A two-stage heuristic for the sequence-dependent job sequencing and tool switching problem

• Published in: Computers & industrial engineering Vol. 163; p. 107813
• Main Authors: Rifai, Achmad Pratama, Windras Mara, Setyo Tri, Norcahyo, Rachmadi
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2022
• Subjects: Adaptive large neighborhood search; Flexible manufacturing machine; Job sequencing and tool switching; Sequence-dependent setup time; Simulated annealing; Job sequencing and tool switching; Sequence-dependent setup time; Adaptive large neighborhood search; Simulated annealing; Flexible manufacturing machine
• ISSN: 0360-8352
• DOI: 10.1016/j.cie.2021.107813

•We study the job sequencing and tool switching (SSP) with sequence-dependent setup.•This study develops a two-stage heuristic for solving the problem.•A combined KTNS and SA is proposed for finding the optimum tool switching pairs.•An Adaptive Large Neighborhood Search (ALNS) is developed for the job-sequencing.•A set of heuristics are developed for the destroy and repair operators of ALNS. The job sequencing and tool switching problem is a combinatorial optimization problem that commonly happens in a flexible manufacturing system environment. In this type of environment, a set of flexible manufacturing machines can be configured with various tools to process different jobs and the requirement to switch tools may correspond to a reduction of productivity. Consequently, research on the job sequencing and tool switching problem has been concentrated on minimizing the number of tools switches. This paper discusses the sequence-dependent job sequencing and tool switching problem. The sequence-dependent job sequencing and tool switching problem extends the standard model by considering non-uniform setup times since industrial applications indicate that a tool setup time might be influenced by the previously installed tool at the same magazine slot. A two-stage heuristic procedure is developed here. In the first stage, an adaptive large neighborhood search is deployed for finding the near-optimal job sequence. Then, in the second stage, a combination of the Keep Tool Needed Soonest policy and simulated annealing is proposed for the tooling sub-problem. Comprehensive computational experiments are carried out to demonstrate the efficacy and robustness of the proposed method for solving the sequence-dependent job sequencing and tool switching problem.