Paul LombaOEE is the best-known metric for the effectiveness of a machine or line. Yet it is often calculated incorrectly, because target times, breaks and rework are treated differently. Anyone who wants to measure and improve OEE therefore first needs one clear formula that applies to every line in the same way.
This article explains the formula, works through a complete example and names the four mistakes we see most often in projects.
The OEE formula: three factors, one value
OEE multiplies three ratios. Each stands for one type of loss: downtime, speed and scrap.
OEE = availability × performance × quality
Availability
Run time divided by planned production time. Planned production time is the shift time without planned stops such as breaks. Unplanned downtime (breakdowns, material shortages, changeovers) reduces availability.
Availability = run time ÷ planned production time
Performance
How much the line produced during its run time, compared with the ideal cycle time. Short stops and slow running reduce performance.
Performance = (ideal cycle time × total count) ÷ run time
Quality
The share of parts that are good the first time through. Scrap and rework reduce quality.
Quality = good parts ÷ total count
Worked example: one shift on an injection moulding line
A shift lasts 510 minutes, including a 30-minute break. Planned production time is therefore 480 minutes. The line had 60 minutes of unplanned downtime, and the ideal cycle time is 30 seconds per part.
| Quantity | Calculation | Value |
|---|---|---|
| Planned production time | 510 − 30 min | 480 min |
| Run time | 480 − 60 min | 420 min |
| Availability | 420 ÷ 480 | 87.5 % |
| Performance | (0.5 min × 756 parts) ÷ 420 min | 90.0 % |
| Quality | 726 ÷ 756 parts | 96.0 % |
| OEE | 87.5 % × 90.0 % × 96.0 % | 75.6 % |
As a check: 726 good parts at the ideal cycle time would have taken 363 minutes. 363 of 480 minutes is also 75.6 %. This cross-check quickly reveals errors in the individual factors.
Four common mistakes when calculating OEE
1. Treating breaks and planned maintenance inconsistently
If one line counts the break as downtime and another does not, their values cannot be compared. Define once what counts as “planned” and apply it everywhere.
2. Using the wrong cycle time as the benchmark
If you use the average cycle time of recent months instead of the ideal cycle time, you calculate performance losses away. If performance is above 100 %, the stored cycle time is too slow.
3. Counting rework as good parts
A part that had to be reworked was not good the first time through. Count it as a quality loss, otherwise the effort of rework stays invisible.
4. Simply averaging the OEE of several lines
The average of two lines ignores how long each was planned to run. Calculate the plant figure from the summed times and quantities, not from the individual ratios. Why comparing with industry figures helps little here is explained in the article “What is a good OEE value?” (in preparation).
From practice
[Our own experience from a project: the starting point, what we changed in the calculation and what that revealed. This section provides the expertise bonus from Gate 1 and is written by a person.]
Frequently asked questions
What is a good OEE value?
It depends on the industry, batch size and number of variants. The trend of the same line over several weeks says more than any benchmark.
Can OEE exceed 100 %?
No. A value above 100 % almost always points to an ideal cycle time that is set too slow.
Can I calculate OEE in Excel?
Yes, the formula is simple. The hard part is data collection: downtime and counts recorded by hand are inaccurate and arrive too late.
The implementer & leader
As a former CIO, he built up digitalization and smart factory strategically and was responsible for its operation. Meet the team
Ready to discuss a start on your line?
Together we check which line is suitable for the start and which data is already available.