Mopgp'06: 7th Int. Conf. on Multi-objective Programming and Goal Programming

Branch-and-bound methods (also called implicit enumeration methods when the branching is done on binary variables) have proved to perform well on many combinatorial optimization problems, provided good bounding functions are known. Quite surprisingly, as emphasized in [3], they have not been studied widely in a multi-objective setting, i.e. when each solution s in the set S of feasible solutions is evaluated by p objective functions f(x) = (f1(x), . . . , fp(x)) to minimize (without loss of generality). Actually, the few existing papers on this topic concern mainly multi-objective integer linear programming [1, 6, 8, 9].