English is currently available.
Industrial Engineering DISCUSSION
I have been asked to report OEE for a packaging machine. For one shift I have: 480 minutes shift length, 30 minutes of planned breaks, 45 minutes of breakdowns and changeover, an ideal cycle time of 0.5 minutes per unit, 700 units produced and 20 of them rejected.
How are the three factors calculated from these, where do changeovers and breaks belong, and what does it mean if my performance figure comes out above 100 percent on some shifts?
Work down from the shift. Planned production time = 480 - 30 = 450 min. Run time = 450 - 45 = 405 min. Availability = 405 / 450 = 90.0 percent.
Performance = (ideal cycle time × total count) / run time = (0.5 × 700) / 405 = 86.4 percent. Quality = good count / total count = 680 / 700 = 97.1 percent. OEE = 0.900 × 0.864 × 0.971 = 75.6 percent.
A quick check: OEE is also good count × ideal cycle time / planned production time = 680 × 0.5 / 450 = 75.6 percent.
Where things belong: planned breaks, and any time the machine is not scheduled to run, are taken out before OEE is calculated, so they do not count against it. Breakdowns are availability losses. Changeovers are also normally counted as availability losses, even though they are planned, because they are time the machine could have been producing; hiding them removes the incentive to shorten them.
Short stops of a minute or two and reduced speed usually go unrecorded as downtime, and that is what the performance factor captures. Rejects and rework, including start-up scrap, are quality losses.
Performance above 100 percent means the ideal cycle time is set too long. It should be the fastest rate the machine can sustain for that product, from the design specification or the best demonstrated rate, not a budget or a historical average. If the standard is loose, performance is inflated and the OEE figure hides a real speed loss.
The standard is per product. A machine running several pack sizes needs an ideal cycle time for each, and the numerator becomes the sum of ideal cycle time × count over the products run in the shift.
Use the three factors separately rather than the single number. The same 75.6 percent can come from a machine that breaks down a lot or from one that runs steadily but slowly, and the actions are completely different. In your shift the biggest loss is performance: 405 minutes of running should have given 810 units and gave 700, so 55 minutes' worth of output went in minor stops and slow running, more than the 45 minutes of recorded downtime.
For the same reason, be cautious comparing OEE between different machines or plants, since the definitions of planned time and ideal rate are rarely identical.
Back to Industrial Engineering community
