Forcing a backtrack

[Nicolas]

Hello,

I have a question on managing the search during the solution of a MIP. Using callback when the search is in a MIP_NODE and the status is OPTIMAL, I retrieve the current fractional and test it against other information (in this case an archive a potentially non dominated solutions as the problem is a multi-objective one).

When the current relaxed solution has some properties compared to this archive, I would like to stop the exploration of this branch but not the search, just to backtrack. Is there a way to achieve that that escapes me ? Initially, I thought about modifying the primal bound but it is not possible as the only possibility is to provide the solution. But in my case, I have a bound (derived from the archive) but it is not associated to a solution (feasible or even relaxed).

Thank you in advance,

Nicolas.

[Daniel Espinoza]

Hi Nicolas,

Unfortunately, Gurobi does not allow to change through callbacks the branching behavior. You can use branch priorities (which are set at the beginning of the optimization).

Best,

Daniel