OEE is the indicator that measures the overall effectiveness of a piece of equipment by multiplying three factors: availability, performance and quality. The acronym comes from Overall Equipment Effectiveness, and the result shows how much of the machine's theoretical productive potential actually became good product.
This guide explains each factor of the formula, shows a complete numerical calculation example step by step, and details the part that is usually left out of articles on the subject: the role of maintenance in the index, and how not to inflate OEE by accident.
Key takeaways
- OEE is the product of three factors. The index is born from multiplying availability by performance by quality, and it drops when any of the three drops.
- Availability is maintenance's share. Breakdowns and unplanned stoppages bring down the first factor of the calculation, and that is where the planning function has direct control over OEE.
- The index is only worth it if the data is honest. Reclassifying a stoppage, overestimating the ideal cycle or reporting from memory inflates OEE and hides exactly the loss it should reveal.
- Public references help calibrate the target. The 85% mark is cited as world-class, but most factories operate well below it, and the useful target is to improve against your own baseline.
What is OEE
OEE (Overall Equipment Effectiveness) is an indicator created in the context of TPM, Japanese Total Productive Maintenance, to answer a simple question: of everything this machine could produce in the available time, how much did it actually produce with quality?
The strength of OEE is in condensing three different losses into a single number:
- Stoppage loss: the equipment was not running when it should have been (breakdown, setup, no operator, no material).
- Speed loss: the equipment ran, but below the ideal pace (microstops, degraded cycle).
- Quality loss: the equipment produced, but part of the output came out defective, as scrap or rework.
An OEE of 100% would mean producing throughout the entire planned time, at the maximum theoretical speed, with no bad parts. None of these three conditions survives on its own in the real world, and that is why OEE is useful: it shows where reality is losing against potential.
The OEE formula
OEE = Availability × Performance × Quality
In plain text: OEE is the result of multiplying availability by performance by quality, with the three factors expressed as percentages. Because it is a multiplication, the final index is always less than or equal to the worst of the three factors, and a small drop in each becomes a big drop in the total.
Availability
Availability is the time in which the equipment actually operated divided by the planned production time. Planned stoppages (meals, meetings, scheduled preventive maintenance) leave the denominator; unplanned stoppages (breakdown, unforeseen setup, lack of material) reduce the numerator.
In plain text: availability equals operating time divided by planned production time. It is the factor maintenance most influences.
Performance
Performance compares what was produced with what the equipment would produce in the same operating time, running at the ideal cycle. In plain text: performance equals actual production divided by the theoretical production of the operating time. Microstops and reduced speed are the typical losses of this factor.
Quality
Quality is the fraction of output that came out good on the first pass. In plain text: quality equals the number of good parts divided by the total quantity produced. Scrap and rework enter here, even if the part is recovered later.
How to calculate OEE: a step-by-step numerical example
The example below is hypothetical, with round numbers chosen to make the math easy. Imagine a filling line on an 8-hour shift.
Step 1: planned production time
The shift has 480 minutes. There were 60 minutes of planned stoppages scheduled (meal and shift change). Planned production time is 480 minus 60, that is, 420 minutes.
Step 2: availability
During the shift, the line suffered 84 minutes of unplanned stoppages: a 60-minute breakdown and an unforeseen 24-minute setup. Operating time was 420 minus 84, that is, 336 minutes.
Availability = 336 ÷ 420 = 0.80 = 80%.
Step 3: performance
The line's ideal cycle is 1 unit per minute. In 336 minutes of operation, theoretical production would be 336 units. Actual production was 302 units.
Performance = 302 ÷ 336 = 0.899 = 89.9%.
Step 4: quality
Of the 302 units produced, 287 came out good on the first pass; 15 were scrapped or reworked.
Quality = 287 ÷ 302 = 0.950 = 95.0%.
Step 5: OEE
OEE = 0.80 × 0.899 × 0.950 = 0.683 = 68.3%.
There is a shortcut to check the math: OEE can also be calculated as good parts multiplied by the ideal cycle time, divided by the planned production time. In the example: 287 units × 1 minute ÷ 420 minutes = 0.683, the same 68.3%. If the two paths do not match, there is a reporting error in one of the factors.
The six big losses behind the three factors
The TPM tradition organizes efficiency losses into six classic groups, and each group attacks a specific OEE factor. Using this lens helps move from the number to the cause:
| OEE factor | Classic loss | Typical examples |
|---|---|---|
| Availability | Equipment breakdown | Functional failure that stops the machine mid-shift |
| Availability | Setup and adjustments | Product changeover, tuning, warm-up beyond what was planned |
| Performance | Microstops | Second-long jams, sensor that locks up, irregular feeding |
| Performance | Reduced speed | Machine running below the ideal cycle due to wear or misadjustment |
| Quality | In-process defect | Scrap and rework during stable production |
| Quality | Startup loss | Bad parts until the machine stabilizes after setup or a stoppage |
The practical use of the table: when OEE drops, the first step is to identify which factor dropped; the second is to identify which of that factor's two losses grew. Only then does the corrective action have an address, whether a revised preventive plan, a standardized setup or a process adjustment.
OEE, TEEP and the calendar trap
A frequent variation in shop-floor discussions is TEEP (Total Effective Equipment Performance). The difference is in the denominator: OEE measures efficiency over planned production time, while TEEP measures over total calendar time (24 hours a day, 7 days a week). TEEP answers "how much of installed capacity are we using", useful for investment decisions; OEE answers "how well did we use the time we planned to produce", useful for managing the routine.
Mixing the two denominators is a common source of numbers that cannot be compared across plants. Before comparing indices, confirm which time convention each one uses.
What is a good OEE
The most cited reference comes from the creator of the concept: an OEE of 85% has been treated as world-class since Seiichi Nakajima's TPM book, from 1984. The OEE.com portal, maintained by Vorne, tempers this yardstick with two reality checks: most factories operate with an OEE closer to 60%, and plants below 45% are more common than plants above 85%.
The practical reading: 85% is a long-term direction, not a starting target. The healthiest use of OEE is to compare the equipment with its own baseline and attack the factor that loses the most, instead of chasing an absolute number borrowed from another context.
The role of maintenance in OEE
Of the three factors, availability is maintenance's direct territory: every breakdown and every emergency corrective job consumes planned production time. That is why OEE talks to the classic planning indicators, as we show in the complete guide to industrial maintenance KPIs:
- High MTBF means fewer breakdowns, therefore fewer unplanned stoppages and more availability.
- Low MTTR means fast repair, therefore shorter stoppages when the failure happens.
- Well-scheduled preventive work shifts the intervention to the planned window, which leaves the availability denominator, instead of bursting in the middle of the shift.
Maintenance's influence does not stop at availability. Misadjusted equipment runs slower (performance) and produces more scrap (quality). A machine that "works, but not like before" can keep availability standing while quietly eroding the other two factors.
How not to inflate OEE: 5 measurement traps
OEE only guides decisions if the measurement is honest. These are the most common distortions:
- Reclassifying an unplanned stoppage as planned. The breakdown that becomes "scheduled maintenance" in the record improves the index and hides the loss. The classification criterion has to be defined beforehand, not at the moment of reporting.
- A generous ideal cycle. If the theoretical cycle used in the calculation is slower than the machine's real capacity, performance appears higher than it is. The ideal cycle should come from the manufacturer or from the best proven sustained time, not from the historical average.
- Measuring only availability and calling it OEE. Without performance and quality, the number is not OEE and it drops two thirds of the losses.
- Reporting from memory. A stoppage recorded hours later becomes a rounded estimate, and microstops disappear. The time data has to be born close to the event, in the field.
- An aggregate average hiding the bottleneck. A factory-wide average OEE dilutes the critical equipment. The index decides better when tracked by machine or by line.
How to start measuring OEE on a line
You do not need a big project to get the first reliable OEE. A lean roadmap to start:
- Choose a bottleneck piece of equipment, not the whole factory. The measurement learning happens faster at a point where the result matters.
- Define the conventions before measuring: what counts as a planned stoppage, what the ideal cycle is and where the good-parts count comes from. A written convention avoids adjusting the criterion midway.
- Ensure stoppages are recorded at the source, by the operator or the technician, at the time of the event. It is the most fragile input of the calculation and the first to degrade when reporting is late.
- Close the index per shift and review it per week. The value of OEE is in the trend and in the breakdown by factor, not in a single day's snapshot.
With one or two months of historical series, the conversation shifts from "what is our OEE" to "which loss are we going to attack first", which is where the indicator pays off the effort.
Frequently asked questions about OEE
What is OEE?
OEE (Overall Equipment Effectiveness) is the indicator of a piece of equipment's overall effectiveness. It multiplies availability, performance and quality to show how much of the theoretical productive potential became good product. It was born inside Japanese TPM and is used to locate losses from stoppages, from speed and from defects.
How do you calculate OEE?
Calculate the three factors and multiply them: availability (operating time divided by planned production time), performance (actual production divided by the theoretical production of the operating time) and quality (good parts divided by the total produced). A piece of equipment with 80%, 89.9% and 95% in these factors has an OEE of 68.3%.
What is a good OEE?
The classic reference treats 85% as world-class, a mark spread from Seiichi Nakajima's work in TPM. According to the OEE.com portal, from Vorne, most factories operate closer to 60%, with many below 45%. The useful target is to evolve against your own baseline, factor by factor.
What is the difference between OEE and availability?
Availability is one of the three factors of OEE: it measures only the time in which the equipment operated relative to what was planned. OEE goes further and also discounts the speed losses (performance) and defect losses (quality). A piece of equipment can have high availability and low OEE, running slowly or producing scrap.
Measure the share that maintenance controls
OEE points to the loss, but what gives availability back is the maintenance routine: a breakdown avoided, a fast repair and preventive work fulfilled in the window. None of that is visible without quality reporting, done at the time and place of the job.
Discover the PM Run maintenance software and schedule a demonstration with our team: you will see how real-time mobile reporting turns every work order into reliable data on stoppage, repair and availability for your OEE.
