Mario Ruthmair

Hi Eunseok Kim, As you are already using NumericFocus=3 you seem to be aware that your model might suffer from numerical issues. Actually the coefficient ranges are quite large and you should think...

Hi Daniel, Indeed, formulating this function in a Gurobi model is a bit involved. First, you do not need addGenConstrPow() for quadratic terms, those are directly supported in the constraint formul...

Thanks, I got the files! We will continue the discussion in the ticket.

Hi Zheren, Would you be willing to share the complete model such that we can reproduce the difference with IntegralityFocus=0 vs. 1 on our side? If yes, please create a support request (and refer t...

Hi Ximeng Fan, As often, the answer is: It depends! The first question is what "large-scale" actually means for you (in terms of numbers of variables and constraints)?We have seen models with milli...

Apart from that, your model seems numerically quite challenging. You could try to set parameter NumericFocus=1 (or 2 or even 3).

Hi Zheren, If both solutions are feasible for your problem, and the solution from the complete model solved by Gurobi is clean (no significant violations) and better then I would say that your Bend...

Hi Zheren, If I understood your last post correctly, then your solution values of binary variables only deviate by a value less than 1e-5. If this is correct, then this is expected behavior. You ca...

This line should definitely not appear in your license file. I guess with a correct license file you should be fine.

Hi Zheren, Based on your post it is hard to get a complete picture of your setup.I guess you have a complete model with binary variables that you solve with Gurobi.Additionally, you decompose the m...