Equipment Availability: Formula, Benchmarks & How to Improve It

What is equipment availability?

Equipment availability is the percentage of planned production time during which a machine or line is actually running and able to produce. It is the Availability factor inside OEE and follows a clear formula: run time divided by planned production time. Synonyms in common use: machine availability, plant availability, asset availability, Anlagenverfügbarkeit.

The definition sounds trivial; the measurement almost never is. Two plants quoting "92% availability" may in fact be measuring two entirely different things, because what counts as planned production time — and what counts as a stop — varies enormously from one shift report to the next. Most of the value in improving availability comes from first agreeing on how to measure it honestly.

How is equipment availability calculated?

The ISO 22400 formula is straightforward:

Availability = Run Time / Planned Production Time

Run time is planned production time minus all unplanned stops. Planned stops — scheduled changeovers, breaks, planned maintenance, meetings — are typically excluded from the denominator, not counted as availability losses. This is the single most consequential design choice in any availability measurement: what goes into the denominator decides whether a plant reports 65% or 92% for the same physical reality.

What counts as an availability loss?

Availability losses are unplanned stops longer than the plant's microstop threshold (typically 60 or 120 seconds — anything shorter drops into Performance loss, not Availability loss). Five categories cover almost everything that happens on a real shop floor:

• Breakdowns and faults — mechanical, electrical, control-system failures.
• Material starvation — the line stops because upstream didn't deliver.
• Quality holds — the line is stopped waiting for inspection, release or rework.
• Changeover overruns — the portion of setup time beyond the planned target.
• Organisational stops — shift handover gaps, missing operator, waiting for supervisor decision.

In plants I have walked through on four continents, the distribution is remarkably consistent: technical breakdowns are 30–40% of availability loss, material and organisational causes combined are another 30–40%, and the rest is split between quality holds and changeover overruns. The plants that believe their availability problem is purely mechanical are almost always wrong.

What is a realistic availability benchmark?

The important column is the last one. In every plant where self-reported availability is quoted at 85–90% and no automatic measurement exists, the real number is almost always in the 60–75% range. This is not because people are dishonest; it is because short stops, waiting time and quality holds disappear into the rounding errors of paper-based tracking.

What drags availability down that nobody notices?

Three loss categories show up in every honest measurement and surprise every plant the first time they see them. Microstops just above the measurement threshold — 90 to 300 seconds — add up to 5–10% availability loss across a shift and almost never get logged manually. The second is morning ramp-up: the first 30–60 minutes of a shift rarely deliver full throughput, and operators routinely count the shift as "running" the moment the first part comes out. The third is waiting-for-quality: the line is technically operable but paused while a sample is inspected. In paper reporting, these three categories are usually zero. In automatic measurement, they are typically 15–20% of total loss.

How do you actually improve equipment availability?

Four steps, in strict order, from thirty years of watching this go right and wrong.

1. Measure automatically, not by hand. Until the number comes from machine signals, every improvement is arguing with a guess. This alone usually reveals 20–30 percentage points of hidden loss.
2. Pareto the top three loss reasons. In any plant, three reason codes cover 60–70% of availability loss. Fix those before anything else.
3. Apply the right strategy per asset, not blanket prevention. Bottleneck assets: predictive. Standard assets: preventive. Low-criticality: reactive. See the maintenance strategy framework for the matrix.
4. Close the loop with reason-code discipline. Every stop gets categorised at source. Without this, the Pareto goes stale within weeks.

Skipping step 1 is the most common and most expensive error. Plants jump directly to predictive maintenance software, expensive condition monitoring or Six Sigma workshops, on top of a measurement system that is systematically wrong by a factor of two. The result is an improvement programme that looks professional and produces nothing.

FAQ

Is availability the same as uptime?
No. Uptime is total time the machine is physically powered on and potentially able to run. Availability compares actual run time against planned production time. A machine can have 99% uptime and 70% availability if most of the uptime happens outside planned production — or if the line is powered but idle due to starvation.

How does availability relate to MTBF and MTTR?
There's a direct formula: Availability = MTBF / (MTBF + MTTR). Increasing mean time between failures or decreasing mean time to repair both improve availability. In practice, most plants get more leverage from MTTR improvement (spare parts, standard work, clear escalation) than from MTBF improvement, because MTTR is more controllable.

Should planned maintenance count against availability?
The mainstream convention is no — planned maintenance is excluded from planned production time, so it doesn't drag availability down. Plants that want to challenge their maintenance teams sometimes include it, which gives them a stricter but less comparable number. Pick one convention and stick with it.

Why does availability always drop when a plant first installs an MES?
Because it was wrong before. Paper-based tracking under-reports stops by 20–40%. The initial drop when real measurement starts is not a performance regression — it's the first time the number is true. Plants that don't understand this often roll back the system, believing it "broke" their numbers. The ones that stay the course see availability recover and then exceed the old inflated figure within 6–12 months.

What's the fastest way to lift availability by 5 percentage points?
Install automatic measurement, run the first Pareto, address the top reason code. In plants starting from manual logging, a 5-point lift inside 90 days is routine — because the first reason code is almost always something obvious that nobody had data to prove.

Does high availability always mean high OEE?
No. OEE multiplies availability by performance by quality. A line running 95% availability at 75% performance with 92% quality hits 65.6% OEE — below the commonly cited "85% world-class" threshold. Availability matters, but it's one of three factors.

How does SYMESTIC measure equipment availability?
Via OPC UA, MQTT or digital-I/O gateways reading PLC-level run/stop signals directly, with configurable microstop thresholds and reason-code capture at source. The Production Metrics module shows live availability per machine and line, benchmarked against planned targets. Most plants see the first honest availability number — which is almost never what they expected — within days of go-live, not months.

Related: OEE · MES · Machine Downtime · MTBF · MTTR · Maintenance Strategy · Process Interruptions · Setup Processes · Production Metrics · Alarms.

About the author

Christian Fieg

Head of Sales at SYMESTIC. 25+ years in manufacturing — Six Sigma Black Belt and PLC engineer at Johnson Controls JIT Center of Excellence, global MES and traceability lead for 900+ machines and 750+ users across seven countries, Manager Center of Excellence for the global MES programme at Visteon. Author of "OEE: One Number, Many Lies" (2025). · LinkedIn