Tobias Achterberg

You are adding variables with model.addVar() from within your callback? This just cannot work. What you have in your hand as a user is the original user model. But the branch-and-cut is performend ...

This should get fixed with our new Gurobi 10.0 release, which is scheduled for first half of November 2022.

This is a delicate issue. We only promise that we will be able to detect the following two cases as convex: 1. x^T Q x <= d with PSD matrix Q2. sum xj² - y² <= 0 with y >= 0 But Gurobi will also be...

You need to add each of your constraints with m.addConstr().

In your formulation, I think that you always need to use (1 - bi) for all the big-M constraints as you have formulated them.

I think that your big-M constraints are wrong. For example, b1=0 will force I[v][i] <= 120*SOC_avg[v][i] due to the first big-M constraint. This is not what you want. You only want to force constra...

For example, say you have four different regions based on the relationship of x and y in which you have different objective functions. Then you need four binary variables to indicate in which regio...

Yes, you can do this manually. But you can also just use the "and" general constraint: addConstr(z == and_(x,y)). Moreover, you can just stick with your quadratic objective function as indicated in...

In very rare cases it may even help to also use barrier for the local node relaxations. There is of course the very big downside that barrier cannot benefit from the warm start information that you...

Hi Austin, yes, your observation is correct. The fix in Gurobi 9.5 just means that those solutions are discarded from the solution pool. It does not mean that they are already pruned in the search ...