A discrete whale swarm algorithm for hybrid flow-shop scheduling problem with limited buffers. (April 2021)

• Record Type: Journal Article
• Authors: Zhang, Chunjiang; Tan, Jiawei; Peng, Kunkun; Gao, Liang; Shen, Weiming; Lian, Kunlei
• Is Part Of: Robotics and computer-integrated manufacturing. Volume 68(2021)
• Publication Date: 2021-04
• Subjects: Whale swarm algorithm -- Hybrid flow-shop scheduling -- Limited buffers -- Assembly line; Robots, Industrial -- Periodicals; Computer integrated manufacturing systems -- Periodicals; Robotics -- Periodicals; Robots industriels -- Périodiques; Productique -- Périodiques; Robotique -- Périodiques; 670.285
• DOI: 10.1016/j.rcim.2020.102081
• Languages: English
• ISSNs: 0736-5845

Abstract

Highlights: The HFSP with limited buffer and two process routes is studied for the first time. A new discrete WSA (DWSA) is proposed to solve the LBHFSP efficiently. A real case from a diesel engine plant is studied to prove DWSA is helpful. Abstract: This paper studies a hybrid flow-shop scheduling problem with limited buffers and two process routes that comes from an engine hot-test production line in a diesel engine assembly plant. It extends the classical hybrid flow-shop scheduling problem by considering practical constraints on buffer area resources and alternative process routes. Because of its NP-hardness and large scale, traditional optimization methods and heuristic rules cannot obtain satisfactory solutions. A discrete whale swarm algorithm (DWSA) is proposed to identify near-optimal solutions efficiently. The proposed algorithm adopts an encoding method based on the problem characteristic and a greedy delayed decoding strategy to avoid infeasible solutions. A hybrid initialization is used to ensure the quality of the initial population and diversity. A new way of computing distances and a movement rule between individuals are designed. Five mutation operators and a deduplication strategy are proposed to improve the population diversity. The effectiveness of the proposed DWSA is validated on three groups of instances and a real-world industrial case.