What the Overall Equipment Effectiveness Calculator does
This calculator works out overall equipment effectiveness from one shift's data: planned time, downtime, the ideal cycle time, how many pieces were made and how many were rejected. It returns availability, performance and quality separately, the OEE itself, TEEP when you give it calendar time, and the six big losses ranked in minutes so the biggest one is not a matter of opinion.
It uses the ISO 22400-2 definitions, with availability measured against planned production time rather than the calendar. That is stated on the result, because the same machine can be reported at 68% or at 23% depending on which denominator someone chose, and the two numbers get compared in meetings as though they meant the same thing.
Everything runs in your browser. Shift data stays in the tab.
How to use it
- Enter planned production time in minutes: the shift, less breaks and any period the asset was not scheduled to run.
- Enter the downtime, then the ideal cycle time in seconds per piece. The ideal cycle time is the nameplate figure - the fastest the machine can go - not the average it achieved, which would make performance 100% by construction.
- Enter the total pieces made and how many were rejected. Rework counts as a reject if it had to be touched again.
- Add the calendar time if you want TEEP. Leave it at zero and TEEP is simply not reported rather than guessed.
- Open the advanced section to split the downtime into breakdowns and setup, and to say how much of the performance loss was minor stops. Anything you do not attribute stays with breakdowns and reduced speed, so the totals always add up.
Reading the results
The three factors multiply, which is why OEE falls so fast. Three respectable-looking figures of 90%, 95% and 99% give 84.6%; three of 85% give 61%.
Availability counts stopping, performance counts running slowly, quality counts making scrap. A machine that never stops but runs at 70% of rated speed has perfect availability and a serious problem.
The loss table converts everything into minutes of planned time, which makes the six losses directly comparable. A pieces-equivalent column shows the same loss as product not made.
Because OEE equals good pieces times ideal cycle time divided by planned time, recovering a loss of N minutes raises OEE by exactly N divided by planned time. The page uses that to size the prize on the biggest loss instead of implying that the whole gap to 100% is available.
Performance above 100% is always an error, never good news. It means the ideal cycle time is set too slow or the piece count is wrong, and the page refuses to treat it as a result.
Worked example: a packing shift at 68% OEE
A packing line is planned to run 480 minutes. It is down for 60 minutes, so run time is 420 minutes and availability is 420 / 480 = 87.5%.
The ideal cycle time is 1.0 second per piece and 20,000 pieces were made, so the machine only needed 20,000 seconds - 333.3 minutes - of running to make them. Performance is 333.3 / 420 = 79.4%: for 86.7 minutes of the shift the line was running but not at rated speed.
Four hundred of the 20,000 pieces were rejected, so quality is 19,600 / 20,000 = 98.0%.
OEE is 0.875 x 0.794 x 0.980 = 68.1%.
In minutes of planned time the losses are: 60 to stopping, 86.7 to reduced speed, and 6.7 to defects (400 rejects at one second each). Reduced speed is the biggest single loss, which is the opposite of where the breakdown log would have pointed, and recovering it would add 86.7 / 480 = 18.1 percentage points of OEE.
With a 1,440-minute day, utilisation is 480 / 1440 = 33.3%, so TEEP is 68.1% x 33.3% = 22.7% - the figure that matters if anyone is proposing to buy a second machine.
Formulas and scoring rules
- Availability
availability = run time / planned production timerun time = planned - downtime.- Performance
performance = (ideal cycle time x total count) / run timeAbove 1 means the ideal cycle time or the count is wrong.- Quality
quality = good count / total count- OEE
OEE = availability x performance x qualityEquivalently, good count x ideal cycle time / planned production time.- Utilisation and TEEP
utilisation = planned production time / calendar time; TEEP = OEE x utilisation- Value of removing a loss
OEE gain = loss minutes / planned production timeFollows directly from the OEE identity above, so it needs no assumption.- Pieces equivalent of a loss
pieces = loss minutes x 60 / ideal cycle time (seconds)
The six big losses, and why two of them hide
The six losses pair up with the three factors. Breakdowns and setup or adjustment reduce availability. Idling with minor stops and reduced speed reduce performance. Process defects and reduced yield at start-up reduce quality.
Breakdowns, setup and defects get recorded because somebody has to react to them. Minor stops and reduced speed usually do not: nobody raises an event for a ninety-second jam, and nobody logs that the line ran at 80% of rated speed all afternoon because the product was sticky. Those two are therefore invisible in most downtime systems and are frequently the largest loss on the line.
This page derives them rather than asking for them. Performance loss is whatever running time was not needed at nameplate speed, and you can attribute part of it to minor stops if you have counted them; the rest is reduced speed.
OEE and TEEP answer different questions
OEE asks how well the asset performed during the time it was asked to run. It is the maintenance and operations number, and it excludes the hours nobody staffed.
TEEP asks how much of the asset's total capacity was converted into good product, calendar hours included. It is the capital number: a line at 85% OEE on one shift a day is at about 28% TEEP, which means there is a second and third shift of capacity available before anybody needs to buy another machine.
Quoting one when the audience means the other is the most common OEE argument there is, which is why this page prints the utilisation factor next to TEEP rather than folding it in silently.
Limitations: what the result does not prove
- OEE is only as honest as the ideal cycle time. Setting it to what the line usually achieves makes performance 100% and turns the whole measure into a availability-and-quality figure.
- The figure is not comparable between plants, lines or even products. Different definitions of planned time, different rules about changeovers and different cycle times make cross-comparison meaningless; comparing the same asset with itself over time is what it is for.
- It counts pieces, not value. A machine making the wrong product very efficiently scores extremely well.
- The loss split shown here is derived from the totals you entered, not measured. Attribute what you have actually counted and leave the rest unattributed rather than distributing it by guesswork.
Privacy: where your data goes
Everything you paste, type or drop is processed in this browser tab. It is not uploaded, logged, stored or sent to analytics. Session recording and tag-manager scripts are switched off on this page.
Standards and sources
- ISO 22400-2 - Key performance indicators for manufacturing operations - checked 19 Sep 2026
- IEC 60300-3-11 - Dependability management: reliability centred maintenance
- IEC 60812 - Failure modes and effects analysis (FMEA and FMECA)
Frequently asked questions
How do I calculate OEE?
Multiply three factors: availability (run time divided by planned production time), performance (ideal cycle time times total count, divided by run time) and quality (good count divided by total count). A shift at 87.5%, 79.4% and 98.0% gives an OEE of 68.1%.
What is a good OEE?
The often-quoted figures of 85% for world class and 60% for typical discrete manufacturing come from the original TPM literature and are not standards. They are also not comparable across industries or definitions. The only reliable benchmark is the same asset's own trend.
What is the difference between OEE and TEEP?
OEE measures the asset during the time it was scheduled to run. TEEP multiplies OEE by utilisation - scheduled time divided by calendar time - so it measures against all 168 hours in a week. A line at 85% OEE running one shift a day is at about 28% TEEP.
Should breaks count as downtime?
No. Breaks are normally taken out of planned production time, so they affect TEEP but not OEE. What matters is consistency between periods: moving breaks in or out of planned time changes availability by several points without anything changing on the shop floor.
Why is my performance over 100%?
Because the ideal cycle time is too slow or the piece count is too high. Performance is capped by definition: the machine cannot make pieces faster than its nameplate rate. This page flags it as an input error rather than showing an OEE above what the machine can do.
What are the six big losses?
Breakdowns and setup/adjustment (availability), idling with minor stops and reduced speed (performance), and process defects and reduced yield at start-up (quality). The two performance losses are the ones almost nobody records, and they are frequently the largest.
Does rework count as a reject?
Yes, for OEE purposes. Quality counts pieces that came out right first time, because a piece that had to be touched again consumed capacity twice. Counting reworked pieces as good hides exactly the loss the measure exists to expose.
Can I compare OEE between two plants?
Not usefully. Different definitions of planned time, different cycle-time conventions and different rules about changeovers move the number by more than any real difference in performance. Compare each asset against its own history, and compare the loss breakdowns rather than the headline figure.
Last reviewed by the A2Z.Tools team against the sources listed above.
