OEE & bottleneck:
Gain capacity without investment
Setup times, micro-stops, cycle losses – the largest free capacity sits in the machines you already have. Process mining exposes where it is lost, and optimisation brings it back.
01 The problem – The most expensive machine is the one that waits
OEE losses are scattered across many small causes
An OEE of 68% means: almost a third of possible production does not happen. The loss is rarely one big block – it sits in setup times, micro-stops, reduced speed and ramp-up losses, scattered across shifts and machines.
That is exactly why reporting does not show it. Process mining reconstructs the actual material flow from MES and PPS data and reveals where the bottleneck truly is – often not where you suspect it.
Capacity gained on the bottleneck machines – without a single new machine.
02 The model – The real process, not the target layout
Process mining on machine events, bottleneck detection across the throughput chain
Ø OEE Engpass-Anlagen: 68.0% -> Ziel: 74% Gewonnene Anlagenstunden/Jahr: 2.160
Process mining shows the bottleneck is not static: depending on product mix it shifts between machines. Optimisation therefore prioritises where the bottleneck stands most often – and setup-time reduction at that point acts directly on overall throughput.
03 Business impact – Capacity as value
The lever: bottleneck hours gained × contribution margin per hour
No round number: every assumption comes from the sample plant and is stored centrally. With your real figures only the input changes, not the method.
| Item | Value |
|---|---|
| Bottleneck machines | 6 |
| OEE increase: 68% → 74% | +6 pp |
| Baseline: Hours gained (6 × 6,000 h × 6 pp) | 2,160 h |
| Lever: Contribution margin per machine hour | €280 |
| Result: capacity value / year | €604,800 |
Assumptions of a sample plant – in a real project your data replaces these values.
Bottleneck hours gained are not an end in themselves: they mean additional orders without extra investment or pulling delayed deliveries forward. Valued with the contribution margin per hour, this becomes a clear euro figure.
04 Next steps
From the bottleneck picture to ongoing control
We use the machine events your MES already logs – without new sensors.
For one production line the real process model is built, with named loss sources and optimisation levers.
Live view of the moving bottleneck and the biggest setup and cycle losses – as a decision basis for shift management.