Market share per region

[Younis]

I would like to introduce a user constraint to impose market shares per region. By that I mean bounds on the production of a commodity defining the contribution of each region to the total production of the commodity from all regions.

For example, a model with two regions REG1 and REG1, each has two processes PRC1 and PRC2, producing commodity COM1. The condition should be that 20%-30% of the production of COM1 should come from REG1 (regardless of which process), whereas the remaining 70%-80% should come from REG2. 

As per the attached screenshot, my thinking was to use UC_COMPRD for COM1 combined with UC_FLO for the target regions and processes, while summing over regions (UC_Sets: R_S). However, the intended effect is not achieved. The constraint is not met. I wonder if this is a problem with the logic of the constraint formulation or it is more about the syntax.

Your feedback and tips are much appreciated.

[Antti-L]

As far as I can see, your LO constraint can be read as follows:

  SUM(r∈{REG},p∈{'PRC1','PRC2'}, VAR_FLO(r,t,p,'COM1'))  ≥
    0.2×VAR_COMPRD('REG1',t,'COM1') + 0.7×VAR_COMPRD('REG2',t,'COM1')

So, you are summing the output flows of COM1 over all regions and processes PRC1,PRC2, and require that sum to be higher than or equal to the sum of the commodity productions in REG1 and REG2 multiplied by 0.2 and 0.7, respectively. If I read that correctly and that is not what you intended, then you indeed have a problem with the logic of the constraint formulation.

Based on your description "20%-30% of the production of COM1 should come from REG1 (regardless of which process), whereas the remaining 70%-80% should come from REG2" I think you would need two LO equations, such as the following:

   VAR_COMPRD('REG1',t,'COM1')  ≥ 0.2 × SUM(r∈{REG}, VAR_COMPRD(r,t,'COM1'))
   VAR_COMPRD('REG2',t,'COM1')  ≥ 0.7 × SUM(r∈{REG}, VAR_COMPRD(r,t,'COM1'))

The corresponding UP equations would then be redundant, and thus not needed at all in this 2-region case (or vice versa).
These constraints would be straightforward to specify under a VEDA UC_T table. I hope that might help you.

[Younis]

Hello Antti, thanks a lot for your response and for pinpointing the shortfall of my initial logic. Just an update for anyone interested in this topic: I took some time to formulate and test your suggested approach using COMPRD variable only. However, there remained an issue that, as far as i understand, it affects both gross production (including import), whereas I wanted the constraints to exclusively limit domestic production. So I applied your suggested formulation but then on VAR_ACT for the domestic production processes, and it seems to work quite well.

[Antti-L]

Yes, correct, COMPRD does involve total commodity production, including any imports. I am glad you found a satisfactory solution referring to domestic production only.