An immunity-based hybrid genetic algorithms for permutation flowshop scheduling problems

Abstract

This paper investigates novel GA-based hybrid artificial immune system (AIS) for the permutation flowshop scheduling problems (PFSP) that minimizes the makespan. The proposed approaches aim to show that the efficiency of GAs in solving flowshop problems can be improved significantly by tailoring the various AIS operators to suit the problem structure. The proposed hybridization scheme is applied in two ways: (1) the first hybrid of GA and AIS introduces vaccination (Jiao and Wang, IEEE Trans Syst Man Sybernetics Part A Syst Hum 30(5):552–561, 2000) into the field of GAs based on the theory of immunity in biology, (2) the second takes its inspiration on the immune network theory (Perelson, Immunol Rev 110(1):5–36, 1989), and applied it to the field of GAs. The proposed hybrid metaheuristics produce high quality solutions as proved by the tests performed over Taillard’s (Eur J Oper Res 64(2):278–285, 1993) well-known flowshop scheduling benchmarks and corroborated by the comparisons we did with the most frequently referred in the related literature and recently developed hybrid GAs, including genetic algorithms, particle swarm optimization, and other advanced and recent techniques. Furthermore, the effects of some parameters are discussed.