The OEE accuracy problem
Overall Equipment Effectiveness is the gold standard metric for manufacturing productivity. Yet most factories calculate it from manual production logs filled in by operators at shift end. These logs are estimates at best, fiction at worst. A machine that ran at 72% gets logged at 85% because nobody tracked the micro-stops.
The gap between reported OEE and actual OEE is typically 10-20 percentage points. That gap represents lost capacity, untracked downtime, and improvement opportunities that remain invisible.
How real-time OEE works
When machines are connected to the MES via OPC-UA, Modbus, or simple I/O signals, the system captures every cycle, every stop, and every quality event automatically. Availability, performance, and quality are calculated from machine data, not human memory.
- Automatic cycle counting — every part, every cycle, timestamped to the second
- Downtime categorisation — operators tag the reason, system records the duration
- Quality integration — reject data from inspection stations feeds directly into quality rate
- Shift-level granularity — compare operators, shifts, and machines in real time
- Pareto analysis — automatically ranks loss categories by impact
From measurement to improvement
Real-time OEE is not just about accuracy. It is about speed of response. When a machine drops below target, the supervisor sees it on the dashboard in seconds, not in tomorrow's report. When a quality issue emerges, the line can be adjusted before a full batch is affected. The data drives action, not just reports.
You cannot improve what you cannot measure. And you cannot measure OEE accurately with clipboards and spreadsheets.