Tobias Achterberg

Can you say something on how many iterations each loop in your code has? How many value pairs does big.iterrows() return? As far as I can see, for each value pair you generate 3600*[2 to 5] constra...

You just read in the *.sol file: model.read('model.sol')

The conclusion of an indicator constraint must be a linear constraint. But you can easily get around this by introducing another set of variables that are defined identically to `a`; let's call the...

This is a tough question. How to choose parameters to find good solutions quickly depends heavily on the type of problems you are solving. As far as I understand, a main issue with your models is t...

Sure. You can install a callback, which is then called at every single node in the search tree. And you can call the abort() method to stop the MIP search. But I don't understand how this would hel...

No, unfortunately not. Gurobi's data structures and algorithms are designed for a fixed number of variables during the solving process. Hence, if you add more variables, you need to start from scra...

I am not sure whether I understand your question correctly, but maybe the following explanation is of help. First, Gurobi does not support branch-and-price, so you cannot solve an MIP to optimality...

The issue usually arises from parallel solves. In our parallel MIP solver we first process a bunch of nodes without any user interaction in parallel. Then, at the synchronization point, we present ...

Sure, you can just use an all-zero objective function. Then the problem is a pure feasibility problem, which of course is fine for KKTs.

It should not matter as long as presolve is enabled. From a practical point of view, range constraints in Gurobi are a bit painful because adding a range constraint will actually add an equation an...