Micro-Stops & the Hidden Factory: The Losses Your OEE Doesn't Show
Key takeaways
- A micro-stop is a sub-5-minute stoppage that's cleared in seconds and almost never logged.
- Together they form the hidden factory - 5–20% of capacity, invisible to your reports.
- They hit performance, not availability, so they're hard to attribute and easy to ignore.
- You can't fix what you can't see: finding them needs automatic detection + the true cause.
Every plant has a second factory hiding inside the first one. It runs on the same machines, employs the same people - and produces nothing. It's built entirely from stops so short that nobody writes them down. This is the hidden factory, and micro-stops are its bricks.
What is a micro-stop?
A micro-stop (or "small stop") is a brief, usually operator-cleared interruption - typically under five minutes, often under one. A carton jams. A sensor false-trips. A bottle tips on the conveyor. The operator clears it in fifteen seconds and the line runs again. No work order, no log entry, no trace.
The problem isn't any single stop. It's the frequency. Ten micro-stops an hour at 30 seconds each is five minutes lost every hour - over 8% of your capacity, gone, with nothing in the report to show for it.
Enter your stop frequency and rate to see the lost units, euros per year, and the OEE points it's hiding.
Why they hide from OEE
This is the crucial part. In the OEE model, a logged stop reduces availability. But a micro-stop is too short to log - so its lost time doesn't reduce availability. Instead it silently drags down performance (the line produced fewer parts than its run time should allow).
Performance loss is a vague bucket. It blends speed loss and small stops together, with no breakdown of cause. So the loss is real, it's in your OEE - but it's unattributable. That's why measured OEE typically runs 10–18 points above reality: the small losses are present in the math but invisible in the analysis.
What they cost
A moderate example - 10 stops/hour, 30 seconds each, a line running 3,600 units/hour at €0.40 margin, 4,000 operating hours/year:
| Metric | Value |
|---|---|
| Lost time per hour | 5.0 min (8.3%) |
| Lost units per year | 1.2 million |
| Hidden OEE points | ~8.3 |
| Annual hidden loss | €480,000 |
And these defaults are conservative - most plants undercount both how often micro-stops happen and how long they last. Pair this with the cost of downtime and the picture gets sharper still.
How to find them
You can't clipboard your way out of this - the stops are too short and too frequent for manual logging. Finding the hidden factory takes two things working together:
- Automatic detection. Read stops directly from the machine (PLC), so every interruption is captured - including the one-second ones.
- The true cause. Detection alone gives you a count; you need to see what happened. Machine vision captures the moment of each stop, turning "performance loss" into "the labeller jams on condensation."
Once each micro-stop has a cause, the hidden factory becomes an ordinary fix list - usually a handful of recurring failure modes that account for most of the loss.
Fabrico detects every stop from the machine and uses computer vision to show the true cause - automatically, no manual logging.
Frequently asked questions
What's the difference between a micro-stop and a breakdown?
Duration and logging. A breakdown is long enough to record and usually needs maintenance; a micro-stop is cleared by the operator in seconds and rarely logged. Breakdowns hurt availability; micro-stops hurt performance.
Are micro-stops the same as speed loss?
They're related but distinct. Both reduce the performance factor of OEE. Speed loss is running below nameplate rate continuously; micro-stops are brief halts. Together they make up the performance losses among the six big losses.
How much can fixing them gain?
Because micro-stops commonly cause 5–20% performance loss, addressing them is often the single biggest OEE lever available - and it usually costs less than chasing big breakdowns.
← More OEE Lab guides · Open the Hidden-Factory Calculator →