Exact makespan minimization of unrelated parallel machines

We study methods for the exact solution of unrelated parallel machine problem with makespan minimization, generally denoted as R||C max . Our original application arises from automotive assembly process where tasks needs to be distributed among several robots. This involves solutions instances, which proved hard a MILP solver since objective induces weak LP relaxation bounds. To improve these bounds and enable larger we propose branch–and–bound method based on Lagrangian assignment constraints. For this derive criterion variable fixing prove zero duality gap property case two machines. computational studies indicate that proposed algorithm is competitive state-of-the-art different types instances. Moreover, impact each feature analysed.