New benchmark algorithm for hybrid flowshop scheduling with identical machines

Abstract

With many industrial applications in the production sector, the hybrid flowshop scheduling problem (HFSP) has received wide recognition in the scheduling literature. Given the NP-hard nature of the HFSP, which is characterized by highly intractable solution spaces, effective solution approaches are of particular interest to facilitate its real-world use cases. This study proposes a new benchmark metaheuristic, the Chaos-enhanced Simulated Annealing (CSA) algorithm to minimize makespan in the HFSP with identical machines. A recently published testbed is used to evaluate the performance of CSA against that of the upper bounds/best-known solutions in the literature. In addition to improving the upper bounds, the computational results revealed that CSA performed very well in terms of computational efficiency and stability. The proposed CSA can serve as a strong benchmark algorithm for developing more algorithms to effectively and efficiently solve HFSPs and its extensions.