OEE and Smart Factory

How to Calculate OEE in a Manufacturing Plant

OEE = Availability x Performance x Quality. This guide covers the exact formula, what each component measures, world-class benchmarks, and how to use the number to drive improvement.

OEE equals Availability x Performance x Quality. Availability = (Planned Run Time - Downtime) / Planned Run Time, Performance = (Actual Output x Ideal Cycle Time) / Run Time, and Quality = Good Parts / Total Parts. World-class OEE is generally 85% or above, while many plants operate closer to 60% to 65%. If a machine is 90% available, runs at 95% performance, and produces 99% good parts, OEE = 0.90 x 0.95 x 0.99 = 84.6%. That one number matters because it combines lost time, lost speed, and lost quality into a single view of usable output.

Planned run time should be scheduled production time minus planned stops such as breaks, lunch, and scheduled maintenance. Availability losses come from breakdowns, changeovers, and startup delay. Performance losses come from minor stops and running slower than the ideal cycle. Quality losses come from scrap and rework, and the underlying data usually comes from the machine counter, downtime log, and first-pass inspection records. On high-speed packaging lines, Performance is often the biggest loss, while job shops usually lose more on Availability because of changeovers and waiting.

The most common OEE mistake is defining planned run time incorrectly. If a shift is 8 hours with two 10 minute breaks and a 30 minute lunch, planned run time is 7 hours, not 8. Another common error is counting reworked parts as good parts in the Quality term. Only parts that pass without rework belong in the good count if you want OEE to show real process loss. Plants also hide minor stops under 5 minutes, which inflates Performance and makes the true bottleneck harder to see.

Use OEE to rank improvement priorities by loss category, not just to post a single score on a board. If Availability is 70%, Performance is 95%, and Quality is 99%, OEE is only 65.8%, and Availability is clearly the first place to attack. Raising Availability to 85% lifts OEE to 79.9% without touching speed or scrap. That tells you changeover reduction, spare parts readiness, and breakdown prevention deserve attention before speed tuning. OEE by component turns a vague productivity conversation into a clear improvement plan.

Track OEE at the machine or cell level, especially at the constraint, because plant averages hide the problem. A 15% gain on the bottleneck machine can produce roughly 15% more throughput for the whole line, while the same gain on a non-constraint may not change shipments at all. Also pair OEE with downtime cost, scrap cost, and schedule attainment so teams do not chase a better score that does not improve business results. Related metrics such as MTBF, MTTR, and line balance help explain which part of OEE is weak. The best use of OEE is not reporting, it is prioritizing the next improvement dollar.

Published 2026-05-28.