It can happen that presolving removes all rows and columns of a problem. This means all constraints are removed and variables are fixed, so no further LP or MIP solving techniques (like branch-and-bound, heuristics or cutting planes) are required. The optimal solution - if the model is feasible - could already be determined using the various reductions, implications, and propagations that are performed during the presolving phase of Gurobi.
Articles in this section
- Why does Gurobi perform differently on different machines?
- How to use MIP starts
- Is there a way to predict how long it will take Gurobi to solve a MIP?
- How to query best bound or best objective value in multi-objective optimization?
- How does Gurobi compute the IIS for infeasible models?
- Can I access the MIP cuts generated by Gurobi?
- What does "Presolve: All rows and columns removed" mean?
- How can I return from multi-objective optimization back to single-objective mode?
- Why can't I get the Pi values for a MIP problem?
- Is Gurobi Optimizer deterministic?