OEE losses explained: the six big losses and how to find them
Overall Equipment Effectiveness (OEE) answers one question: of the time a line was scheduled to run, how much of it produced good units at full speed? Everything that isn't, is loss. The six big losses tell you what kind.
The formula
OEE = Availability × Performance × Quality
- Availability: run time ÷ planned production time. Stops take it down.
- Performance: (ideal cycle time × total units) ÷ run time. Running slow or stopping briefly takes it down.
- Quality: good units ÷ total units. Scrap and rework take it down.
The six big losses
| Loss | Hits | What it looks like |
|---|---|---|
| 1. Breakdowns | Availability | Equipment fails and the line stops. |
| 2. Setup and adjustments | Availability | Changeovers, warm-up, tweaking until the first good unit. |
| 3. Small stops | Performance | Jams, blocked or starved conveyors, sensor trips. Often too short to get logged. |
| 4. Reduced speed | Performance | Running below the ideal rate "to be safe" or because of wear. |
| 5. Startup rejects | Quality | Scrap and rework while the line stabilizes after a start or changeover. |
| 6. Production rejects | Quality | Scrap and rework during steady running. |
A worked example
A shift is planned for 480 minutes. The line is down for 60 minutes (a breakdown and a changeover). Its ideal cycle is 1 minute per unit. It builds 350 units, and 340 of them are good.
- Availability: (480 − 60) ÷ 480 = 420 ÷ 480 = 87.5%
- Performance: (1 × 350) ÷ 420 = 83.3%
- Quality: 340 ÷ 350 = 97.1%
- OEE: 0.875 × 0.833 × 0.971 = 70.8%
A useful check: 340 good units × 1 minute = 340 minutes of fully productive time, and 340 ÷ 480 = 70.8%. The other 140 minutes are the hidden factory on this shift. Availability lost 60, performance lost 70, quality lost 10.
Using OEE without fooling yourself
- Look at the losses, not just the score. Two lines can both show 70% for completely different reasons. The fix for a changeover problem is nothing like the fix for a speed problem.
- Break it down by line and shift. A plant average hides the one shift or machine where most of the loss lives.
- Use the real ideal cycle. Setting the "ideal" slower than the machine can run makes performance look great and hides reduced speed.
- Don't chase a benchmark. 85% is often quoted as "world class," but plants and processes differ. Your own trend and your biggest priced loss matter more than someone else's number.
- Price the losses. OEE tells you where time goes. Putting a dollar value on each loss tells you which one to fix first. See how to put a price on downtime.
What is one minute of your line worth?
Put your own numbers into the free line-cost calculator, then get the full method in the book.