Plant problems
OEE Measured Honestly: Common Mistakes That Make the Number Meaningless
An OEE of 85% that nobody believes helps nobody. The definitions, data sources and traps that decide whether OEE is useful.
Trikaan Research & Engineering4 min read
KEY TAKEAWAYS
- OEE = Availability × Performance × Quality, measured against planned production time.
- Most distortions come from excluding losses or using the wrong ideal speed.
- Compare a machine with itself over time, not with a world-class benchmark.
- The losses behind OEE matter more than the headline number.
Overall equipment effectiveness is one of the most useful numbers in manufacturing — and one of the most abused. Plants report impressive OEE that nobody on the floor believes, because the definitions quietly exclude the losses that matter.
This guide explains the calculation and the traps.
The calculation
OEE is the product of the three. An hour of planned time in which the machine ran 45 minutes at 90% of ideal speed and made 98% good parts gives 0.75 × 0.90 × 0.98, roughly 66%.
| Factor | Formula | Loss it captures |
|---|---|---|
| Availability | Run time ÷ planned production time | Breakdowns, changeovers, waiting |
| Performance | (Ideal cycle time × total count) ÷ run time | Slow running, small stops |
| Quality | Good count ÷ total count | Rejects and rework |
Mistake 1: Shrinking planned time
If changeovers, cleaning or waiting for material are treated as ‘planned’ and removed from the denominator, availability looks excellent. Decide clearly what counts as planned downtime — usually only scheduled breaks and no-demand periods — and keep everything else in.
Mistake 2: The wrong ideal speed
Using today’s typical speed as the ideal hides performance loss. Use the design speed or the best sustained speed achieved for that product. Different products may need different ideal cycle times.
Mistake 3: Counting rework as good
Product that passes only after rework is not first-pass good. Counting it as good inflates quality and hides cost.
Mistake 4: Manual data
Hand-recorded stops miss short stops and round durations. Automatic run signals and counters from the PLC give an honest picture; operators then add reason codes.
Getting operators to trust the number
OEE becomes useful when operators and supervisors believe it. Involve them in defining reason codes, show them the raw data behind each shift’s figure and let them challenge it. When a disputed stoppage turns out to be a sensor fault, fix the sensor publicly. Trust in the number is what turns OEE from a report into a daily improvement tool.
Display OEE and its main losses by shift near the line, updated live or hourly, and discuss the top loss in each shift handover.
Use the losses, not the score
- Break OEE into its six big losses: breakdowns, setups, small stops, slow running, start-up rejects and production rejects.
- Rank them by lost time on the bottleneck.
- Run improvement projects on the top two.
- Track the same machine month over month.
Frequently asked questions
What is a good OEE?
It varies widely by industry and process. Track your own trend and focus on the largest losses.
Should OEE be calculated for every machine?
Start with bottlenecks and critical machines, where improvement has the most effect.
Is OEE the same as utilisation?
No. Utilisation compares run time with total time; OEE also accounts for speed and quality losses.
Can OEE be calculated from PLC data?
Yes — run status, counts and reject signals are often available directly.
- #how to calculate OEE
- #OEE calculation
- #overall equipment effectiveness
- #OEE mistakes
- #OEE availability performance quality
Trikaan Research & Engineering
Trikaan
Written by the researchers and engineers who build Anvaya, ASTR and Seed Track, with input from Trikaan's delivery and engineering teams.
