many objective particle swarm optimization (mopso) algorithm

Multi-agent and deteriorating scheduling has gained an increasing concern from academic and industrial communities in recent years. Multi-agent scheduling problem is a subset of multi-objective scheduling problems in which each agent has a set of jobs and its aim is to optimize its own objective function. This study addresses a three-agent flow shop deteriorating scheduling problem. In the investigated problem, the actual processing time of jobs is a linear function of their normal processing time and starting time. To make the proposed problem more realistic, two practical assumptions such as sequence-dependent setup times and release date of jobs are considered.A mixed integer programming model has also been developed for the problem, which is solved using the Augmented ε-constraint exact method. Also due to the complexity of the model and its inability to solve large-scale problems, Multi Objective Particle Swarm Optimization (MOPSO) algorithm is developed. Since the parameters of meta-heuristic algorithms affect their overall performance and output, the Taguchi method has been used to adjust the parameters of the MOPSO algorithm. Finally, in order to evaluate the performance of the proposed algorithm, numerical sample problems of different structures are solved using this algorithm as well as the Non-dominated Sorting Genetic Algorithm (NSGA-II) and the exact augmented ε-constraint method. The computational results confirm the acceptable convergence and very good dispersion of the solutions of the MOPSO algorithm as well as the better performance of this algorithm compared to the augmented ε-constraint method and the NSGA-II algorithm.