A month has 720 hours. Your line truly produces in 331.
Of the month's 720 hours, 600 are planned for production — and even those leak. What OEE is, how to calculate it in 4 steps and why every efficiency point changes the cost of each ton that leaves your plant.
One number to answer: how much of the time became good product?
OEE (Overall Equipment Effectiveness) measures how much of the planned production time actually turned into good product, at ideal speed. It combines three questions into a single calculation:
The month's 720 hours, loss by loss
A month has 720 hours. Remove the time with no production scheduled (weekends, no demand, planned cleaning and maintenance) and 600 planned hours remain. See how they “leak” until only what truly produced good product is left.
Why 600 and not 720? OEE measures efficiency during the time the line should produce. Comparing with the 720 calendar hours is another KPI, TEEP (total effective equipment performance): 331 ÷ 720 = 46%. TEEP shows how much capacity the plant still has available.
Two losses almost nobody sees in time
Speed: the loss that never turns on a red light
The machine doesn't stop — but it doesn't produce at the speed it could either. No alarm goes off, nobody calls maintenance. The loss only shows up at the end of the shift, when output doesn't add up. In the example, 90 hours a month disappear this way.
Quality: the most expensive loss of all
When a product is scrapped at the end of the process, it takes with it everything already invested in it — not just machine time, but raw material, packaging, energy and the labor of every step. In the example, 66 tons a month cost all of that and never reach the customer.
Three facts about OEE
OEE was popularized by Seiichi Nakajima, the “father of TPM” (Total Productive Maintenance), as a way to expose the hidden losses of equipment.
The famous “world class” benchmark comes from 90% availability × 95% performance × 99.9% quality.
Plants that start measuring rigorously often find their OEE in this range — well below what the team expected.
The 6 big losses that drag OEE down
Each one hits one of the three factors. Knowing which is biggest on your line is half the battle.
Breakdowns and failures
Unplanned stops caused by equipment failure.
Setup and adjustments
Format and product changeovers and tuning.
Minor stops
Jams of seconds or minutes that almost nobody records.
Reduced speed
The line runs below its ideal rate “to avoid trouble”.
Start-up scrap
Bad product until the line stabilizes after a stop or changeover.
Scrap and rework
Out-of-spec product during normal production.
How to calculate OEE in 4 steps
Using the same line: 600 planned hours, ideal rate of 2.2 tons per hour.
Availability
450 h ÷ 600 h = 75%Running time divided by planned time.
Performance
792 t ÷ (450 h × 2.2 t/h) = 80%What was produced divided by what the ideal rate would give.
Quality
726 t good ÷ 792 t ≈ 92%Good product divided by total produced.
OEE
75% × 80% × 92% ≈ 55%Or directly: good product ÷ the maximum possible (600 h × 2.2 t/h).
Same line, same cost — more tons
The line costs practically the same per hour whether it runs well or badly: people, energy, depreciation, overhead. Let's assume $700 per hour — that is, $420k per month. What OEE changes is how many tons share that bill.
The detail your controller will ask about: that 15% is absorbed cost — fixed cost spread over more tons. It becomes cash savings when the extra output is sold or when the same output is made in fewer hours (fewer shifts, less overtime and energy). In the model of a real reference operation, this meant about 92 hours freed per line per month.
On this line, each OEE point is worth about 13 extra tons a month with the same structure. Multiply by the plant's lines and OEE stops being a production KPI and becomes a financial one.
Calculate your line's OEE and cost
Move the three factors and the line data. It's the same math we used above.
Simplified simulation: treats the hourly cost as fixed. For a full calculation of your plant, talk to our consulting team.
What 20 years of measuring OEE in real plants taught us
OEE theory fits in a formula. What separates a nice number from a real gain lies in details that only show up after many lines, plants and countries.
OEE is a rule before it is a number
What counts as a stop, what the ideal rate is, how changeovers are treated: the definition changes the result more than the operation itself. Standardize the rule before comparing lines or plants.
A “realistic” rate hides the biggest loss
When the speed target is adjusted to what the line already does, the performance loss disappears from the KPI — but stays in the cost. Use demonstrated capacity, not the comfortable one.
Minor stops only show up with automatic data
With manual reporting, 1- or 2-minute jams are almost never recorded. Added up, they become hours. PLC or sensor capture usually reveals losses the team didn't see.
Average OEE is not the average of percentages
Adding the lines' OEE and dividing by the number of lines distorts the result. Consolidated OEE must be time-weighted — or recalculated from hours and volumes.
OEE doesn't move alone
Our research with real data confirmed OEE's association with safety, quality and maintenance. Preventive work on time, CIL done and trained operators show up in the KPI.
Sustained gains come from routine, not from the screen
The +10 to +14 point gains we measured came from system and management: daily meetings with the stop Pareto, actions with owner and deadline, and the same rule every day.
20 years on the shop floor · plants of world leaders
What you hear on the shop floor
“Our OEE is 90%, we're great.”
It is almost always a sign of incomplete calculation: unrecorded minor stops or an underestimated ideal rate. Honest OEE usually scares at first.
“Two equal OEEs can hide different problems.”
55% with low availability calls for maintenance; 55% with low quality calls for process work. That's why OEE is always broken into its three factors.
“Planned downtime counts in OEE.”
It depends on the company rule. Periods with no planned production are usually left out — what matters is defining the rule and applying it to every line.
Common questions about OEE
What does OEE stand for?
OEE stands for Overall Equipment Effectiveness. It shows how much of the planned production time became good product at ideal speed, by multiplying availability, performance and quality.
What is the OEE formula?
OEE = Availability × Performance × Quality. It can also be calculated directly as good output divided by the maximum possible output in the planned time at the ideal rate.
What is a good OEE?
The world-class benchmark is 85%. In practice, what matters most is measuring consistently and tracking progress: plants that start measuring rigorously often find values between 40% and 60%.
How does OEE affect product cost?
Since much of a line's cost is fixed per hour, raising OEE means producing more tons with the same structure. In this article's example, going from 55% to 65% cuts the absorbed cost per ton by about 15%.
Do I need software to measure OEE?
You can start with a spreadsheet, but it misses minor stops, depends on retyping and doesn't show results in real time. A MES records stops and output on the shop floor, validates the data and calculates OEE on the spot.
Want to see your plant's OEE like this, in real time?
ProMOS® MES records stops, losses and output shift by shift and calculates OEE on the spot — with gains of up to +14 points documented in real plants.
