Bi-Objective Optimal Scheduling With Raw Material's Shelf-Life Constraints in Unrelated Parallel Machines Production

• Published in: IEEE transactions on systems, man, and cybernetics. Systems Vol. 50; no. 11; pp. 4598 - 4610
• Main Authors: Wang, Ming-Zheng, Zhang, Ling-Ling, Choi, Tsan-Ming
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.11.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Bi-objective optimization; Completion time; discrete particle swarm algorithm; Evolutionary algorithms; Integer programming; Job shop scheduling; Job shops; Optimal scheduling; Parallel machines; Particle swarm optimization; Performance measurement; Raw materials; raw material’s shelf life; Scheduling; Shelf life; unrelated parallel machines scheduling
• ISSN: 2168-2216; 2168-2232
• DOI: 10.1109/TSMC.2018.2855700

This paper studies a challenging optimal scheduling problem considering the raw material with shelf-life constraints in unrelated parallel machines production. The aim is to optimize the assignment and sequencing of jobs to achieve tradeoffs between minimizing total completion time and the minimization of raw material costs. We formulate a bi-objective nonlinear 0-1 integer programming model for it. To solve the bi-objective problem which is NP-hard, we propose an evolutionary discrete particle swarm optimization algorithm (EDPSO) with a hybrid-greedy method (for generating the initial population), a new particle updating strategy, and an SPT-local search method (which has been proven for improving solutions' quality theoretically and practically), and the Pareto archive updating strategy for storing good solutions. Computational experiments verify the effectiveness of EDPSO and show that it can obtain better solutions compared to other competing algorithms based on four important performance metrics.