Effective Population-Based Meta-heuristics with NEH and GRASP Heuristics Minimizing Total Weighted flow Time in No-Wait Flow Shop Scheduling Problem Under Sequence-Dependent Setup Time Constraint
Said Aqil
DOI: https://doi.org/10.1007/s13369-023-08642-7
IF: 2.807
2024-01-19
Arabian Journal for Science and Engineering
Abstract:The optimization of scheduling in industrial manufacturing is a crucial concern in operations research. In this paper, a range of of methodological approaches for solving the permutation flow shop scheduling problem with a sequence-dependent setup. The goal is to minimize the total weighted flow time while ensuring no waiting jobs in the production process. The proposed solution methodology encompasses eight population-based meta-heuristics and the mixed integer linear programming (MILP) method. The employed population-based meta-heuristics include migratory bird optimization, the artificial bee colony algorithm, iterated greedy and iterative local search algorithms. Various approaches, incorporating local search procedures, are developed to explore neighboring solutions. Furthermore, two constructive algorithms namely the Nawaz-Enscore and Ham (NEH) algorithm and the greedy random adaptive search procedure (GRASP), are proposed for generating best sequences in the initial phase. In total, eight population-based meta-heuristics : PIG , PIG , PILS , PILS , MBO , MBO , ABC and ABC are presented to investigate the manufacturing optimization problem. To evaluate the efficiency of the approach proposed for the algorithms, a comprehensive set of tests is conducted on instances of varying sizes . The simulation shows that the population-based iterated greedy algorithm enhanced by the NEH procedure (PIG ) performs best compared to other algorithms. Indeed, the PIG meta-heuristic demonstrates a success rate of and achieves a minimum percentage relative deviation value of .
multidisciplinary sciences