Most people learn OEE as a formula: availability times performance times quality. That is correct, but it hides the more intuitive way to understand it, as a story about time. Every shift starts with a block of available time, and a series of losses chips away at it until only a fraction is spent making good product at full speed. The OEE time model maps that journey, and once you see it, where to focus your improvement becomes obvious.
OEE is what survives after each layer of time loss is subtracted.
Picture time as a stack that gets smaller at each level:
Plant operating time — all the calendar time available.
Planned production time — what remains after planned shutdowns (no shifts scheduled, holidays, planned maintenance). This is the time OEE actually judges you on.
Run time — planned production time minus availability losses (breakdowns and changeovers). The ratio here is Availability.
Net operating time — run time minus performance losses (minor stops and slow running). The ratio versus run time is Performance.
Fully productive time — net operating time minus quality losses (defects, rework, startup scrap). The ratio is Quality.
OEE is fully productive time as a share of planned production time, the slice of planned time spent making good units at full speed, first time.
Each drop between levels is caused by specific losses, the well-known six big losses:
Availability losses: breakdowns (unplanned stops) and setup/changeover.
Performance losses: minor stops/idling and reduced speed.
Quality losses: production defects and reduced-yield/startup losses.
Seeing losses against the time model tells you not just that time was lost, but at which layer, which points straight to the fix.
A frequent source of argument is what counts as "planned" time and therefore whether it should hurt OEE. Planned shutdowns (no demand, scheduled maintenance) are excluded from planned production time, so they do not lower OEE, that is what TEEP is for, since it judges against all calendar time. OEE focuses on how well you used the time you intended to produce. Consistent definitions of these buckets are essential, or two teams will report different OEE for the same shift.
The time model turns a single OEE percentage into a diagnosis. A low number caused mostly by availability losses calls for different action than one driven by speed or quality losses. Without capturing where time goes, the percentage is just a verdict with no explanation, and unrecorded losses become dark data you cannot act on.
Fabrico captures machine data in real time and automatically attributes lost time to the right layer and loss category, so you see exactly where your minutes go, breakdown, changeover, minor stop, slow running or quality, rather than a bare number. That turns the OEE time model from a textbook diagram into a live map of where to act.
A hierarchy showing how total available time is reduced layer by layer, plant time to planned production time to run time to net operating time to fully productive time, by successive losses, with OEE as the final share.
Run time minus performance losses (minor stops and slow running). Comparing it to run time gives the performance factor of OEE.
No. They are excluded from planned production time, so they do not reduce OEE. Their effect on total capacity is captured by TEEP instead.
See exactly where your minutes go. Fabrico attributes every lost minute to the right loss, turning OEE into an action plan. Book a demo today.
Login
Blog
