OEE Calculator (Equipment States)

Every minute of the period is placed in one of six states: non-scheduled, unscheduled downtime, scheduled downtime, engineering, standby and productive. The tool converts all six to minutes, adds them up in a few levels, and builds every ratio from those sums.

1. Time levels. Total time is the sum of all six states. Operations time is total time minus non-scheduled time. Uptime is the time the equipment was able to run: productive, standby and engineering time together. Downtime is the time it was not: scheduled and unscheduled downtime together.
2. Ideal cycle time. The ideal rate is turned into minutes per unit: seconds per unit divided by 60, one divided by units per minute, or 60 divided by units per hour. Total units multiplied by this cycle time gives the theoretical production time, the time the units would have taken at the ideal rate.
3. Four factors. Each factor compares one level with the next one down.

AE asks how much of the period the equipment was up. OE asks how much of that uptime was spent on production. RE compares the actual pace with the ideal one while producing, and QE is the share of the output that was good.

Most terms cancel in the product, so OEE is also good units × ideal cycle time / total time: the share of the whole period that the good units alone would have needed at the ideal rate.

Three-factor view. The familiar availability × performance × quality form is the same product grouped differently:

Units per hour. Actual UPH is total units divided by productive time in hours. Theoretical UPH is 60 divided by the ideal cycle time in minutes. Their ratio is RE.

Units and rounding. One time unit, minutes or hours, applies to all six time fields; the details table always shows minutes. Percentages and UPH are rounded to one decimal; minutes and seconds in the details table show at most two decimals. OEE is calculated from the unrounded factors, so multiplying the rounded factors on screen can differ from it in the last digit.

Example, using the default inputs: total time 1440 min, uptime 1260 min, productive time 1080 min, 500 units at an ideal 30 units per hour (2 min each), 490 good. AE = 1260 / 1440 = 87.5 %, OE = 1080 / 1260 = 85.7 %, RE = 1000 / 1080 = 92.6 %, QE = 98.0 %. OEE = 68.1 %, the same as 490 × 2 / 1440.

• Engineering time is uptime, not downtime. The equipment could run; it was used for experiments or qualification instead of production. It therefore lowers OE, not AE.
• Blank non-scheduled time changes the base. With non-scheduled time left blank, total time equals operations time and OEE is measured against the scheduled period only. Entering it measures against the whole calendar period, which lowers AE and OEE. When comparing figures, check that they use the same base.
• 85 % is a reference value, not a target. Nakajima gave about 85 % as a reference for TPM, from roughly 90 % availability × 95 % performance × 99 % quality. It is not an acceptance criterion, and semiconductor equipment is usually not set against this number.
• One machine, one period. The tool does not combine several machines into a line or fleet figure.

• Entering engineering time as downtime. This lowers AE and raises OE, so neither factor shows where the time went.
• Using a historical average as the ideal rate. An average already includes slow running, so RE can rise above 100 % and the speed loss disappears. Enter the theoretical rate the equipment is built or qualified for.
• Counting shipped units as good units. Shipments can include stock made in earlier periods and leave out units still waiting to ship. Count the units from this period that met the specification.
• Mixing time units. All six time fields use the unit chosen at the top. Entering some in hours and others in minutes gives wrong totals without any error message.