The partial inverse minimum spanning tree problem when weight increase is forbidden

In a partial inverse optimization problem there is an underlying optimization problem with a partially given solution. The objective is to find a minimal perturbation of some of the problem’s parameter values, in such a way that the partial solution becomes a part of the optimal solution.

We consider the partial inverse minimum spanning tree problem in an undirected weighted graph under the constraint that edge weights can not be increased: by decreasing one or more edge weights as little as possible, a given forest must be presented in the new minimum spanning tree. Under a quite general criterion function, evaluating the proposed decreases of weight, we show that this problem can be solved in polynomial time.