OEE (Overall Equipment Effectiveness) condenses into a single number how well a machine produces compared to its theoretical potential. It is a powerful indicator precisely because it is composite: it is the product of three factors — Availability × Performance × Quality — so it doesn't just say "how much", it helps you understand "why". The problem, in SMEs, is that all too often it is calculated by hand, once a month, when it is no longer useful for deciding anything.
The three components, concretely
- Availability: the ratio between the time the machine actually produced and the time it was planned to. Unplanned stops, setups, changeovers and micro-stops all weigh on it.
- Performance: the real speed against the ideal cycle time. It is the sneakiest component, because slowdowns accumulate invisibly: the machine is "running", but more slowly than it could.
- Quality: the ratio between good parts and total parts. Scrap and rework erode efficiency as much as a stop does, yet they are often not accounted for as a loss.
An example makes it clear: a machine at 90% availability, 85% performance and 98% quality has an OEE of roughly 75% (0.90 × 0.85 × 0.98). That looks like a good number, but it means a quarter of the productive potential is lost — and OEE tells you where: in this case, mostly in performance.
The typical mistakes
The first mistake is measuring OEE only after the fact: it describes the past and doesn't let you react. The second is not coding downtime reasons: without reason codes you know the machine stopped but not why, so you cannot act. The third is leaving quality out of the calculation, getting optimistic numbers. The fourth, finally, is comparing OEE across machines or processes that aren't comparable, drawing misleading conclusions.
From the Excel sheet to real time
An OEE calculated in real time, at the machine, changes nature: it becomes a reaction tool. The operator sees the performance drop as it happens; the manager receives the alert about the recurring micro-stop; the reason code is recorded while the context is still fresh and therefore reliable. The difference from the monthly sheet is not the number itself but the granularity and timeliness: from the shift to the single cycle, from "how much" to "why", from an after-the-fact tale to a decision in the moment.
Where to start
You don't need to digitalise everything at once — in fact that is counterproductive. Start from the bottleneck machine — the one that sets the pace of the line — collect states and cycle times, code the five or six most frequent downtime reasons and watch for a few weeks. In a short time OEE stops being a meeting-room estimate and becomes a concrete lever of continuous improvement, shared between the office and the shop floor.
This is exactly what VMES45 was built for: making machine data visible, timely and actionable, even (and especially) in companies without a dedicated IT department. Get in touch to see how.
