maintenance planning and management communication, where the intuitive time-between framing — expect a failure roughly every so many hours — is easier to act on and explain. Both are equally valid, and a reliability-aware operation uses whichever fits: failure rate for the engineering analysis, MTBF for the maintenance schedule and the management report. The one caution is consistency — make sure everyone knows which is being used and that the underlying failure definition and operating-time basis are the same, since a failure rate and an MTBF that disagree usually do so because of inconsistent definitions, not because the asset behaves differently.
Failure rate and MTBF both measure reliability, which drives the availability factor of OEE — a higher failure rate (lower MTBF) means more frequent breakdowns and more unplanned downtime, the biggest availability loss. They are the frequency and interval views of the same reliability that, combined with repair time (MTTR), determines availability, as covered in availability vs reliability. Reducing the failure rate — equivalently, raising the MTBF — directly lifts availability by cutting the frequency of failures, and is the reliability lever (versus the maintainability lever of faster repairs) in reliability vs maintainability. Whichever framing you use, improving it shrinks the unplanned downtime that dominates the six big losses. Failure rate and MTBF are two ways of measuring the reliability that OEE availability depends on.
Fabrico computes reliability from the failure and downtime events it captures — whether you prefer to read it as a failure rate (how often) or an MTBF (how long between), both fall out of the same logged data, tied to specific assets. Seeing how often each asset fails next to its live OEE is what flags the equipment whose failure rate is rising before it becomes a major availability problem, and confirms whether a reliability improvement actually reduced the failure frequency. Book a demo to see your reliability from real failure data.
Failure rate is how frequently an asset fails per unit of time (failures per hour or year). MTBF (Mean Time Between Failures) is the average operating time between failures. For a constant failure rate, MTBF is the reciprocal of the failure rate. Failure rate emphasizes frequency; MTBF emphasizes the interval.
For a constant failure rate, MTBF is the reciprocal of the failure rate: MTBF is about 1 divided by the failure rate. A failure rate of one per 1,000 hours equals an MTBF of 1,000 hours. They are two reciprocal views of the same reliability, from opposite directions.
Failure rate is favoured in reliability engineering and statistical work, where the frequency view fits probability models, the bathtub curve, and combining component reliabilities. MTBF is favoured in maintenance planning and communication, where the intuitive time-between framing is easier to act on.
Only for a constant failure rate — the steady, useful-life portion of the bathtub curve. During early-life (infant mortality) or wear-out, the failure rate changes over time, so the simple MTBF-equals-one-over-failure-rate relationship becomes an approximation.
Both measure reliability, which drives the availability factor of OEE. A higher failure rate (lower MTBF) means more frequent breakdowns and more unplanned downtime, the biggest availability loss. Reducing the failure rate, or raising MTBF, directly lifts availability.
Anmeldung
Blog
