OEE Cost Structure
OEE Cost Structure
The OEE cost structure is based on the OEE loss-structure and thus follows the line of settings in OEE Coach.
For every item that we define and track in OEE Coach, we can now determine its cost-effect.
In the description below we will go through the loss-structure and describe -block by block- how it should work.
[A] 24 x 7 overhead cost
TEEP looks from this perspective.
A machine, simply by being present, is costing money 24 x 7. Used or not.
Those are the base-costs. Whatever time-frame is taken to look at, the base-cost per hour are always there.
Such cost include the following cost-items:
1. Finance & Insurance
Those are cost related for financing the equipment and fulfilling obligations like insurances.
Example:
The machine has a lease of 12.500 €/month.
A month has on average 730 hrs
The hourly financial base costs are 12.500€ / 730hrs = 17,12 €/hr
2. Housing & Facilities
The machine is placed in a building, consuming space.
It is connected to electricity, gas, pressurized air. Such connections cost money.
There may be security measures and surveillance.
The building may be heated or cooled permanently.
Assuming the total monthly costs of all those facilities are 6000€,
that means 6000€ / 730hrs = 8,22 / hr
3. Corporate Overhead
In an average factory, on every operator there is appx 1 other person working as indirect labor. These employees need workspace. All indirect costs, not related to the machine, are called ‘overhead’
The factory has 10 machines. The total non-machine/production related costs are 5 Mio € per year.
We assign 10% to our machine: 500k / year.
This gives an overhead for this machine of 500k / 12 = 41.667 €/Month
41.667€ / 730hrs = 57,08 € / hr.
4. Where are 24/7 costs stored
The cost of overhead is stored in the mastertable Machine. The values are given PER YEAR.
OEE Coach will calculate the values to HOURS and add the result to each calendar hour assigned to a machine (Scheduled, Not-scheduled or Un-scheduled).
Time Frame
OEE Coach will determine the timeframe by the filter that is applied: The smallest time frame will be 1 day in order to be always able to calculate TEEP. Other filters applied will use the same calculations.
[B] Scheduled vs not-Scheduled time
- In the timeframe of the filter the ‘scheduled time’ is all the registered time: the sum of all shift-durations within the time-frame.
- The difference between total time in the filter and the scheduled-time is the non-scheduled time
- The last nightshift in te filter will go over the filter limit (shift of 23 March ends at 24 March). This is compensated with the ‘gap’ that is created at the beginning of the filter:

[C] Shifttime vs Unscheduled time
Within a ‘normal’ shift, time can be unscheduled (activity = [U]).
In that case the remaining shift-time consists of all activities of types [P]+[W]+[F]+[L]
Unscheduled time is a registered activity type and behaves as the other activity type described in block [D]
When time in a shift is being UNscheduled, normally the operator goes elsewhere. However when he is present and can not work at another machine, the cost will go on. Therefore, OEE Coach allows actual operators to be registered during UNscheduled time.
[D] Activities: [Production] [Waiting] [Failure] [Line-restraint]
In this block we find the regular activities as registered in the OEE Toolkit.
In the MAP is (per activity) defined how many operators are planned to be scheduled (PO). In the shift-data the actual amount of operators per activity is registered.
The calculation uses the ACTUAL amount of operators (Op).
The [cost per operator per hour] is stored in master Team. Depending on the salary of the team-members, this might fluctuate per team
The normal operator costs are calculated as: (Activity Minutes x (CostOperatorHour/60)) x ShiftFactor
(ShiftFactor: See [G])
The activity itselves might lead to extra costs;
e.g. A [P]roduction run might consume extensive energy that does not occur while waiting.
A certain [W]aiting might bring extra cost due to expensive tools or personel needed.
To allow such variances, for this activity an extra cost per hour can be added in the MAP.
These extra costs are calculated: Activity Minutes x (CostExtra/60)
[E] Speed Settings
Speed settings
- Max Speed: Comes from MAP
- Set Speed: Default form MAP, overruled in Shift Data
- Actual Speed: To be derived from Shift Data
[F] Performance: Actual Output – Minor Stops – Reduced Speed
To be derived from Shift Data
[G] Shift Factor
The hourly rate per operator per hour is the regular rate (defined per team). However for shifts in the weekend and night, usually there is an extra cost on top of the normal rate.
Per shift a factor can be given. If the nightshift is 25% more expensive as normal, the factor would be 1,25.
The factor for a normal shift is 1.
The cost factor of a shift is applied to all activities with operator(s):
(Activity Minutes x (CostOperatorHour/60)) x ShiftFactor
[H] Conversion Costs
So far all cost to operate the machine have been defined.
The results of the calculated costs from block A – G are added
Since at this point we know the total amount of produced items, we can calculate the conversion cost per item: Total conversion costs / total output.
[I] Output: Good – Scrap – Rework – SubSpec
The produced output is identified in four quality categories:
- [G]ood: the product being in spec, it can be sold.
Rejects: The product does not meet the specifications:
- [S]crap: It has to be completely rejected (Landfill, burned, …)
- [R]ework: It can pass the process (or a rework process) once again to be ‘repaired’
- [SubSpec]: It does not meet the spec, however it can be sold as a lower spec product (‘B-choice’).
Output is counted and qualified at the end of the machine/process, usually by sensors or manually.
(In the simulator a percentage of the total output is given)
Each has its own cost-profile, consisting of the following elements:
Cost factors
- Conversion costs:
At this point in the process all costs to convert the product have been spent. Whatever the result of the conversion process may be (G-S-R-SS) the conversion cost will be the same.
- Material costs:
In most cases, at this point all raw materials have been added to the product and the cost will mostly be the same for each category of output. However it may be possible that the product has been rejected due to missing parts or elements, in which case the cost of added materials may be less. Or more when more product has been used.
- Extra Handling:
Good product never has EXTRA handling. But rejected product cannot follow the normal process. And this usually brings extra costs. E.g. getting rid of scrap product may cost 800-1200 € per 1000Kg to dump or burn it.
Also reworking brings extensive extra cost since the product has to follow a separate route and being processed once more. Even SubSpec (the product can be sold, however against a lower price) will have an extra cost component since it needs an extra logistical route that needs to be managed and it may need some (re)labeling of (re)packing.
Value
Besides the different costs of the different sorts of output, there is the actual generated value of this output.
1. The value of the [G]ood product: this is what we all worked so hard for.
This value can be established in several ways:
- The actual sales price
- The intercompany transfer price
- internal calculated value etc…
2. The value of [S]crap: we invested cost in this product but in the end we cannot sell it. In some cases there is a way to get some money for the scrap e.g. the price of Kg metal. But usually the value is zero.
3. The value of [R]ework: when the product can be re-processed we ‘gain’ the cost of raw material. Be careful! A commonly seen mistake is that the value of the reprocessed materials is validated LOWER than the original raw material. If this would be true, you should consider making only rework product and sell it as cheap raw material in the market (yes, nonsense!). Meaning: the value of reworkable material should normally never be lower than the value of the original raw material.
Example: From plastic grains you are making plastic film, however the film is too thin. The film can be shredded and the shredded pieces can perfectly be used instead of grains. To gain this value you have to add the cost of extra handling being the logistic process and the cost of shredding.
4. The value of [Ss]ubspec: The product does not meet the intended specification, but can be sold for its lower specifications at a lower price
Example: you intended to make 80% alcohol, however the output was alcohol 70%. Makes a fine alcohol, but not at 80%. And can perfectly be sold for its 70% at a lower price.
Extra cost of handling may apply.
Value per category
Per category of output P-S-R-SS we know the amount produced and the value that was assigned to each item. Multiplying those, results in the value per category
Result per category
Value per category minus cost per category = result per category
[J] OEE – OOE – TEEP
Here the regular KPI’s are calculated as usual
[K] Cost – Result summary
This is the ultimate result of our calculations:
Part 1
Based on the realized amount of output per category and the calculated cost and result per item,
this overview shows the results per category, leading to the total realized result for the whole calculated period.
Part 2
What we want is GOOD products. The rest is ‘collateral damage’.
If we now take all the costs and all the results and we assign those to the GOOD products only, this shows the best possible indication what the financial impact of this specific pile of products really had.
Of course it is important to assign costs and result as correct as possible to the categories.
But even if those are rough estimates, the analyses shows unexpected (or even astonishing…) results when comparing several situations. The actual results in euro’s may not be correct, the effect of changing one of the parameters however will be very clear since the change of the results will clearly indicate the scale of the effect.