Optimization of job shop scheduling problems using particle swarm and ant colony algorithms

摘要

This paper proposes a prominent approach to solve job shop scheduling problem based on Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO). The steps to generate the solution are grouped as planning, scheduling and optimization. Initially, fuzzy logic is applied for planning and then the scheduling stage is optimized using PSO and ACO. The processing order of jobs for each machine is scheduled with an objective to find a feasible plan that minimizes the makespan, completion time and waiting time. The well known Fisher and Thompson 10×10 instance (FT10) and Adams, Balas, and Zawack 10×10 instance (ABZ10) problems are selected as the experimental benchmark problems. The results of the applied optimization techniques are compared with the computed parameters like makespan, waiting time, completion time and elapse time. The performance evaluation of optimization techniques are analysed for both benchmark problems and the PSO technique is found superior.