Multiple Machines
Alf Kimms · 1997
This chapter discusses several heuristic solution procedures for the PLSP-MM. We briefly report about some unsuccessful attempts in Section 6.1. A backward oriented construction scheme is underlying all methods described in here. Hence, in Section 6.2 we provide those construction principles that are common to all heuristics. Afterwards, we refine the presented concept to give a randomized regret based sampling procedure in Section 6.3, a cellular automaton in Section 6.4, a genetic algorithm in Section 6.5, a disjunctive arc based tabu search method in Section 6.6, and a so—called demand shuffle heuristic in Section 6.7. Section 6.8 provides a summary of the computational studies and compares the procedures. In Section 6.9 we run some tests with large instances to reveal the applicability to real—world instances. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.