Solutions methods for m-machine blocking flow shop with setup times and preventive maintenance costs to minimise hierarchical objective-function

• Published in: International journal of production research Vol. 61; no. 19; pp. 6308 - 6335
• Main Authors: Miyata, Hugo Hissashi, Nagano, Marcelo Seido, Gupta, Jatinder N. D.
• Format: Journal Article
• Language: English
• Published: London Taylor & Francis 02.10.2023; Taylor & Francis LLC
• Subjects: Blocking; blocking flow shop; Completion time; Computational efficiency; Computing time; Greedy algorithms; Heuristic methods; Integer programming; Linear programming; Maintenance costs; maintenance operations; Mixed integer; Preventive maintenance; Scheduling; setup times; Swarm intelligence; total completion time; Tradeoffs
• ISSN: 0020-7543; 1366-588X
• DOI: 10.1080/00207543.2022.2127959

In this article, maintenance operations were incorporated to the sequence-dependent setup blocking flow shop to minimise total completion time subject to total maintenance costs. A mixed integer linear programming and procedures to incorporate maintenance to job sequence were developed. A constructive heuristic and three metaheuristics, greedy randomised adaptative search procedure (GRASP), discrete artificial bee colony (DABC), variable block insertion heuristic (VBIH) and iterated greedy algorithm (IG), designed to the blocking flow shop with total completion time minimisation were adapted to minimise total maintenance costs and the hierarchical function, respectively.All the methods were applied to solve small and medium and large size instance sets, with respective 1920 and 2200 problems. Experimental results shows that for small size instances set, DABC with (DABC(20)) obtained the best trade-off between effectiveness and efficiency. For medium and large size instances set, DABC(20) VBIH with (VBIH(20)) generated the best trade-off between quality of solution and computational time. Considering both instances set together, both DABC(20) and VBIH(20) obtained the best performance between quality of solution and computational time.