Particle swarm optimization algorithm embedded with maximum deviation theory for solving multi-objective flexible job shop scheduling problem

Abstract

Due to intense competition in the market place, effective scheduling has now become an important issue for the growth and survival of manufacturing firms. To sustain in the current competitive environment, it is essential for the manufacturing firms to improve the schedule based on simultaneous optimization of performance measures such as makespan, flow time, and tardiness. The current paper presents a novel particle swarm optimization (PSO) algorithm for solving multi-objective flexible job shop scheduling problem with the goal of finding approximations of the optimal Pareto front. The Pareto-optimal solutions obtained through multi-objective particle swarm optimization (MOPSO) have been ranked by the composite scores obtained through maximum deviation theory (MDT) to avoid subjectiveness and impreciseness in the decision-making. The results are compared with non-dominated sorting genetic algorithm-II (NSGA-II) and multi-objective evolutionary algorithm (MOEA) in terms of four performance metrics. Twenty-eight benchmark instances from literature are solved by the proposed algorithm. It is observed that MOPSO outperforms NSGA-II and MOEA in four performance metrics in most of the instances.