cost subclasses have to be rescructured according to #1875

[stap-m]

Description of the issue

We decided to make cost a process attribute, see oeo-dev 86 and #1902
Cost is a process attribute that describes the amount of money needed to buy, make, or do a thing.

Now we have to restructure its subclasses:

• which are process attributes?
• what are the corresponding quantity values?
• check existing axioms!

Ideas of solution

If you already have ideas for the solution describe them here

Workflow checklist

• I discussed the issue with someone else than me before working on a solution
• I already read the latest version of the workflow for this repository
• The goal of this ontology is clear to me

I am aware that

• every entry in the ontology should have a definition
• classes should arise from concepts rather than from words

[madbkr]

I took a look at these classes. If I am not doing this wrong - which is possible - then none of them have unique axioms associated with them except social cost (and one of those is marked as obsolete).

From the reading I did I feel like those would be process attributes:
delivery cost
investment cost (and all it's subclasses)
variable cost (and all it's subclasses)
levelised cost of electricity

While these classes don't seem to fit process attribute for me:
fixed cost because by definition those should apply to several processes and therefore not existentially depend on it.
social cost of emission because this is linked to the emission value rather than a specific process.

[stap-m]

From the reading I did I feel like those would be process attributes:
delivery cost
investment cost (and all it's subclasses)
variable cost (and all it's subclasses)
levelised cost of electricity

I agree. Maybe we can find some meaningful axioms anyway.

While these classes don't seem to fit process attribute for me:
fixed cost because by definition those should apply to several processes and therefore not existentially depend on it.
social cost of emission because this is linked to the emission value rather than a specific process.

fixed cost: There are no fixed costs without an operation or maintencance process. Therefore, I see an existential dependency of processes here, too.

social cost of emission: this looks like an inconsitency in the def and axioms. I'll take a closer look.

[areleu]

Looking at: #958 I understand that the reason social cost of emission is there is because of carbon certificates. And to my understanding in this context the value is attached to the act of emitting and not to the emission itself, thus is not tighly coupled to the emission quantity.

This is important because often in this context we speak of things like CO2 equivalents and the actual emission may have happened already in the past or it may happen in the future so I would be careful of attaching the social cost of emission to an specific process and I would rather handle it in an agent/social part of the ontology.

So my suggestion is to completely remove the axiom relating the social cost of emission from the actual emission, so neither emission rate nor emission value should be related to it. And if they are to be related to each other is through parallel processes, namely the physical act of emitting (within a time frame, i.e. a year) and the accounting of emissions during an operational time frame (also a year, for example), both of which can be parts of a wider process (i.e. the operation of a cement plant). And I think this is not so crazy, as you often see that the companies separate between operations and accounting.

Even if you look into wikipedia they classify the social cost of carbon as a marginal cost https://en.wikipedia.org/wiki/Social_cost_of_carbon , without going to deeply into the validity of this claim, it seems reasonable.

[stap-m]

So my suggestion is to completely remove the axiom relating the social cost of emission from the actual emission, so neither emission rate nor emission value should be related to it. And if they are to be related to each other is through parallel processes, namely the physical act of emitting (within a time frame, i.e. a year) and the accounting of emissions during an operational time frame (also a year, for example), both of which can be parts of a wider process (i.e. the operation of a cement plant). And I think this is not so crazy, as you often see that the companies separate between operations and accounting.

I see. Would you suggest to remove the axioms completely, or add othes instead? E.g. to a newly created accounting process of emissions? @areleu

[madbkr]

For the tasks module it may be interesting to add operating cost and maintenance cost. In my opinion those could both be children of fixed cost.

[LillyG901]

I agree with the assessment of the classes above, but I don't think anyone has discussed system cost yet.

If my understanding of the term is correct, I wouldn't classify it as a process attribute.
While different processes may add to system cost, the system itself is a continuant and may have an inherent cost.

We could maybe model it as a process attribute that is dependent on the "lifespan" of a system?

[madbkr]

The removal of the social cost of emission axioms sounds reasonable to me.

As for system cost:
It is defined as System cost are total costs of a system. I agree with Lilly that this seems to depend on a continuant - the system - rather than a process and therefore should not be a process attribute. My first instinct would be to consider it a quality but that may be wrong too.

[stap-m]

While different processes may add to system cost, the system itself is a continuant and may have an inherent cost

Ok, I see that the def suggests the relation to system directly, though being the sum of different kinds of costs (e.g. fix costs, devilery costs, ...). However, can you give an example for costs of a system that is not related to a process?
Maybe the definition needs an update to something like: System cost are the total costs that occur in a system due to processes the system participates in.? + system participates in some process

[madbkr]

I think that definition would definitely make it easier to see system cost as definitely related to a process.

[LillyG901]

I also like the new definition.

I think that it probably caputeres the idea better than what I was thinking of.

I imagined that e.g. a computer would fulfill the definition of a system and while most of it's costs, like delivery, are definitively related to processes, the materials have value by themselves. The value of the materials are of course also related to labour that was needed to harvest them/deliver them/..., but they also have a cost factor that is related to e.g. their rarity or desirability.
I'd argue that such factors are not necessarily related to processes, though I guess you could say that their desirability for example depends on current economical processes.

But I assume this isn't very relevant to the OEO anyway, so I think your proposed change of definition gets rid of this problem much more nicely.

[stap-m]

I imagined that e.g. a computer would fulfill the definition of a system and while most of it's costs, like delivery, are definitively related to processes, the materials have value by themselves.

We the idea is to distinguish between costs caused by processes, and prices referring to the (quantity) value of contiuant entities.

[LillyG901]

I see, that makes sense.
In that case I don't think it's possible to come up with an example that isn't reliant on processes.
I still think the definition change would help avoid misunderstandings.

[stap-m]

Since the discussion on social cost of emission popped up here, I close #1036

[LillyG901]

To summarize:

the following classes are process attributes and can therefore be moved as subclasses of cost:
delivery cost
investment cost (and all it's subclasses)
variable cost (and all it's subclasses)
levelised cost of electricity
fixed cost
system cost

It has been suggested to add the following classes as SubClasses of fixed cost for use in the task-module:
operating cost
maintenance cost

social cost of emission needs to be handled separately (maybe in a social part of the OEO), but we are still waiting on feedback of whether or not we should replace its axioms with new ones or delete them entirely.

If we can agree on the last two points I would start to implement.

[stap-m]

It has been suggested to add the following classes as SubClasses of fixed cost for use in the task-module: operating cost maintenance cost

There is a separate issue #1974

social cost of emission needs to be handled separately (maybe in a social part of the OEO), but we are still waiting on feedback of whether or not we should replace its axioms with new ones or delete them entirely.

I put that on the agenda for the next oeo-dev meeting.

[LillyG901]

I moved social cost of emission alongside the other subclasses for now, because I assumed it would be easier to find as a part of the hierarchy it currently belongs to instead of leaving it as a quantity value