OEE calculator

How OEE is calculated

Overall equipment effectiveness (OEE) is the share of the planned production time in which a machine or line made good parts at its ideal speed. It splits the gap to 100 % into three losses, so that a single number points to where the time went: stops (availability), slow or interrupted running (performance) and bad parts (quality).

Formulas

• Planned production time = shift time − planned stops (breaks, meetings, planned maintenance, no orders).
• Run time = planned production time − unplanned downtime (breakdowns, setups and adjustments).
• Availability (A) = run time / planned production time.
• Performance (P) = ideal cycle time × total count / run time. The numerator is the net run time: the time the units made would have taken at the ideal speed.
• Quality (Q) = good count / total count. Good means right first time: reworked units are not good.
• OEE = A × P × Q = ideal cycle time × good count / planned production time (the fully productive time over the planned production time).
• TEEP (total effective equipment performance) = OEE × utilization, utilization = planned production time / calendar time. TEEP shows how much of all the time in the period (1440 min a day) was fully productive, so unscheduled shifts and planned stops lower it.
• Times are in minutes, the ideal cycle time in seconds per unit; an ideal rate is turned into a cycle time (60 / units per minute, 3600 / units per hour). One unit per cycle: for a mould with several cavities count shots, or use the cycle time per part.

Worked example (the tool's example)

1. An 8-hour shift is 480 min. Two breaks and a meeting take 40 min, so the planned production time is 480 − 40 = 440 min.
2. The downtime list has a pump fault (22 min, breakdown), a mould change (25 min) and a colour change adjustment (8 min): 55 min of downtime, so the run time is 440 − 55 = 385 min and A = 385 / 440 = 87.5 %.
3. The ideal cycle time is 24 s and 820 shots were made: net run time = 24 × 820 / 60 = 328.0 min, P = 328.0 / 385 = 85.2 %. The missing 57 min are performance loss: 18 min of recorded minor stops and 39 min of running slower than ideal.
4. 31 of the 820 were rejects, 789 good: Q = 789 / 820 = 96.2 %. The rejects cost 31 × 24 s = 12.4 min (12 startup rejects 4.8 min, 19 production rejects 7.6 min).
5. OEE = 0.875 × 0.852 × 0.962 = 71.7 %, the same as 24 × 789 / 60 = 315.6 fully productive minutes out of 440.
6. With a calendar time of 1440 min the utilization is 440 / 1440 = 30.6 % and TEEP = 315.6 / 1440 = 21.9 %.

Six big losses

• The losses are grouped as in Nakajima's total productive maintenance: availability — equipment failures (breakdowns) and setups and adjustments; performance — idling and minor stops, and reduced speed; quality — process defects (production rejects) and reduced yield (startup rejects).
• The tool expresses every loss in minutes of the planned production time, so the losses and the fully productive time add up to it. The loss chart is a waterfall from the shift time down to the fully productive time; the calendar time only enters TEEP.
• The split inside each loss is optional: give a downtime list with the loss of each stop (Breakdown, Setup; anything else is other downtime), the minutes of minor stops that are not in the downtime record (the rest of the performance loss is reduced speed), and the number of startup rejects. Where to draw the line between a minor stop and recorded downtime (often 5 min) is a choice of each plant; keep it the same over time.

Several shifts or lines

• Each row gets its own A, P, Q and OEE. The total is calculated from the sums: OEE = Σ fully productive time / Σ planned production time, availability = Σ run time / Σ planned time, performance = Σ net run time / Σ run time.
• Quality of the total is weighted by the ideal cycle time (Σ good × ideal / Σ total × ideal), so that A × P × Q equals the OEE of the totals also when the rows make parts with different cycle times. The plain count ratio Σ good / Σ total is shown beside it.
• The mean of the row OEEs is shown too, to make the difference visible: it gives a short shift as much weight as a full one. In the example the short late shift on line B (240 min, 47.8 %) pulls the mean down to 75.0 %, while the OEE of the totals is 77.1 %.

Common mistakes

• Using the planned or average cycle time as the ideal cycle time: performance then looks better than it is (and can exceed 100 %, which the tool flags).
• Counting breaks or planned maintenance as downtime, or leaving short stops out of the record and then wondering about a low performance.
• Counting reworked units as good: quality is first-pass yield.
• Averaging OEE percentages over shifts, machines or days instead of adding the times.
• Comparing the OEE of different machines or plants as a ranking: the ideal cycle times and the loss definitions differ. OEE is best for following one asset over time.

Input and files

• Numbers may use a decimal comma or point; a comma in the fields gives commas in the results. Planned stops, downtime, calendar time and the loss split may be left empty.
• Table columns are found by their header names (Planned, Downtime, Ideal, Total, Good or Rejects, Shift / line), otherwise from the right in that order. With both Good and Rejects the good count is used and a mismatch is reported.
• Files: CSV, TSV and TXT are read as if pasted (UTF-8, UTF-16, otherwise Windows-1257). XLSX: choose the sheet and the columns or a cell range; formula cells use the value Excel last saved, dates are kept as text. Files up to 20 MB are read in this browser only and never uploaded.
• XLSX files are read with SheetJS Community Edition 0.20.3, Apache License 2.0 (license text at the end of this file).

Questions

What is a good OEE?

There is no universal target: OEE depends on the process, the product mix and how the ideal cycle time was set. A figure of 85 % is often quoted as world class for discrete manufacturing, but comparing OEE between different machines or plants says little. Use it to follow one machine or line over time and to see which loss is the largest.

Are breaks counted as downtime in OEE?

No. Breaks, meetings and other planned stops are taken off the shift time before OEE starts: OEE is measured against the planned production time. They do lower TEEP, which is measured against the calendar time.

Why can performance be above 100 %?

Only when the ideal cycle time is set too slow (too long), or when the counts or the downtime are recorded wrongly. The ideal cycle time must be the fastest cycle the equipment is designed for or has shown, not an average or a planned standard.

How do I combine the OEE of several shifts or machines?

Add up the times and counts first and calculate OEE from the sums: the sum of the fully productive time divided by the sum of the planned production time. The plain average of the OEE values gives a short or idle shift the same weight as a full one, so it is not the OEE of the whole period.