Improved particle swarm optimization algorithm based novel encoding and decoding schemes for flexible job shop scheduling problem

The flexible job shop scheduling problem (FJSP) is a typical scheduling problem in practical production and has been proven to be a NP-hard problem. The study of FJSP is important to remarkably direct actual manufacturing processes. The paper proposes an improved particle swarm optimization (PSO) algorithm for solving FJSP and obtains beneficial solutions by improvement on encoding/decoding scheme, communication mechanism between particles, and alternate rules of candidate machines of operations. The innovation of encoding/decoding scheme proposes a novel designed chain encoding scheme and a corresponding effective decoding scheme. The chain-based encoding scheme can reasonably convert FJSP to an appropriate operation linked list and the novel designed decoding scheme owns the capacity of further explorering the solution space. The improvement of traditional PSO focuses on the innovation of information communication between particles, besides the modification of algorithm architecture. The amelioration of rules on operated machine selection is carried out based on the critical path of operations research (OR). It promotes algorithm efficiency by only alternating the candidate machines of operations on the critical path. In addition, much parameters tuning work is involved in a series of experiments. The study proposes some tuning schemes of parameters with exact mathematical methods, and these schemes can effectively help find more appropriate parameters. The final experiment results prove that the improved PSO exhibits remarkable ability to solve FJSP. (C) 2020 Elsevier Ltd. All rights reserved.