This article is for plant managers, production managers, industrial controllers and continuous improvement leads who talk about the same line in different units: minutes and OEE points on the shop floor, cost per part in finance. It explains how to translate stops, micro-stops, changeovers and scrap into cost per part, and how to feed cost accounting with measured machine time rather than standard times.
- Unit cost often relies on standard times, not on the time actually consumed.
- Every minute of stop, micro-stop or changeover has a cost you can calculate.
- Spreading those minutes over good parts explains the cost variance, cause by cause.
- Measured machine time makes cost accounting more reliable.
What cost accounting expects from the shop floor
Tree Partners, a French chartered accounting firm, published Expert-comptable industrie : pourquoi la comptabilité analytique est clé on its website, written by Ali Tazi (24 June 2026). The article covers the accounting specifics of manufacturing: the weight of inventory on the balance sheet, the choice of valuation method, the rational allocation of indirect production costs and, above all, calculating unit cost by product, customer, line or site, which, according to the author, requires an information system that tracks resource consumption. It separates margin on direct cost, margin on production cost and net margin, and concludes plainly: “Une comptabilité qui ne reconstitue pas ces marges ne sert pas à piloter, elle sert seulement à clôturer.” (in other words: an accounting system that does not rebuild these margins is not a steering tool, it only closes the books).
The accounting method is in place. One question remains that an accountant cannot settle alone: how much machine time did each product really consume? If that time comes from a theoretical routing or a hand-entered average, the calculated margin is right on paper and wrong on the floor. That is where OEE measurement becomes financial data.
Standard time or measured time: where the gap comes from
In many plants, unit cost relies on a nominal rate and a loading assumption set once a year. Yet the time actually spent producing a part number depends on what happens on the line: breakdowns, material waits, adjustments, micro-stops of a few seconds, slow running, changeovers that overrun, parts scrapped on restart. None of these events appear in the standard time.
Controllers know the result: a production cost variance that shows up at month end, booked as a lump, with nobody able to say which line, which part number or which cause produced it. Micro-stops are the most misleading: too short to be logged, they can add up to a lot over a shift. Changeovers are often spread evenly across part numbers, even though a short run carries far more setup time per part than a long one.
Calculating the cost of a minute of downtime, step by step
Translating OEE into money starts with one simple value: what one minute of the line costs. Here is a five-step method to build with your controller.
- Set the hourly cost of the line. Add up the costs that run whether the machine produces or not: depreciation, assigned labour, base energy, maintenance, share of overheads. Divide by planned production time.
- Derive the cost of one minute. Hourly cost divided by 60. This is the cost of a lost minute, whether it is lost to a stop, a micro-stop, a changeover or making a part that ends up as scrap.
- Add lost margin if the line is a bottleneck. On a saturated line, a lost minute is also a lost sale. On a line with spare capacity, the cost is limited to the resources consumed. That distinction completely changes which projects come first.
- Measure lost minutes by cause and by part number. Stops, micro-stops, changeovers, slow running, scrap: each OEE loss category is a number of minutes you can put a value on.
- Spread those minutes over good parts. The real cost of a good part is the cost of planned time divided by the number of good parts. The gap to standard cost then breaks down cause by cause.
Example: breaking down the cost of a part
For example, take a line with a full hourly cost set at 300 euros, or 5 euros per minute, a nominal rate of one part per minute, on a 480-minute shift. The figures below are illustrative: they show the method, not a real situation.
| OEE loss (example) | Lost minutes | Cost per shift | Cost per good part |
|---|---|---|---|
| Stops (breakdowns, waits) | 50 min | 250 € | 0.71 € |
| Changeovers | 40 min | 200 € | 0.57 € |
| Micro-stops and slow running | 30 min | 150 € | 0.43 € |
| Scrap (10 parts) | 10 min | 50 € | 0.14 € |
| Total losses | 130 min | 650 € | 1.86 € |
In this example, the line makes 350 good parts instead of 480, an OEE of about 73%. The shift costs 2,400 euros whatever the output. The machine cost of a good part goes from 5 euros in theory to 6.86 euros in reality, including 1.86 euros of losses. If the budget routing assumed 85% OEE, the standard cost would be about 5.88 euros: the gap of nearly one euro per part now has named causes. For scrap, lost material must be added on top, which accounting already tracks.
This reading changes decisions. A SMED project can be priced in cost per part on short runs. A micro-stop plan becomes an investment comparable to any other. And the margin by part number, which cost accounting is trying to rebuild, rests on the time actually consumed.
Feeding cost accounting with measured time
For production data to serve cost accounting, a few simple rules are enough: the same loss breakdown in engineering and in finance, times linked to the part number and work order produced, a month-end close that uses the measured times of the period rather than routing times, and a joint review of the cost variance by production and finance. Standard time remains useful for budgeting; measured time explains the variance and shows where to act.
TEEPTRAK measures OEE, stops and micro-stops in real time. A module wired in the electrical cabinet and a sensor at the edge of the line record each machine’s activity automatically, with no connection to the PLC. Installation takes about one hour per machine and OEE is available within 48 hours. Stop, changeover and micro-stop times are timestamped and assigned a cause, giving controllers measured machine time to reconcile with their allocation keys. The costing method remains that of the company and its accountant; TEEPTRAK provides the measurement that feeds it. The solution is deployed in more than 700 plants in more than 30 countries.
Further reading
- Expert-comptable industrie : pourquoi la comptabilité analytique est clé (Tree Partners, in French)
- Cost of downtime: how to calculate your losses and protect your profitability
- Value stream mapping with measured machine data: filling in the data boxes
FAQ
Should measured times replace standard times in unit cost? Not necessarily. Standard time remains the reference for the budget and inventory valuation, depending on the method chosen with your accountant. Measured time is used to analyse the variance, link it to causes and correct standards that are persistently wrong.
How do you put a value on a minute of micro-stop? Like any lost minute: at the cost of one minute of the line, plus lost margin if the line is a bottleneck. The hard part is not the calculation but the measurement, because micro-stops are too short to be logged by hand.
How should changeover cost be split between part numbers? By the time actually measured for each changeover, assigned to the part number being started. An equal split across part numbers underestimates the cost of short runs and overestimates that of long ones.
0 Comments