An NEH-based heuristic algorithm for distributed permutation flowshop scheduling problems

Abstract

Distributed Permutation Flowshop Scheduling Problem (DPFSP) is a newly proposed scheduling problem with a strong engineering background. Whereas there is a body of work on scheduling problems in the past decades, the literature on DPFSPs is scant and in its infancy. Motivated by the good performances of some heuristics in a very recent work, we propose a constructive heuristic algorithm enhanced through a novel dispatching rule to deal with the DPFSP. Given multiple factories in a DPFSP, our dispatching rule will insert a group of jobs to the factories at one time instead of inserting one job at one time like the original rules. The time complexity of the proposed heuristic algorithm is the same as that of the NEH with the original rule. To validate the proposed heuristic, intensive benchmark experiments are carried out on the large problem instances, and the results show that the proposed algorithm outperforms the existing heuristics in terms of tradeoff between solution quality and running time.   Key words: Distributed scheduling problem, flowshop, heuristic, branch and bound.