Ruibing Wang

Hi Jaromił,Thank you very much! I have reformulated the model, which is non-convex MINLP, and it works well in Gurobi. Now, I convexified the model and obtained a convex mixed integer version. Hwev...

Hi Riley,Thanks. If the solution quite depends on seeds, does it mean the global optimum is not trustworthy?Best,Ruibing 

Hi Jaromił,Many thanks! However, when I did a sensitivity analysis under the ru model, I found that for some instances, the “gloabl” optimum is still different for different seeds. Here are the res...

Hi Jaromił,Many thanks! Unfortunately, I am not able to reformulate the other model in the same way as in the ro case. I therefore re-checked the ru model carefully and revised all potential zero-d...

Hi Jaromił,Many thanks for pointing out the strange behavior! The model that I sent you yesterday (conic_model2) corresponds to one case – let’s call it the ru case. I I also have another model, co...

Hi Jaromił, Many thanks for your help! Here is the newly reformulated model conic_model2.mps. I applied your suggestion and reformulated the denominators. Since I upgraded to Gurobi v13, I only hav...

Sure. Here is the model file: conic_model.mps. Thank you again.

Thanks. I have upgraded to v13. I compared the solution of Model B with the additional constraint obtained via v12 and v13. For v12 and v13, my solver setting is:ampl.setOption('solver', 'gurobi') ...

Thanks. I still do not fully understand your answer to my second question, so let me clarify what confuses me. My objective is to maximize revenue. After I added an additional constraint to Model B...

Also, after I added a constraint to the model, the objective from Gurobi increases. Does this mean the solution from Gurobi is not global optimal?