I'm currently building a model, where it seems to be running without any dummy imports.
However, I just recently noticed that I have a capacity of ships, which the first year of running meets the demand exactly, but still meets the next without investing in new capacity even though i have created a demand projection, which states a larger demand.
To specify a bit more: I have calculated it so that the capacity exactly meets the demand the first year, therefore you could expect that the model would have to invest in new capacity already the next year.
Is there any explanation to this? And furthermore, anyway to solve this problem?
Hard to say without additional info. Have you checked that there are no dummy imports satisfying the growth in the demand?
Could you post a picture showing all the process attributes, and also the demand projection?
I have added an Excel file, which shows the technology used, new technologies which are available and the demand projections.
It also includes the attributes used.
But as you can see the capacity drops over the first year, and without investing in new technologies it is still able to supply the demand (Even so it should no be, from the given attributes). I'm not sure where the problem is though.
To that I also have an additional question. So I have tried to make a sensitivity analysis on the side line, on hydrogen (TRABRNT), which is only able to supply from 2020. Here i try to increase the price of the technology seen in the Excel sheet by 5% in different scenario files, using the attribute: INVCOST, and just overwriting the old INVCOST in the NewTechs file. However, when I increase the price of Hydrogen, it suddenly invests in new capacity in 2016, for the HFO.
Which i do not understand.
I tested your model, using exactly the data you provided.
In the results, after 1 year I get an output from TSEACEHFO exactly as you said it should be: 0.21203582.
Therefore, I cannot reproduce your findings: The existing capacity is not satisfying the full demand, but produces only a part of the demand. The rest of it is satisfied by the new technologies.
If you want me to look at it in more detail, please provide me with the full model (*.dd, *.run files).
Ok, thank you for providing me with the full model *.dd and *.run files, for taking a closer look.
I ran the case defined in the model files, and obtained the results shown in the following table (Excel pivot table):
As you can see, the existing capacity process TSEACEHFO produces the demand commodity DTSEAC 0.22319560 in 2015, and then one year after it produces 0.21203582 in 2016. This is exactly as expected, and exactly what you said yourself in your file it should be producing. And subsequently, the output from TSEACEHFO decreases along with the decreasing capacity (the capacity is assumed to decrease linearily because the Stock value is defined only for a single year).
The remaining demand is satisfied by the new technologies. And no dummy imports. So, all looks fine to me.
Consequently, unfortunately it seems that I am still not able to reproduce your findings. But I can see the model is quite badly scaled, and your problem could be due to that. Can you tell me which solver you are using? I would suggest to try using an aggressive scaling option, if such is supported by your solver.
