A return calculation for a production line looks simple: investment cost divided by annual benefit. Its reliability depends entirely on the inputs. Most of them are estimates.
A process model does not replace the investment analysis. It replaces some of its assumptions with verified values, and makes clear which ones remain assumptions.
Required output and takt
The starting point is the output the investor requires. Takt follows from it:
Takt = available working time / required output. 90 pieces per hour in continuous work gives a takt of 40 seconds. A lower required output gives a longer takt, and potentially a simpler and cheaper line.
If the investor is considering several output levels, the sizing of the line is checked against each of them.
From gross capacity to effective output
Gross capacity assumes that the line never stops and makes only good parts. The usual measure of the gap between this and reality is OEE, the product of three factors:
- availability: failures, changeovers and waiting for material;
- performance: micro-stops and running below nominal speed;
- quality: scrap and rework.
Three factors of 90% each give about 73% OEE, so a line with a gross capacity of 90 pieces per hour effectively delivers about 66. These percentages are for illustration, and the actual values depend on the process. New lines also do not reach their target immediately. The ramp-up period, during which output is lower and scrap higher, should be included in the first year’s figures.
What the model verifies, and what it does not
| Input to the investment case | Source | Role of the model |
|---|---|---|
| Takt and cycle times | Process design | Verified on the running process |
| Bottleneck station | Process design | Identified, variants can be compared |
| Buffers and pallet count | Process design | Checked for their effect on output |
| Staffing per station | Process design | Tasks defined, to be confirmed in the plant |
| Equipment list | Design, stage 2 | Parts list with manufacturer codes, as a basis for supplier quotes |
| Effect of scrap on flow | Assumed rate | Line behavior checked at the assumed rate |
| Actual availability and scrap | Plant experience | Not determined by the model |
| Required output | Investor | An input: the model checks whether the process reaches it |
| Labor, energy and maintenance cost | Finance, plant | Outside the model, the model only provides quantities |
| Market assumptions, financing, discount rate | Investor | Outside the model |
The model is strongest where traditional analysis relies on the supplier’s declaration or on engineering judgment.
Technical risk in the calculation
Investment cases often contain an implicit assumption that the line will reach its nominal parameters on schedule. A model reduces this technical risk in concrete ways:
- the process is confirmed to reach the output before machines are ordered;
- logic errors are removed before commissioning, which reduces work on site;
- changes are made while they are still design decisions;
- suppliers quote against an unambiguous specification, which reduces scope differences.
We do not put a percentage on these effects. Published figures vary widely, and they depend on the project. It is more reliable to treat them as risks to assess in the customer’s own analysis: how long commissioning usually takes in this plant, and what a week of delay costs.
How to use the results in a decision
A useful investment case presents ranges and not a single number. In practice:
- Fix the required output, and each level of it if the investor is considering several.
- Derive the required takt for each level, and check whether a single line design covers them.
- Take capacity from the model, then apply an OEE assumption and a ramp-up curve that fit the plant.
- Price the equipment on the basis of the parts list and supplier quotes.
- Check sensitivity: what happens to effective output if OEE is 10 points lower, or if ramp-up takes twice as long.
Staged decisions
The split into stages also limits the amount at risk. Stage 1 is a relatively small commitment compared to the investment, and it ends with a complete result. If the analysis shows that the investment does not hold, the decision to stop is taken before the costs of detailed design and equipment.
What we provide for the analysis
From stage 1, the achieved output, bottlenecks and the effects of the variants studied. From stage 2, the equipment list as a basis for quotes and the signal list as a basis for pricing the control system. The market figures, margins and financing assumptions remain with the investor. We can review the technical assumptions in a finance model, but the financial conclusions belong to the people who are responsible for them.
