Multi Agent Demand Share

[Mia McLachlan]

Hi all, 

Thank you for the meeting this morn. We had a quick question when running with multiple agents:

We want to represent the different population size of the agents, considering that not all women get a say on the choice of stove and there is a much larger rural population than urban population. We were thinking the way to do this would be by changing the quantity value from 1 in the Agent file. 

Is this correct and if so can we just use relative values to reflect the different demand shares?

Best wishes,

Avi & Mia

[Alexander Kell]

Hi both,

That's correct.

Say Agent A1 made up 30% of the population, you would set Quantity to be 0.3 in both the new and retrofit agents. 0.7 if you wanted the second agent to make up the rest of the population (70%).

Thanks,

Alex

[Mia McLachlan]

Aha great, thank you!

[Mia McLachlan]

We tried this with two agents. We wanted to check that it was taking into account the populuation so we did a run with only one agent (A1) and then did a run with 2 agents with the population A1= 1 and A2=0 to see if we would get the same result. However, there are capacity additions corresponding to both agents and the total capacity has been doubled. Do we need to multiply the raw data by the population split?

Thanks, 

Avi and Mia

[sara giarola]

When you have two agents, they are in fact always 4, as each one would come with a retrofit and a new one. The technodata file needs to associate the capacity to the retrofit agent of each agent you are modelling and using the same split of demand

Normally yu would run into errors if you have agents with zero capacity.

[Mia McLachlan]

Even when we adjust to A1=0.9 and A2=0.1, the total capacity for each year is doubled (this is just considering the retrofit type).

[Alexander Kell]

Could you share your model inputs? I've just tested this using your other case study and don't seem to be getting the same behaviour.

Thanks,

Alex

[Mia McLachlan]

Yes sure, here is the Agent File. I can attach all the files if need be!

[Alexander Kell]

Thanks for this. If you could upload the two scenarios with 1 agent and 2 agents so I could compare, that would be helpful. I don't seem to be getting a doubling of capacity.

[Mia McLachlan]

Yes sure, I have uploaded all three files: A1, A2 and 2Agents. Just to explain what I mean by 'doubling capacity', this (2Agents.JPG) is a screenshot from the MCACapacity.csv which attributes the same capacity for both agents which when summed is doubled that when considering agents individually (A1.JPG).

[Alexander Kell]

Thanks for highlighting this. There is an oversight in the documentation which I will amend. 

Basically, as Sara mentioned, your agent file should begin like this:

AgentShare,Name,AgentNumber

Agent1,A1,1

Agent2,A1,2

Agent3,A2,3

Agent4,A2,4

Then you should go into the residential/Technodata file and amend it so the end of the file looks like this (you do not need Agent1 or Agent3 in the technodata file):

Agent2,Agent4

Retrofit,Retrofit

0.9,0.1

0.9,0.1

0.9,0.1

0.9,0.1

0.9,0.1

0.9,0.1

0.9,0.1

Where, in this example Agent A1 makes up 90% of the population and Agent A2 makes up 10% of the population. You have to keep this consistent with the population share in the Agents file.

This should then fix your problem.

Please let me know if any of this is confusing!

Thanks,

Alex

[Mia McLachlan]

Yes great, that seems to be working now, thank you!