Many-to-many assignment problems: Lagrangian bounds and heuristic

Abstract

Modified Lagrangian bounds and a greedy heuristic are proposed for many-to-many assignment problems taking into account capacity limits for tasks and agents. A feasible solution recovered by the heuristic is used to speed up the subgradient technique to solve the modified Lagrangian dual. A numerical study is presented to compare the quality of the bounds and to demonstrate the efficiency of the overall approach. © 2012, IGI Global.