OEE needs downtime reasons and output definitions. Stabilize reporting and downtime capture before OEE boards, or numbers will not drive action.
Part 1 — OEE factors—availability, performance, quality—each need capture points.
Part 2 — Downtime reason lists must be short and actionable; past thirty options nobody picks seriously.
Part 3 — Calibrate shift output and good-part definitions before ranking boards.
Prioritize verifiable scenarios: main path, auditable permissions/logs, and clear interfaces.
| Decision lens | Common trap or cost | Safer cut |
|---|---|---|
| Reporting/downtime first | Reliable definitions | Then OEE |
| Short reason list | Actionable classes | Better fills |
| Morning meeting loop | Named owners | Numbers become action |
Trusted data before management theater
OEE without auto-capture?
Approximate yes, with accuracy limits; plan sensors/PLC later.
Low OEE always equipment?
No. Changeovers, material waits, and QC holds often dominate.
One plant-wide target?
Use bands by line type; a flat 85% is usually politics.
Link to lean tools?
OEE shows loss structure; SMED/TPM supply methods; the system locks the habit.