LimitLine

Learn · updated 2026-10-21

How to calculate OEE

Overall equipment effectiveness tells you how much of the time you planned to produce actually made good parts at full speed. It multiplies three losses together, so one number shows the whole picture and the three factors show where it went.

Availabilityrun time ÷ planned timePerformanceideal cycle time × total count÷ run timeQualitygood count ÷ total countOEEA × P × Q

Worked example: one shift

The example from OEE.com: 420 minutes planned, 47 minutes down, an ideal cycle time of 1.0 second, 19,271 pieces made and 423 rejects.

FactorCalculationResult
Run time420 − 47 minutes373 min
Availability373 ÷ 42088.8%
Performance(1.0 s × 19,271) ÷ (373 × 60 s)86.1%
Quality(19,271 − 423) ÷ 19,27197.8%
OEE88.8% × 86.1% × 97.8%74.8%

Where the time went

LossMinutes
Downtime (availability)47
Slow cycles and small stops (performance)51.8
Rejects (quality)7.1

The biggest loss is performance, about 52 minutes, slightly more than the downtime. Small stops and slow running rarely get logged, so they hide; time them for a week before buying anything to fix downtime.

What is a good OEE?

The often-quoted world-class benchmark, attributed to Seiichi Nakajima, is 85%: availability 90% × performance 95% × quality 99.9%. Many plants measuring for the first time land well below that. Use OEE to track one line against itself over time; comparing different lines or plants is misleading because planned time and ideal cycle time are defined differently.

Three common mistakes

  1. A padded ideal cycle time. Use the fastest the machine can truly run, not the standard. A padded ideal hides performance loss and can push performance over 100%.
  2. Counting planned stops as downtime. Breaks and scheduled maintenance come out of planned time, not run time.
  3. Counting reworked parts as good. Good means right the first time.
Calculate OEE in LimitLineIts OEE example starts with this shift (a Pro tool; Free users can try it on the example).

Sources

  1. OEE.com, “Calculating OEE” (worked example)
  2. S. Nakajima (1988), Introduction to TPM: Total Productive Maintenance, Productivity Press