Particle filter and Levy flight-based decomposed multi-objective evolution hybridized particle swarm for flexible job shop greening scheduling with crane transportation

Binghai Zhou,Xiumei Liao
DOI: https://doi.org/10.1016/j.asoc.2020.106217
IF: 8.7
2020-01-01
Applied Soft Computing
Abstract:Since greening scheduling is arousing increasing attention from many manufacturing enterprises, this paper focuses on a flexible job shop greening scheduling problem with crane transportation (FJSGSP-CT). Distinguished from the traditional scheduling model which merely concentrates on machining processes, FJSGSP-CT takes the comprehensive effect of machining and crane transportation processes into consideration. Due to the NP-hard nature of the problem, an efficient hybrid algorithm, particle filter and Levy flight-based decomposed multi-objective evolution hybridized with particle swarm (PLMEAPS), is developed to find feasible solutions. The proposed PLMEAPS benefits from the synergy of decomposed multi-objective evolutionary algorithm (MOEA/D) and particle swarm optimization (PSO). Particle filter and Levy flights are then creatively fused into the framework of PLMEAPS to enhance the computational performance of the algorithm. The introduction of particle filter enriches the diversity of the population and makes it possible to predict the near optimal solutions at each iteration, and the combination of Levy flights has beneficial effect on escaping from local optimum and accelerating convergence speed. The performance of the proposed PLMEAPS is evaluated by comparing with two other high-performing intelligent optimization algorithms, the multi-objective genetic local search (MOGLS) and the multi-objective grey wolf optimizer (MOGWO). The computational results reveal that the proposed PLMEAPS outperforms the other two algorithms both in solutions' quality and convergence rate when solving FJSGSP-CT. (C) 2020 Elsevier B.V. All rights reserved.
What problem does this paper attempt to address?