A comparison of GA and PSO algorithm for multi-objective job shop scheduling problem

This paper compares two metaheuristic solution methods for multi-objective job shop scheduling problems. The two methods compared are 2S-GA (a two-stage multi-objective genetic algorithm) and 2S-PSO (a two-stage particle swarm optimization algorithm). Both algorithms are implemented with three criteria: minimize makespan, minimize total weighted earliness, and minimize total weighted tardiness. Both algorithms applied the random keys representation and the schedules are constructed using a permutation with m-repetitions of job numbers. Performance of the algorithms is tested on benchmark instances for both the single objective and multi-objective cases. In general, the solutions obtained via 2S-PSO dominate those from 2S-GA in that it provided equal or better solution quality in much shorter computational time.