Development of pareto-based differential evolution for multiobjective flexible job shop scheduling problems

• Main Authors: Warisa Wisittipanich, Thiarat Sootsuk
• 格式: Conference Proceeding
• 出版: 2020
• 主題: Engineering
• 在線閱讀: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85084085931&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/70575
• 機構: Chiang Mai University

© 2016 Proceedings of the 2016 Industrial and Systems Engineering Research Conference, ISERC 2016. All rights reserved. This paper proposes an algorithm, called MODE-3FJSP, based on multi-objective differential evolution to find a Pareto frontier for multi-objective flexible job shop scheduling problems. The objective is to find the schedules that simultaneously minimize makespan, total workload of all machines and maximum workload on a critical machine. The MODE-3FJSP framework adopts the idea of the Elite group to store non-dominated solutions and utilizes those solutions as the guidance of the vectors to search for a high-quality Pareto front. A novel mutation strategy is proposed in MODE-3FJSP to use the best solution of each objective function as the vector guidance in order to improve the search efficiency. This study presents efficient solution mapping procedures to hierarchically sequence job operations and assign operations to machines to generate an active schedule. The performances of the proposed algorithm are evaluated on a set of benchmark problems and the numerical experiments show that the MODE-3FJSP is a highly competitive approach which is capable of providing a set of diverse and high-quality non-dominated solutions compared to those obtained from existing algorithms.