On the robustness of optimal solutions for combinatorial optimization problems

We consider the so-called generic combinatorial optimization problem, where the set of feasible solutions is some family of subsets of a finite ground set with specified positive initial weights of elements, and the objective function represents the total weight of elements of a feasible solution. We assume that the weights of all elements may be perturbed simultaneously and independently up to a given percentage of their initial values. A feasible solution which minimizes the worst-case relative regret, is called a robust solution. The maximum percentage level of perturbations, for which an initially optimal solution remains robust, is called the robustness radius of this solution. In this paper we study the robustness aspect of initially optimal solutions and provide lower bounds for their robustness radii.