The key figures for manufacturing, with sources and no dressing up.
One page with the figures most often asked for: OEE benchmarks, formulas, the six big losses, and median figures from the OEE Index, as they accumulate. Every figure has a source. We don’t invent statistics.
Real figures — from the field
The index continuously collects anonymous OEE figures from manufacturers. median figures appear here as soon as the data basis is large enough to be both statistically meaningful and anonymous (a minimum of 5 contributions per segment). Until then, we refer to the established international reference figures below.
The international reference figures
World-class OEE
The established benchmark for world-class OEE is around 85% — corresponding to ~90% availability × ~95% performance × ~99.9% quality.
Source: S. Nakajima, Introduction to TPM, Productivity Press, 1988
Typical unoptimised OEE
Typical production plants without systematic loss reduction sit around 60% OEE — i.e. about 40% of the planned production time doesn’t create good parts. The figure is an industry rule-of-thumb, not a measurement; estimates range between 40% and 60%.
Source: Industry rule-of-thumb among OEE vendors, estimates 40–60% — not Nakajima
The six big losses
Equipment losses are classically categorised into six: breakdowns, changeover/adjustment, minor stops, reduced speed, start-up defects and process defects. The pairs correspond to the OEE factors availability, performance and quality.
Source: S. Nakajima, Introduction to TPM, Productivity Press, 1988
The OEE formula
OEE = Availability × Performance × Quality. Availability = running time ÷ planned time; Performance = (ideal cycle time × parts) ÷ running time; Quality = good parts ÷ total parts.
Source: Standard definition, cf. Nakajima (1988)
How to cite the figures
The 85% and the six big losses are attributed to the original source: S. Nakajima, Introduction to TPM, Productivity Press, 1988. The ~60% does not belong there: it is an industry rule-of-thumb among OEE vendors with estimates between 40% and 60%, and it should not be attributed to Nakajima.
median figures are attributed to: The OEE Index, Prozanta Operations (prozanta.dk/en/oee-benchmark) — stating the number of contributions and the date, as the index is updated continuously.
Prozanta does not publish client names or cases and releases no figures without a source or data basis.
Where does your OEE stand?
Calculate your OEE with a loss breakdown in 5 minutes, and see immediately where you stand relative to the index.