NSGA-II, NSGA-III and exit variables

I have a problem that have 8 output variables. My problem is required to use the maximum of the profit and use the minimum of the area (output variables). 4 areas have better productivity and have 0 cost. The other 4 are a lower produtivity, and have a value to use, different from the first 4. Also, all areas have a limit to use. I played my code in NSGA-II and NSGA-III and the better areas was not using at 100%, but the other areas have at least 60% to use, since the are worst, because have a lower productivity and have a cost.

A lower productivity in this case means more use in areas to maximize profit. A higher productivity means less use of areas to maximize profit. Again, the objetive is maximize profit and minimize the sum of output variables (areas).

So, why NSGA-II and NSGA-III is not using 100% of the areas which have no cost and have a better productivity, and using some of the areas that have a less productivity and have a cost to use? Is related to how that algorithms works?

Another important detail: the problem does not have a restriction related to minimum ammount of each area to use, but have a restriction related to maximum use of the area.

I need to search for an answer to my question. I have doubts to how using NSGA-II and NSGA-III and I study that algoritms from only a few time.

Hello @RedRHPS ,
Welcome to the KNIME Community!

Would it be possible for you to share the workflow with the data or with dummy data(if your data is confidential) so we can better understand the problem?


1 Like

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.