ThingConnect

Free tool

OEE calculator

Enter one shift’s numbers for one machine and get availability, performance, quality, and OEE — with the math shown, not hidden.

OEE71.9%
Availability
87.5%
Performance
85.7%
Quality
95.8%

A result in the 40–60% range is typical for plants measuring honestly for the first time. The gap to your target is the hidden capacity you already own.

How the calculation works

Availability = run time ÷ planned production time. Run time is planned time minus all stops. Whether breaks and planned maintenance count against availability is a policy decision — pick one definition and apply it everywhere.

Performance = (ideal cycle time × total parts) ÷ run time. It captures speed losses: slow cycles, micro-stops, reduced feeds. If it comes out above 100%, your ideal cycle time is set wrong — a common and quietly corrosive problem, because it hides real speed losses.

Quality = good parts ÷ total parts, counting only right-first-time as good.

OEE is the product of the three. The point is not the headline number — it is which of the three components is eating your capacity, because each has a completely different fix.

This calculator is for a single machine over a single period. ThingConnect computes it continuously, for every machine, from controller data — with your plant’s own availability rules.

Questions

Frequently asked questions

OEE = Availability × Performance × Quality. Availability is run time divided by planned production time. Performance is ideal cycle time multiplied by total parts, divided by run time. Quality is good parts divided by total parts. Each is a ratio between 0 and 1, so the product is the share of planned time that produced good parts at the rated rate.

Same three-factor formula as any OEE calculation, but the inputs need CNC-specific definitions: planned production time should have a fixed, written policy on whether tool-change time and scheduled maintenance count against it, and ideal cycle time should come from a demonstrated sustainable rate, not the CAM program's theoretical time — otherwise performance loss absorbs errors that have nothing to do with actual machine speed.

There's no universal target. The commonly cited 85% figure comes from Total Productive Maintenance literature describing high-volume discrete manufacturing, and it doesn't transfer cleanly to a job shop running short batches with frequent setups. The more useful question is whether OEE is stable or improving for a given machine and part mix, measured the same way every time — an absolute number is only meaningfully comparable against itself.

Almost always a definitions mismatch: planned production time that includes or excludes breaks inconsistently, micro-stops too short to make it onto a paper log, or an ideal cycle time pulled from the program instead of a rate the machine actually sustains. Reading machine state directly from the controller removes the manual-logging gap, but the definitions still need to be agreed once and held constant.

Against a rated cycle time you can defend and reuse every period. Leave ideal cycle time at the theoretical program value and performance loss ends up absorbing everything the number can't otherwise explain, which makes OEE stop being diagnostic. Set it from a demonstrated sustainable rate instead, and performance loss keeps meaning one specific thing: the gap between what the machine can do and what it did.