OEE Measurement
An OEE measurement makes equipment losses visible and prioritisable: Availability × Performance × Quality, tallied honestly and consistently, coupled to daily reaction. The goal is not the number — it is the conversation and the actions the number forces.
The definition is agreed IN WRITING before the first measurement: what counts as planned time, and what doesn’t
The ideal cycle time per product is fixed (design/nameplate speed — not “what we usually run”)
The recording is realistic for the operators: better simple and complete than detailed and full of holes
Decide once: planned production time = total time minus DECIDED exclusions (no orders, planned maintenance if kept outside). Changeovers, cleaning during operation, breaks with the line running, waiting for material — all of that is LOSS, not exclusion. Write the definition down; it must never be negotiated when the number disappoints.
Stops are recorded with a cause (breakdown, changeover, material, etc.) — including the short ones. Minor stops under, say, 5 minutes may be recorded with a tally button rather than text; what matters is that they are counted, for they typically sum to more than the big breakdowns. Quality loss: scrapped + reworked parts, incl. start-up scrap.
Availability = running time ÷ planned time. Performance = (ideal cycle time × parts produced) ÷ running time. Quality = good parts ÷ parts produced. OEE = the product. Always report the three factors together with the total — it is the factors that show where the effort should go.
2-4 weeks of measurement before concluding: the natural variation must be known before a change can be called a change. Use the period to validate the recording — compare recorded time with total time; holes in the accounting are recording errors.
Yesterday’s OEE and biggest loss cause on the board every morning. Weekly/monthly: a Pareto over loss causes — the two or three biggest get an A3 or SMED effort. Without this coupling, OEE is an accounting record; with it, it is a management tool.
Compare primarily with your own baseline. The established references (world class ≈ 85%, typical unoptimised ≈ 60%, Nakajima 1988) are used as a bearing — and contribute anonymously to the Danish OEE Index for a Danish reference.
How it goes wrong in practice — and this is where most implementations part ways.
The temptation to move changeovers out of planned time is constant. Each move makes the number prettier and more useless — the capacity is the same, only the visibility is gone.
If the number is used to judge operators, the recording’s honesty dies in a week — and with it the whole measurement. OEE measures the system, not people.
An OEE number no one acts on is wasted recording work — and the operators know it before management. The reaction cadence (board + Pareto + efforts) IS the purpose of the measurement.
The method only works when it becomes daily routines and leadership behaviour. The operations check shows in 4 minutes where your operations stand — and what should come first.