Which costs per component belong in a robust calculation?

A unit cost calculation forces you to allocate all relevant process steps to a quantity produced or processed. This makes visible the ancillary times that easily disappear in simple comparisons: set-up, positioning, inspection, filter changes, rework or internal transport.

The technical productivity of a process is already described in the article How does laser ablation work when cleaning metal?. This article is solely about developing a commercially usable per-unit calculation from it.

Why unit costs are easier to compare

Hourly rates are useful for many calculations.

They can be misleading, however, if two processes have different ancillary times or unit output rates.

An example:

Machine A costs €100 an hour to run and processes ten components.

Machine B costs €150 an hour and processes twenty components.

Looking at the hourly rate alone would make machine A appear cheaper.

Per component, however, the result is:

A: €10 of machine costs.

B: €7.50 of machine costs.

Set-up time, labour and rework are still missing – but even this simple example shows why the unit “cost per released component” often provides the better basis for a decision.

Where component sizes differ, costs per square metre, per metre of processing length or per batch can be used instead.

The only important thing is that the same functional unit is compared.

Separating fixed and variable costs

A robust calculation starts with a simple separation.

Fixed costs

They arise regardless of how many components are actually processed.

These can include:

• Depreciation or financing
• certain insurance policies
• Basic floor space costs
• Parts of the maintenance
• Training costs

Variable costs

They rise with use, or with the number of units.

These can include:

• Energy
• Filters
• Wear parts
• Consumable media
• active staff time
• Disposal

This separation matters because, as utilisation rises, fixed costs are spread across more units.

A company with 20 productive hours a year therefore has different unit costs from the same company with 1,000 productive hours.

The machine investment must therefore not be judged in isolation from the planned use.

Machine costs per productive hour

For a simple internal calculation, the annual machine cost burden can be divided by the expected productive hours.

The annual burden can include:

• Depreciation or financing
• regular maintenance
• imputed interest
• fixed service costs

The result is an internal machine cost rate.

The important word is productive.

If a system would theoretically be available for 2,000 hours but actually only has 400 hours of orders, the fixed costs should not be spread artificially across 2,000 hours.

That would present the unit costs as too low.

With a new technology in particular, the calculation should be conservative.

The question is:

How many productively chargeable or internally value-adding hours are realistic?

Labour per component

The next block is operating time.

A distinction should be made here between:

• Set-up
• active processing
• Component changes
• Inspection
• Cleaning

Not every process ties up the employee for the entire machine running time.

An automated process can run for ten minutes, for example, while the operator takes on another task in parallel.

With manual hand-held processing, by contrast, the employee is typically tied up directly.

For unit costs this means:

Machine time does not automatically equal staff time.

Both figures should be recorded separately.

This also shows whether automation could be worthwhile cost-effectively.

Spreading set-up time across the batch size

Set-up time has a particularly large influence with small batch sizes.

Suppose a system needs 30 minutes to set up.

Processing then takes five minutes per component.

For a single part, the total time is 35 minutes.

With 100 identical parts, by contrast, the half hour of set-up time comes to just 18 seconds per component.

Batch size therefore changes unit costs drastically.

For companies with many one-off jobs, a short and flexible set-up is therefore particularly valuable.

In series production, by contrast, a more elaborate initial set-up process can be cost-effectively acceptable.

This logic applies regardless of the process used.

Energy and media consumption

Energy is a classic variable cost item.

Actual consumption should be measured wherever possible and not derived from the nominal laser power.

The wattage of the laser source is not the same as the total electrical energy consumption of the system.

Additional loads can be:

• Cooling
• Extraction
• The control system
• Peripherals

On the other side, other processes also have energy and media consumption.

Compressed air, for example, can cause a considerable energy demand.

Wet chemical processes can additionally require drying or temperature control.

Fraunhofer therefore examines industrial cleaning technologies explicitly from ecological and economic resource perspectives.

For the per-unit calculation, the relevant figure is:

Total energy and media consumption divided by the units processed.

Filters, wear and maintenance

Laser-based removal requires no conventional abrasive or blasting medium.

Even so, there are still consumables.

Process extraction, for example, requires filters.

Optical protective components have to be checked and replaced where necessary.

Maintenance costs can also arise.

There are two options for a per-unit calculation.

Recurring costs can be spread annually across the expected number of units.

Consumption-dependent components can be calculated on the basis of real replacement intervals.

Example:

A filter costs €200 and lasts for 2,000 components under the specific conditions.

The simplified filter item then comes to €0.10 per component.

With other coatings, the same filter may be saturated considerably sooner.

Practical operating data are therefore more valuable than blanket assumptions.

Inspection and quality assurance

In economic terms, a component is not finished when the machine stops.

It is finished when it has reached the required condition and can be released.

That can include an inspection.

Depending on the application, this can be:

• purely visual
• A dimensional check
• A roughness measurement
• A check for residual coating
• documented process control

This time also belongs in the unit costs.

In a series application, inspection can be automated or carried out on a sampling basis.

With individual parts, every component can be checked separately.

If this effort is not taken into account, processes with a higher inspection requirement appear artificially cheap.

Rework as a hidden cost item

Rework is particularly critical because it is often not attributed to the original process.

For example, a component is processed first.

An employee then notices that coating is still present in some places.

They re-insert the component and correct the surface.

In the production statistics, the main processing time could still appear as the original machine time.

In economic terms, however, the rework is part of the same job.

On productivity, Fraunhofer ILT explicitly stresses that a process should not be judged on maximum output alone. Low rework and stable quality are just as decisive.

For the unit cost calculation this means:

Record the average rework time per component.

Do not measure only ideal parts.

Judging scrap correctly

Scrap can be even more expensive than rework.

If a high-value component can no longer be used because of an unsuitable process, the costs are not only those of processing.

The value of the component itself can also be lost.

For established, stable processes the scrap rate may be very low.

During an introduction phase, or with widely varying applications, the risk should nevertheless be considered.

With high-value components in particular, slower but more robust processing can make more economic sense.

Optimising purely for seconds per part then falls short.

Batch size changes the calculation

The unit costs of a process are not constant.

They change with the batch size.

Large series benefit from:

• distributed set-up time
• stable parameters
• feeding that can be automated
• repeatable movements

One-offs and small series, by contrast, benefit more from:

• flexible set-up
• simple programming
• quick adjustment
• low fixture complexity

A company should therefore not calculate only a theoretical unit price.

Useful figures are, for example:

• Costs at batch size 1
• Costs at batch size 10
• Costs at batch size 100
• Costs at a typical annual quantity

This shows from what volume onwards a process develops its economic strengths.

Utilisation and idle time

Even a perfect unit cost calculation can become misleading if the system is barely utilised.

Fixed costs continue to run.

Machine costs should therefore never be calculated solely on the basis of the maximum possible number of units.

Real demand is what counts.

Anyone also wanting to use a system for external customer orders should plan internal and external utilisation separately.

Possible additional revenue is attractive, but it is not yet guaranteed capacity utilisation.

For a conservative investment calculation, the applications that are certain should carry it first.

A calculation model for practice

A simple unit cost formula can be built up as follows:

Machine costs per part

plus

Labour costs per part

plus

Energy per part

plus

Consumables and filters per part

plus

Quality inspection per part

plus

average rework

plus

other relevant costs

=

Total cost per released component

The same target condition is then calculated for the existing or alternative process.

Only these two values should then be set against each other.

For a fair comparison, quality must be defined identically.

What data a material test should provide

For a unit cost calculation, a material test should not merely produce an attractive before-and-after result.

It should provide measurable data:

• Size of the test area
• Processing time
• Set-up effort
• Number of passes
• parameters required
• rework required
• target condition achieved

The differences between pulsed lasers and CW lasers are already covered in the corresponding technology comparison.

For unit costs, all that matters is which system does the job within the required quality and at what real process times.

Conclusion: the finished component is the right unit of calculation

Which costs per component belong in a robust calculation?

More than the machine minute alone.

A complete unit cost calculation takes into account:

1. The fixed costs of the system
2. real utilisation
3. Set-up time
4. Machine time
5. Staff
6. Energy
7. Filters and wear
8. Inspection
9. Rework
10. possible scrap costs

Particularly where processes differ, the unit cost calculation can produce surprising results.

A more expensive machine hourly rate can be more cost-effective thanks to higher output.

A slower main process can work out cheaper thanks to less rework.

A high investment can produce very low unit costs where utilisation is sufficient.

It is therefore not the cheapest machine that is decisive.

In surface processes, what is decisive is not the cost of the machine minute but the cost of the released component.

Sources

• Fraunhofer ILT – “More productive with light” — a view of productivity through cycle times, ancillary times, rework and stable process quality.
• Fraunhofer IGCV / Fraunhofer IVV – assessment of industrial cleaning technologies — economic and ecological analysis of resources and process parameters.
• Fraunhofer Business Unit Cleaning – cleaning technologies — different processes positioned along industrial cleaning tasks.

Related content