Multistage robust discrete optimization via quantified integer programming

Decision making needs to take an uncertain environment into account. Over the last decades, robust optimization has emerged as a preeminent method produce solutions that are immunized against uncertainty. The main focus in discrete been on analysis and solution of one- or two-stage problems, where decision maker limited options reacting additional knowledge gained after parts have fixed. Due its computational difficulty, multistage problems beyond two stages received less attention. In this paper we argue can be seen through lens quantified integer programs, powerful tools reduce search tree size developed. By formulating both programming formulations, it is possible compare performance state-of-the-art solvers from worlds. Using selection, assignment, lot-sizing knapsack testbed, show with up nine solved optimality reasonable time.