What is OEE and how do you calculate it?
OEE means Overall Equipment Effectiveness. It is one percentage that shows how much of your planned production time produced good parts at full speed. You get it by multiplying three numbers: Availability, Performance and Quality. A score of 100% means no stops, no slow running and no rejects.
By Lalit FauzdarPublished: Last updated:
How do you calculate OEE?
You multiply three percentages: OEE = Availability × Performance × Quality. Each one answers a simple question about the machine.
- Availability: did the machine run when it was planned to run? Run time divided by planned production time.
- Performance: did it run at full speed? Pieces made divided by the pieces it could have made in that run time at its ideal cycle time.
- Quality: was what it made saleable? Good pieces divided by total pieces.
Because the three are multiplied and not averaged, OEE falls quickly. A machine that scores 90% on each factor still ends up at about 73%.
What does a worked OEE example look like?
Here is one machine on one 8 hour shift, with every step shown. Change the numbers to your own and the method stays the same.
Inputs for the shift
Sample numbers, not customer data
| Item | Value |
|---|---|
| Shift length | 480 minutes |
| Planned breaks (tea, lunch) | 45 minutes |
| Unplanned downtime (breakdowns, waiting) | 55 minutes |
| Ideal cycle time | 30 seconds per piece |
| Pieces made | 650 |
| Rejected pieces | 26 |
Every step
Sample numbers, not customer data
| Step | Working | Result |
|---|---|---|
| Planned production time | 480 − 45 | 435 minutes |
| Run time | 435 − 55 | 380 minutes |
| Availability | 380 ÷ 435 | 87.4% |
| Ideal pieces in run time | 380 × 60 ÷ 30 | 760 pieces |
| Performance | 650 ÷ 760 | 85.5% |
| Good pieces | 650 − 26 | 624 pieces |
| Quality | 624 ÷ 650 | 96.0% |
| OEE | 0.874 × 0.855 × 0.960 | 71.7% |
A quick check: 624 good pieces at 30 seconds each is 312 minutes of perfect production. Divide by 435 planned minutes and you get 71.7% again.
Notice the pattern. Availability lost about 13 points, Performance about 14 and Quality about 4. The big leaks are stops and slow running, not rejects.
You can try your own numbers in the OEE calculator on the product page. OEE software and calculator
What are the six big losses behind OEE?
The six big losses are the standard list of reasons a machine does not reach 100%. Each loss lowers one of the three factors.
| Loss | Hits | Usual example |
|---|---|---|
| Breakdowns | Availability | Motor fault, hydraulic leak, tool crash |
| Setup and adjustments | Availability | Job changeover, waiting for first piece approval |
| Small stops | Performance | Jam, sensor trip, operator steps away for a minute |
| Reduced speed | Performance | Running slower than the ideal cycle time |
| Startup rejects | Quality | Scrap while the machine settles after a start |
| Production rejects | Quality | Parts that fail inspection during normal running |
Small stops are the hardest to see. Nobody writes down a 40 second stop in a register, yet twenty of them in a shift add up to a long stoppage.
What is a good OEE score?
The commonly quoted benchmark is 85%, and it comes from the OEE literature. Treat it as a guide and not as a target for every plant.
What is good depends on the industry, the product mix and how often you change jobs. A job shop with many short runs will score differently from a line that makes one part all day.
One more honest point. A first measurement is often lower than people expect. That is not a failure. It usually means stops and slow running were invisible before, and now you can see them.
The useful comparison is your own machine against itself, week after week. A score that moves up after you fix a cause is worth more than a number from a textbook.
What is the difference between OEE and machine utilisation?
Utilisation only asks how long the machine was running. OEE also asks how fast it ran and how many parts were good.
Using the sample shift above, the machine ran 380 minutes out of 480. Utilisation on the full shift is about 79%. OEE for the same shift is 71.7%, because it also counts slow running and rejects.
Utilisation is the easier number to collect and a good first step. OEE needs a piece count and a reject count on top.
How do you collect the data for OEE?
You can collect it on paper or capture it automatically. Paper works for a start. It gets unreliable as soon as stops are short or frequent.
On paper, an operator or supervisor writes start and stop times, reasons and counts. It is cheap, but it depends on memory, and small stops go missing.
Automatic capture reads the machine directly. LiveMach fits a small device with sensors outside the machine. It reads the machine signals, such as a stack light or a current clamp, and records when the machine is running, stopped, idle or off. That gives you run time and downtime, which is the Availability side, without anybody filling a register.
- Planned production time comes from your shift and break schedule.
- Run time and downtime come from machine state captured by the device.
- Ideal cycle time comes from your own process sheet or an agreed figure for the part.
- Good and rejected pieces need a counting input. What that input is on your machine is agreed with you at the survey.
For how LiveMach turns machine state into reports and OEE, see the product page. OEE software