Problem on Modelling Uncapacitated SAHLP

Problems with modeling
Post Reply
demir
User
User
Posts: 1
Joined: 1 year ago

Problem on Modelling Uncapacitated SAHLP

Post by demir »

Hi everyone,

As part of my project I need to model an uncapacitated single allocation hub location problem on GAMS. I don't have a lot of knowledge on GAMS, but I tried to model by myself anyways. On CAB10 dataset, I was able to get a solution (I don't know if the solution is true though).
CAB10.gms
The model that works
(3.99 KiB) Downloaded 94 times
However on CAB25 dataset, I get the error "CPLEX Error 1217: No solution exists. --- Problem is integer infeasible". Can anyone help me what would be the issue with my model? Thanks in advance.
CAB25.gms
My model that doesn't work
(14.76 KiB) Downloaded 93 times
I tried to model this (1)
I tried to model this (1)
I tried to model this (2)
I tried to model this (2)
User avatar
bussieck
Moderator
Moderator
Posts: 1033
Joined: 7 years ago

Re: Problem on Modelling Uncapacitated SAHLP

Post by bussieck »

Verifying that a model is implemented correctly is not easy. If you think that the model should not go infeasible for your second data set, there are ways to identifying the source of infeasibility. Then you can decide if this is is okay to be infeasible because of the data or if you can blame a bug in the equation algebra. Trouble is that you need to be the expert of the your problem to do this debugging. So you can't expect much help from outsiders. Here we can only help you to identify the source of the infeasibility. If you search the forum for infeasbility you find many pointers to feasopt, iis etc. In your case I would try to construct any feasible solution for your case, set the level values of the variables before the solve and check in the equation listing which equation is infeasible for this "feasible" solution.

-Michael
Post Reply