I golf maybe once a year and go through 20-plus balls per round. But my score always looks decent — because I take mulligans, accept gimme putts, and may have been known to fudge my handicap. Go by the numbers alone, and I look like a respectable golfer. In reality, I'm terrible.

OEE can work exactly the same way.

When world-class numbers hide real problems

Walk into some plants and you'll see OEE dashboards running at 88%, 91%, 93%. Hourly targets are beaten most of the shift. Management is satisfied. On paper, the plant is world class.

But before we hand out that title, let's check for mulligans.

The mulligan: running through breaks without adjusting scheduled time

If a line runs through a scheduled break or lunch — and nobody updates the scheduled minutes in the OEE calculation — those extra production minutes disappear. The denominator stays the same. OEE goes up. But the performance measurement didn't actually get harder; the plant just didn't count all the time it had available. That's a mulligan.

The fudged handicap: ideal cycle time set below actual maximum capacity

By definition, consistently beating your hourly target shouldn't be possible. OEE Performance is calculated against the ideal cycle time — the theoretical maximum rate. If operators are regularly running above target, the ideal cycle time is set too conservatively. It may have been based on early line projections, customer demand rates, or a number someone picked years ago and never revisited. The machine can go faster. The target just doesn't know it.

Plants don't game the system intentionally. The shop floor changes constantly — schedules shift, lines speed up, breaks get skipped — and the OEE parameters simply don't keep pace.

The result: comfortable numbers that hide real opportunity

The danger isn't fraud. It's false confidence. A plant running an 89% OEE that's actually 74% isn't doing anything dishonest — it's just operating with a measurement system that drifts over time. The losses are real. The improvement opportunity is real. But neither surfaces in the reporting.

What plants need is a tool that keeps the measurement honest. Something that:

  • Detects production occurring during unscheduled time (breaks, lunches) and automatically adjusts the scheduled minutes in real time
  • Uses actual collected cycle time data to determine the historically best hour of production — and flags when the ideal cycle time in the system diverges from that reality

That tool exists. Ask me about it.

But please don't ask me about my golf game.