Case study · Manufacturing · Wind energy
Every minute of downtime costs €7,000. Accuracy is how you keep the line fed.
A global wind turbine manufacturer was losing production time because stock records did not match the floor. Accuracy was never the objective in its own right here. Uptime was, and accuracy turned out to be the cheapest way to buy it.
Background
Sentispec services a global wind turbine manufacturer running several production sites worldwide. Each site holds its own local stock, partly to keep the lines running and partly to protect the profitability of equipment that is expensive to build and expensive to leave idle.
That makes the warehouse a production asset rather than a storage cost. The components sitting in the racking are there to be relieved to the line on demand, and the value of knowing exactly where each one is has almost nothing to do with the cost of counting it.
The challenge
- Low inventory accuracy was causing production downtime. When the system says a component is in a location and it is not, the line waits while somebody looks for it. The search is the visible cost. The stopped line is the real one.
- Every minute of downtime costs €7,000. That single figure sets the accuracy requirement. At €7,000 a minute, seven minutes of stopped production costs roughly what a first year site licence costs, on the published price of €9,950 setup plus €39,550 a year. The business needs 99%+ accuracy to protect output, not to tidy the books.
- Manual stocktaking could not keep pace. High value, high tempo manufacturing moves stock faster than a hand scanner count can describe it. By the time a manual count is reconciled, the floor it described has already changed.
What changed
A recording unit is picked up by a forklift the site already operates. The driver covers the aisles at normal travel speed, one shelf level at a time, and the camera reads every pallet label and registers every location, including the empty ones. The result is reconciled against the WMS and returned as a discrepancy list, with the recording behind each line available as evidence.
At the comparable Cuxhaven site the same method produced 99%+ accuracy after go live, 95% fewer stock errors and production stops eliminated at €7,000 of loss per minute, with break even reached in one month. Counting stopped being an event and became a habit, which is the part that holds the gain in place rather than restoring it once a year.
Uptime closes the case, not counting cost
Most warehouses justify automated stocktaking on the cost of counting. On the published cost model for a medium site, audit related labour falls from about €50,000 to about €5,000 and lost pallet searching from €100,000 to €200,000 down to €10,000 to €20,000. That is why most sites break even inside six months.
On a production plant the arithmetic is shorter. A single avoided stoppage of a few minutes covers the cost of counting for a year, and every euro saved on stocktaking labour is incidental. Cuxhaven, at 5,000 pallet locations, is one of the two fastest break evens in the estate despite being one of the smallest sites in it. Small warehouse, expensive line.
The counting method matters for the same reason. There is no drone to charge, no autonomous robot to route around a working shift, and nothing fitted to the forklifts, so the count itself never becomes a reason the line stops. See continuous cycle counting for how the 20 minute run is scheduled, and the five day deployment for what the first week on a live plant looks like.
Sentispec runs manufacturing sites in Germany and Poland on the same pattern. The Łódź site in Poland reached 99.99% accuracy after implementation, with 90% lower stocktaking cost and 90% fewer loading errors.
The wider picture
What comparable sites measured.
| Location | Pallet locations | Accuracy after go live | Sector | Break even |
|---|---|---|---|---|
| Cuxhaven, Germany | 5,000 | 99%+ | Manufacturing | 1 month |
| Łódź, Poland | 15,000 | 99.99% | Manufacturing, retail and fashion | 3 months |
| Duisburg, Germany | 30,000 | 99.9% | Retail and fashion | 1 month |
| Bremen, Germany | 35,000 | 99.5% | Retail and fashion | 3 months |
More case studies
Other problems, same unit.
100,000 pallets a month
Stock accuracy to 99.97%, missing pallet labour down 95%, productivity up 6 to 8% across the network.
3PL, multi customer hub3,600 hours cut to 80
Two wall to wall scans of 90,000 pallets a year across three sites, and two roles released.
Pharma, cold storeA whole warehouse in four hours
Dry, ambient and frozen zones counted by one employee, with inventory management cost down over 80%.
Next step
Put a number on what a stopped line costs you.
A 30 minute demo using recordings from a production site, plus the accuracy and downtime arithmetic for your own plant.


