Failure vs fault
A failure is an event, the moment a machine loses a function. A fault is the state that follows. A machine without power cannot run, but it has neither a failure nor a fault.
EN 13306 (Maintenance terminology) defines both words. A planned maintenance stop is another case: the machine cannot run, yet it has no failure and no fault.
What is the difference between a failure and a fault?
A failure happens at one instant. A fault lasts until someone restores the function. The span from the failure to the restoration is the time to restoration (EN 13306, 9.21), the input of MTTR.
A fault can be partial (6.3): the machine still runs, but it has lost one function. Gulati's glossary counts reduced speed or missed quality as failure too, so the asset need not stop.
Usually a failure starts a fault. A fault can also exist before any failure, from a mistake in design, manufacture, or maintenance. A latent fault (6.2) is a fault nobody has noticed yet.
Is a stopped machine always in fault?
No. EN 13306 gives four reasons for not running their own terms, because each one calls for a different action:
- Fault. The machine cannot perform its function. Maintenance owns the problem.
- Planned maintenance. The machine is stopped on purpose. It is in a down state (6.6), but the definition of fault (6.1) excludes it.
- Missing external resources. Power, staff, compressed air, or cooling water is not there (10.2). The machine is able to run. The standard calls this the external disabled state (6.8).
- No demand. The machine is ready but not needed. This is the idle state (6.10).
Outage (6.7) is the broad word: unable to perform for any reason, so it covers the first three.
External resources do not lower the availability of the machine in EN 13306 (4.7). From the plant's viewpoint the output is still lost.
What do breakdown, outage, and latent fault mean?
| You hear | It means | It is not |
|---|---|---|
| Failure | The event of losing a required function (5.1) | A period of time |
| Fault | The state of being unable to perform it, complete or partial (6.1, 6.3) | A stop for planned maintenance or for missing power |
| Breakdown | An everyday word for a failure that stops the machine. EN 13306 does not define it. Gulati lists breakdown as another name for reactive maintenance | A defined term |
| Latent fault | A fault nobody has seen yet (6.2), for example in a standby unit that nobody tests | A fault that has already shown itself |
| Outage | The state of being unable to perform for any reason, a power cut included (6.7) | Always a fault |
How should you record them?
Two instants describe each failure: when the function was lost and when it was restored. The span between them is the fault and the input of MTTR. Record a stop for missing power, staff, or air as an external cause. Record a planned stop as maintenance. Uptime and downtime shows how each kind enters the numbers.
Corrective maintenance starts after a fault is recognized (7.9). Types of maintenance sorts the jobs that follow, and the work order lifecycle shows the steps. The Maintenance hub has the rest.
Sources
- EN 13306:2017, Maintenance: Maintenance terminology, CEN, 2017. Clauses 4.7, 5.1, 6.1 to 6.3, 6.6 to 6.8, 6.10, 7.9, 9.21, and 10.2. This page explains the standard in our own words; the standard is the authority. Buy it as DIN EN 13306:2018-02 from DIN Media.
- Ramesh Gulati, Maintenance and Reliability Best Practices, 3rd ed., Industrial Press, 2021, chapters 3 and 6 (failure, breakdown, and reactive maintenance).