Solving multi-objective hybrid flow shop problems with biogeography-based optimization algorithm

Almost all researches in the domain of flowshop scheduling are for single machine, two machines or several identical parallel machines while the number of machines is more than two in practice for flowshop scheduling. This paper studies the problem of multi-objective hybrid flow shop problems with biogeography-based optimization algorithm that in domain of scheduling. This problem is defined as a complete set of N jobs, N=1,2,…,n that are processed on M machine, M=1,2…,m. all N jobs pass with same order for machines and the processing time for each job can be different at each stage. in the flow shop problem the aim is finding jobs process order on the machines so that, in each stage ther are some identical machines and the desired restriction to be satisfied.The job can be processed on at just one machine; mean while one machine can process at just one job, at a time.In this paper The purpose is to schedule job sets so as to minimizing two objective simultaneously: the maximum completion(makespan) time and the maximum tardiness time of jobs. we propose a new idea of the use of Multi-Objective biogeography-based optimization (MOBBO) algorithm to solve certain multi-objective problem. The results of MOBBO is compared with Multi-Objective simulated annealing algorithm (MOSA) and non-dominated sorting genetic algorithm (NSGA II) that indicate it is the best. Numerical experiments, analyzed by advanced multi-objective performance measure and statistical tools(ANOVA and LSD), show the proposed algorithm outperforms the other two algorithms. the algorithm MOBBO, by increasing the size of the solutions are improved and the improvement continues until the large sizes to achieve absolute superiority to the other three algorithms.

Keywords:

Hybrid flow shop scheduling , Multi-objective optimization , biogeography-based optimization