Where the missing minutes go
When OEE comes from a shift report, short stops vanish. An operator clears a jam in forty seconds and nobody writes it down. Do that sixty times a shift and the line has lost forty minutes that the report says it ran.
Speed loss hides the same way. A line running a little under its ideal cycle time looks busy from the aisle. Only a comparison against the ideal cycle, part by part, shows the gap.
What happens when you connect the line
The first week a line reports from its own PLC states, measured OEE typically lands several points below the figure everyone was used to. That is not bad news. It is the first honest list of where the capacity went.
The new number also arrives during the shift, not after it. A supervisor who can see availability falling at ten in the morning can still do something about it.
Start with availability, and with reasons
Take stop events from the machine, then ask the operator for the cause from a short list of eight or ten reasons. Long lists get ignored, and free text cannot be counted.
Within a month you will have your three biggest losses ranked by minutes. Those three are usually worth more than any new machine, and they tell you which improvement project to run first.
In short
- Reported OEE and measured OEE are rarely the same number.
- Capture stops from the machine and reasons from the operator.
- Rank losses by minutes and fix the top three first.