Ask a plant manager how busy their machines are and you'll usually hear "fully loaded". Measure it properly and the answer is often very different. OEE — Overall Equipment Effectiveness — is the simplest honest measure of how well a machine is actually used.
The three parts of OEE
OEE multiplies three percentages:
- Availability = run time ÷ planned production time. It captures breakdowns, setups and changeovers.
- Performance = (ideal cycle time × total parts) ÷ run time. It captures slow cycles and small stops.
- Quality = good parts ÷ total parts. It captures rejects and rework.
OEE = Availability × Performance × Quality.
A worked example
A CNC machine is planned for an 8-hour shift (480 minutes) with a 30-minute break, so planned production time is 450 minutes.
- It stops for 45 minutes (a breakdown and a setup). Run time = 405 minutes. Availability = 405 ÷ 450 = 90%.
- The ideal cycle time is 1 minute per part, and it produced 340 parts. Performance = (1 × 340) ÷ 405 = 84%.
- Of the 340 parts, 323 were good. Quality = 323 ÷ 340 = 95%.
OEE = 0.90 × 0.84 × 0.95 ≈ 72%. The machine "ran all shift", yet more than a quarter of its productive capacity was lost.
The six big losses
Every point of lost OEE falls into one of six buckets — and naming them is what makes OEE useful:
- Breakdowns (availability)
- Setups and changeovers (availability)
- Small stops and idling (performance)
- Reduced speed (performance)
- Startup rejects (quality)
- Production rejects (quality)
Why spreadsheets don't work for OEE
Most factories that "track OEE" do it in Excel at the end of the month. By then nobody remembers why the machine stopped on the 4th, small stops were never written down, and every department calculates it slightly differently. The number becomes a report, not a tool.
What OEE software does: operators log stops and reasons on a tablet in two taps — or the software reads run/stop signals and part counts from the machine where it supports it — and OEE is calculated automatically per machine and shift, with a Pareto of the top losses. See OEE & machine monitoring software.
How to get started
- Start with your bottleneck machine, not the whole plant.
- Agree on reason codes — ten to fifteen clear downtime reasons beat fifty vague ones.
- Get an honest baseline for two to four weeks before setting targets.
- Act on the top two losses — that's where most of the improvement comes from.
What is a "good" OEE?
You'll often see "world-class OEE is 85%" quoted. Treat that as a reference, not a target. OEE depends heavily on process, product mix and how you define planned time and ideal cycle time. The more useful comparison is your own machine against its own past — and between similar machines in your plant. An honest 55% that improves steadily is worth more than an inflated 80% nobody believes.
Common OEE calculation mistakes
- Using a slow "standard" cycle time — this hides performance losses. Use the ideal or design cycle time.
- Excluding changeovers from planned time — changeovers are a loss to reduce, not something to exclude.
- Counting reworked parts as good — first-pass quality is what matters.
- Averaging OEE percentages across machines — weight by planned time, or calculate from totals.
- Inconsistent definitions between shifts or plants — agree definitions and let software apply them.
From OEE to action
The number itself doesn't improve anything. The value is in the loss breakdown. A weekly routine works well: look at the top three losses per critical machine, assign an owner and an action for each, and check the next week whether the loss reduced. OEE software makes this routine easy by producing the loss Pareto automatically.
TEEP and capacity planning
OEE measures effectiveness during planned production time. TEEP (Total Effective Equipment Performance) measures against all calendar time. TEEP is useful when deciding whether to add shifts or buy a new machine — sometimes the cheaper path to capacity is raising OEE and running more planned hours.
OEE in different industries
In discrete manufacturing (machining, moulding, pressing), OEE is measured per machine. In process and packing lines, it's measured per line, with the slowest machine defining the line's speed. In batch processes, performance may be measured against standard batch time. The principles are the same; the definitions should be agreed per process and applied consistently by software.
Getting operator buy-in
OEE can feel like a surveillance tool to operators. Present it as a measure of the machine and process, not of people, and involve operators in identifying causes of losses. When their reports lead to fixes — a recurring jam resolved, a material delay addressed — they become the biggest supporters of the system.
Manual vs automatic data capture
Manual entry on tablets is quick to start and captures reasons well; automatic signals from machines capture every stop and count precisely but need hardware. The best systems combine both: machines detect stops and counts, operators add reasons. Start manual if needed — reliable reasons matter more than perfect timing.
Frequently asked questions
What is the OEE formula?
OEE = Availability × Performance × Quality, where availability is run time ÷ planned time, performance is (ideal cycle time × total count) ÷ run time, and quality is good count ÷ total count.
What are the six big losses?
Breakdowns, setups and adjustments, small stops, reduced speed, startup rejects and production rejects.
Can OEE be measured without connecting machines?
Yes. Operators can record stops, reasons and counts on tablets; machine signals can be added later for automatic capture.
Corposol builds OEE software, downtime tracking and machine monitoring for factories. Talk to us about measuring OEE in your plant.
