Problem Geometry and Problem Robustness

Abstract

We could say an optimization problem is “robust” if its solution is not very sensible to changes in the data. This is an important property when models are used to support decision making in various areas. In this talk we explore the potential explanatory power of some geometric measures of the feasible region on the sensitivity and robustness of the problem. We do this in connection with the computation of robust solutions, showing some theoretical estimates as well as some computational results in simulated as well as real Supply Chain Management problems.