OEE Formula — How to Calculate Overall Equipment Effectiveness
OEE (Overall Equipment Effectiveness) measures how well a machine, line, or factory uses its equipment. It combines three factors — Availability, Performance, and Quality — into a single percentage that tells you how much of your planned production time was truly productive. This guide walks through the OEE formula, each component's sub-formula, two full worked examples, benchmarks, and the most common mistakes.
The OEE Formula
Each of the three factors is a decimal between 0 and 1. Multiply them together and convert to a percentage — that's your OEE. An OEE of 100% would mean the equipment ran the entire planned time, at full rated speed, producing only good parts. That never happens in practice; world-class operations target 85%.
1. Availability — the "uptime" factor
Availability captures unplanned stops — breakdowns, changeovers that ran long, waiting for material, operator unavailability. It answers the question: "Of the time we planned to run, how much did we actually run?"
- Planned Production Time = Shift length − planned stops (breaks, scheduled maintenance, no-demand periods)
- Run Time = Planned Production Time − unplanned stops (breakdowns, setup delays, material waits)
Mini-example: 8-hour shift (480 min), 30 min of planned breaks → Planned Production Time = 450 min. Two breakdowns totalling 45 min → Run Time = 405 min. Availability = 405 ÷ 450 = 0.900 = 90.0%.
2. Performance — the "speed" factor
Performance captures speed losses — the machine is running, but slower than its design speed. This includes small idle periods, micro-stops, reduced feeds, and operator slow-downs. It's the factor most operations under-measure because it requires knowing the ideal cycle time.
- Ideal Cycle Time = the fastest possible cycle time under perfect conditions (from OEM spec or best observed value)
- Total Count = every unit produced during Run Time, including defects
- Keep time units consistent — if Ideal Cycle Time is in seconds, convert Run Time to seconds too
Mini-example: Ideal cycle = 40 s, Total Count = 380 units, Run Time = 405 min = 24 300 s. Performance = (40 × 380) ÷ 24 300 = 15 200 ÷ 24 300 = 0.625 = 62.5%.
3. Quality — the "right-first-time" factor
Quality captures parts that didn't meet spec the first time — scrap and rework both count. Only parts that pass inspection without any touch-up are considered "good."
Mini-example: 380 produced, 15 defective → Good Count = 365. Quality = 365 ÷ 380 = 0.961 = 96.1%.
Worked Example 1 — Discrete Manufacturing, 8-hour Shift
Putting the three mini-examples above together into a single OEE calculation for one shift:
| Input | Value |
|---|---|
| Shift length | 8 h = 480 min |
| Planned stops (breaks) | 30 min |
| Unplanned stops (breakdowns) | 45 min |
| Total parts produced | 380 |
| Ideal cycle time | 40 seconds/part |
| Defective parts | 15 |
| Step | Calculation | Result |
|---|---|---|
| Planned Production Time | 480 − 30 | 450 min |
| Run Time | 450 − 45 | 405 min |
| Availability | 405 ÷ 450 | 90.0% |
| Performance | (40 × 380) ÷ (405 × 60) | 62.5% |
| Quality | (380 − 15) ÷ 380 | 96.1% |
| OEE | 0.900 × 0.625 × 0.961 | 54.0% |
The headline number here is Performance at 62.5% — that's where the biggest losses are, not downtime. The operator might assume "we only had 45 min of breakdown so we did fine," but the speed loss is roughly three times larger than the downtime loss.
Worked Example 2 — Continuous Line, 12-hour Shift
Same formula, different inputs — a packaging line running 12 hours with tighter changeovers but a longer ideal cycle.
| Input | Value |
|---|---|
| Shift length | 12 h = 720 min |
| Planned stops | 60 min |
| Unplanned stops | 90 min |
| Total parts produced | 4 800 |
| Ideal cycle time | 6 seconds/part |
| Defective parts | 120 |
| Step | Calculation | Result |
|---|---|---|
| Planned Production Time | 720 − 60 | 660 min |
| Run Time | 660 − 90 | 570 min = 34 200 s |
| Availability | 570 ÷ 660 | 86.4% |
| Performance | (6 × 4 800) ÷ 34 200 | 84.2% |
| Quality | (4 800 − 120) ÷ 4 800 | 97.5% |
| OEE | 0.864 × 0.842 × 0.975 | 70.9% |
Step-by-step: How to Calculate OEE
- Define the window. One shift, one day, one week — pick something you can measure consistently.
- Record Planned Production Time. Shift length minus all planned stops.
- Record Run Time. Planned Production Time minus all unplanned stops (breakdowns, material starvation, changeover overruns).
- Count Total Count and Good Count. Reject and rework both count against Good Count.
- Find the Ideal Cycle Time. Use OEM spec, or the best cycle observed under ideal conditions.
- Apply the three formulas, multiply, and report as a percentage.
OEE Benchmarks
| OEE % | Rating | What it means |
|---|---|---|
| Below 40% | Very poor | Usually points to measurement gaps rather than genuinely poor operations — verify data first |
| 40–65% | Poor | Significant losses across all three factors — immediate attention needed |
| 65–75% | Average | Typical discrete-manufacturing operation — clear room to improve |
| 75–85% | Good | Well-managed facility with active improvement programs |
| Above 85% | World class | Target for lean manufacturing; rare and hard to sustain |
Benchmark with care: process industries (chemicals, paper) often run at 90%+ because they don't have frequent changeovers, while high-mix job shops may plateau at 55–60%. Compare against yourself over time, not against a universal target.
The Six Big Losses
Every OEE loss maps to one of six categories. Isolating which loss is biggest tells you where to focus improvement:
| Factor | Loss | Typical causes |
|---|---|---|
| Availability | Equipment breakdowns | Tool failure, electrical faults, jams |
| Setup & changeover | Product changes, tooling swaps, cleaning | |
| Performance | Small stops & idling | Sensor trips, blocked chutes, minor jams — usually under 5 min |
| Reduced speed | Running below design rate due to wear, operator caution, or poor inputs | |
| Quality | Startup rejects | Parts scrapped while the process stabilises after a start or changeover |
| Production rejects | Defects produced during stable running |
Common Mistakes When Calculating OEE
- Forgetting planned stops. If breaks are counted as "downtime," Availability will look worse than reality.
- Using a fake ideal cycle time. If the "ideal" is the current running speed, Performance will always come out close to 100% and hide real speed losses.
- Counting rework as good. Anything that needed a touch-up counts against Quality. Measuring right-first-time is the whole point.
- Mixing time units. Run Time in minutes and Ideal Cycle Time in seconds is the single most common arithmetic error.
- Averaging OEEs. You can't average three daily OEEs to get a weekly OEE — you must recompute from raw totals.
Frequently Asked Questions
What does OEE stand for?
OEE stands for Overall Equipment Effectiveness. It is a single-number measure of manufacturing productivity that combines availability, speed, and quality.
What is a good OEE score?
For discrete manufacturing, 85% and above is world-class, 75–85% is good, 65–75% is average, and below 65% indicates significant opportunity for improvement.
How do I calculate OEE in production?
For any chosen time window, compute Availability, Performance, and Quality using the formulas above, then multiply the three values together. The result is OEE for that window.
What is the difference between OEE and equipment efficiency?
Equipment efficiency typically only looks at speed or output versus design capacity. OEE is broader — it combines availability, performance, and quality into a single metric, so it captures downtime and defects that efficiency alone misses.
Can OEE be more than 100%?
No. If your calculation gives a number above 100%, the ideal cycle time you used is too slow — the machine is producing faster than its stated "ideal," which means the ideal needs to be updated.
Which of the three factors should I improve first?
The one with the lowest value. In Example 1 above, Performance at 62.5% drags OEE down far more than the 90% Availability. Always attack the weakest factor first because it has the largest multiplicative effect.
Next: Run your own numbers
Use the OEE Calculator to get your Availability, Performance, Quality, and OEE from shift data in seconds — no manual arithmetic.
📊 Open the OEE Calculator