Calculate your OEE, and see where it leaks.
Two calculators on one page. The classic ISO 22400-2 method needs an ideal cycle time; the MachineWise method needs only what a machine can tell you. Both price every loss in rupees.
THREE TOOLS, ONE ARGUMENT · NOTHING YOU TYPE IS SENT ANYWHERE UNTIL YOU ASK FOR A REPORT
The textbook calculation.
Availability × Performance × Quality, with performance measured against an ideal cycle time. This is the formulation defined in ISO 22400-2, and the basis of the widely quoted world-class benchmark.
One shift, one machine
Method (ISO 22400-2): Availability = run time ÷ planned production time. Performance = (total parts × ideal cycle) ÷ run time. Quality = good parts ÷ total parts. Losses are converted at your machine hourly rate. This is one machine on one shift — MachineWise computes the same three factors continuously across every machine on your floor, with each stop already attributed to a reason.
The calculation a machine can actually answer.
This is the MachineWise variant, not the ISO 22400-2 definition. No ideal cycle time, no standards to maintain. Every input here is something MachineWise measures directly from the machine — which is why it survives short batches, mixed part numbers and long-cycle work that break the textbook method.
Measured inputs
Reporting time = gross period − known breaks. Availability = machine on-time ÷ reporting time. Performance = productive (in-cut) time ÷ machine on-time. Quality = approved parts ÷ parts produced. The three losses plus fully productive time always add back to the reporting time exactly — no unattributed remainder.
Same three factors. Different question.
You run long, stable batches of one part
High volume, a well-maintained standard cycle time, and a genuine need to compare against the 85% world-class benchmark or a customer's OEE clause. Classic performance catches a machine running below its rated speed — the MachineWise method will not.
Your floor is job-shop reality
Short batches, many part numbers in one shift, long-cycle work, or standards that were last updated years ago. It needs no ideal cycle time, so it cannot be gamed by an optimistic standard — and it works from day one, on every machine, without setup.
They answer different questions
Classic asks how fast. MachineWise asks how much of the paid-for time actually made parts. A machine can score 100% MachineWise performance while cutting slowly — which is exactly why the two numbers are worth seeing side by side.
What each loss actually means
| Factor | Classic OEE | MachineWise OEE |
|---|---|---|
| Availability loss | Machine scheduled but stopped — breakdowns, changeovers, waiting | Machine not powered on during reporting time |
| Performance loss | Running slower than the ideal cycle time | Powered on but not in cut — setup, idling, waiting, operator absence |
| Quality loss | Rejected parts | Rejected parts — identical |
| Standard | ISO 22400-2, using an ideal cycle time per part | No — everything is measured |
| Comparable to ISO-based benchmarks? | Yes | No — treat it as your own baseline and improve against it |
You have the number. Now price it, then justify fixing it.
See these three numbers live on your own machines.
A free two-machine pilot measures your real availability, performance and quality for thirty days — then hands you the report and the ROI model.
Book my free pilot → WhatsApp usHow the number is worked out.
Run time ÷ planned time
Planned production time minus downtime, as a share of planned time. Breaks and planned stops are excluded before you start.
(Parts × ideal cycle) ÷ run time
How close the machine ran to its ideal speed while it was running. Minor stops and slow cycles live here.
Good parts ÷ total parts
First-pass good output. Scrap and rework both count as loss.