Reliability in Action

Availability Calculator

Calculate availability from measured uptime and downtime, or inherent availability from MTBF and MTTR. Both methods are shown with the formula substituted.

Use the same time unit for both inputs. The result is a ratio, so hours, days or shifts all work as long as they match.

The two methods

A. Uptime and downtime

Availability = Uptime / (Uptime + Downtime) × 100

This describes what actually happened during a defined observation window. It is only as good as your downtime records and your downtime definition.

B. MTBF and MTTR

Inherent Availability = MTBF / (MTBF + MTTR) × 100

This describes the equipment’s own capability, combining how long it typically runs between failures with how long active corrective repair typically takes.

Assumptions & boundaries

  • The MTBF/MTTR method returns inherent availability. It considers active corrective repair time and excludes logistics and administrative delays such as waiting for spares, waiting for a crew, permit issue, or management decisions.
  • The uptime/downtime method depends on definitions. The result moves with how you define downtime — whether planned shutdowns, minor stops, rate loss and idle periods are counted — and with the observation window you selected.
  • Standby treatment is a definition choice. Whether standby hours count as available time, are excluded from the window, or are reported separately depends on your KPI definition and operating context. There is no single correct convention.
  • Both methods assume consistent units and reasonably complete data. Gaps in downtime recording bias the result upward.

Worked example

A pump is monitored over a 720-hour month. Records show 20 hours of downtime across three events, so uptime is 700 hours.

Availability = 700 / (700 + 20) × 100 = 700 / 720 × 100 = 97.22%

Separately, the reliability engineer reports MTBF of 350 hours and MTTR of 10 hours for the same pump, using active repair time only.

Inherent Availability = 350 / (350 + 10) × 100 = 350 / 360 × 100 = 97.22%

In this example the two agree closely, which suggests little logistics delay in the period. If the observed figure had been 93% against an inherent 97%, the difference would point at waiting time rather than at the pump.

Frequently asked questions

Which availability formula should I use?
It depends on the question. If you want to describe what actually happened over a period, use uptime divided by uptime plus downtime for that period. If you want to describe the equipment's own capability without logistics or administrative delays, use the MTBF and MTTR form, which gives inherent availability. Neither is universal, and they answer different questions.
Is inherent availability the same as the availability my plant reports?
Usually not. Inherent availability considers only active corrective repair time. Reported plant availability typically includes waiting for parts, labour, permits and decisions, so it is normally lower than inherent availability for the same asset.
How should standby equipment be treated?
That depends on your KPI definition and operating context. Standby hours can be counted as available time, excluded from the observation window, or tracked separately. Whichever you choose, apply it consistently, because switching conventions changes the result without anything changing in the field.
Does availability tell me how often the equipment fails?
No. Availability is a ratio of time. One long outage and many short outages can produce the same availability. To describe failure frequency you need MTBF or a failure count, and to describe restoration speed you need MTTR.
What observation window should I use?
Use a window long enough to contain a representative number of events and aligned to how the result will be used, such as a month for routine reporting or a full campaign for turnaround planning. Always state the window alongside the percentage.

Related