I'm designing a iteration algorithm in which I need to solve a MILP model during each iteration.Although the initializaion goes well (with warning 'max constraint violation (2.4386e-06) exceeds tolerance'),the question comes as the model turns out infeasible after iteration.
I tries to turn off the presolve function and that works.But new question comes as I find that some variables doesn't observe the constraints I set,which results in the mistake of final objective computing.
I really want to know why the 'presolve' parameter would affect the feasibility of model and whether exites some solutiuons to keep the model feasible and correct.
Thanks in advance!
Please sign in to leave a comment.