OEE — Overall Equipment Effectiveness — is the closest thing manufacturing has to a universal health score for a machine or production line. It's a single percentage that tells you how much of your planned production time is being used productively. World class is 85%. Most facilities start somewhere between 40% and 60%, which means there's more capacity sitting unused than most managers realise.
The reason OEE is useful isn't the number itself — it's the three factors that make it up. Each one points to a different category of loss, which means OEE doesn't just tell you something is wrong, it tells you roughly where to look.
The OEE formula
If your availability is 90%, your performance is 95%, and your quality is 99%, your OEE is 0.90 × 0.95 × 0.99 = 84.6% — just under world class. That's the compounding effect: each factor below 100% multiplies the losses from the others.
Availability
Planned run time that was actually used. Lost to breakdowns, changeovers, and unplanned stops.
Performance
Actual output vs. maximum possible output at ideal speed. Lost to slow cycles and minor stops.
Quality
Good units as a proportion of total units produced. Lost to defects, rework, and startup scrap.
The targets above — 90%, 95%, 99% — are the world-class benchmarks for each factor. Hitting all three simultaneously gives you OEE ≥ 85%.
Availability — are you running when you should be?
Availability measures the proportion of planned production time in which the equipment was actually running.
If you planned to run for 8 hours and a breakdown took the line down for 45 minutes, your run time is 7 hours 15 minutes and your availability is 90.6%.
The key distinction is planned vs. unplanned downtime. A scheduled 30-minute lunch break doesn't reduce availability — it's already excluded from planned production time. An unplanned stop for a conveyor jam does. This is why accurate downtime tracking is the foundation of a meaningful OEE measurement.
Performance — are you running at full speed?
Performance captures speed losses — when the equipment runs, but slower than its ideal rate.
Performance losses come from two sources: reduced speed (running slower than the ideal rate, often to avoid quality issues or because of worn tooling) and minor stops (brief interruptions of less than a few minutes that are too short to log as downtime but add up significantly over a shift). Minor stops are often the most underestimated OEE loss category — individually invisible, collectively substantial.
Quality — are you making good parts?
Quality measures the proportion of output that meets specification first time.
Quality losses include process defects (parts that fail specification during steady-state production) and startup rejects (parts scrapped during warmup, after changeovers, or when a process is being re-established). Startup losses are often overlooked but can be significant in high-changeover environments — another reason why reducing your EPEI matters beyond just inventory levels.
A worked example
A CNC machine runs a single 8-hour shift. Planned production time is 480 minutes. Here's the data for one shift:
| Metric | Value |
|---|---|
| Planned production time | 480 min |
| Unplanned downtime | 52 min (two breakdowns) |
| Run time | 428 min |
| Ideal cycle time | 0.5 min/part |
| Total parts produced | 780 |
| Defective parts | 16 |
Availability = 428 ÷ 480 = 89.2%
Performance = (0.5 × 780) ÷ 428 = 390 ÷ 428 = 91.1%
Quality = (780 − 16) ÷ 780 = 764 ÷ 780 = 97.9%
OEE = 0.892 × 0.911 × 0.979 = 79.5%
A 79.5% OEE sounds reasonable — but compare it to world class at 85% and you're leaving roughly 5.5% of planned capacity unused. On a single machine running one shift, that's meaningful. Across a whole plant, it's significant.
OEE benchmarks
| OEE score | Rating | What it typically means |
|---|---|---|
| ≥ 85% | World class | Best-in-class performance. All three factors near their targets. |
| 70–84% | Good | Room for improvement. Often one factor dragging the others down. |
| 50–69% | Fair | Significant losses present. Improvement programme warranted. |
| < 50% | Poor | Major losses across multiple categories. Urgent attention needed. |
How to improve OEE
The factor with the lowest score is almost always the right place to start. Some typical improvement actions by factor:
- Availability — preventive maintenance schedules, root cause analysis on repeat failures, faster changeover times via SMED
- Performance — investigate and eliminate minor stops, restore equipment to its designed speed, address tooling wear
- Quality — first-off inspection discipline after changeovers, process control limits, operator training on defect recognition
Once you've identified which factor to target, use the hour-by-hour production tracker to monitor whether improvements are sticking during the shift. When actual output falls below target, operators log the reason — downtime, quality issue, speed loss — which over time builds the data you need to sustain improvement.
OEE is most valuable not as a headline number to report upward, but as a diagnostic framework that points your improvement efforts at the right category of loss. Measure it consistently, break it into its three components, and it will tell you exactly where your capacity is going.
Calculate your OEE instantly
Enter your run time, ideal cycle time, units produced and defects to get your OEE score broken down by factor.