MODEL DEVELOPMENT OF SHIFTING BOTTLENECK HEURISTIC TO SOLVE JOB SHOP SCHEDULING PROBLEM WITH PARALLEL-MACHINES

Abstract

Job Shop Scheduling is a complex problem that has been a challenge for years. Despite its complexity, job shop flow pattern is often used in many manufacturing process. Therefore, job shop scheduling is very important and needed by the industry. In addition, the existence of parallel-machine on the factory floor increases the complexity of scheduling. One of the models that are used to solve job shop scheduling problem with parallel-machines is non-delay algorithm. A research was conducted to prove the effectiveness of this model in solving a job shop scheduling problem using the data gathered from SOHO Group Pharma. The Shifting Bottleneck Heuristic will be developed for parallel machines and tested with the combination of priority rules. The result shows that the model successfully produces a feasible schedule for 59 jobs and 28 machines with the makespan of 798. Furthermore, this paper is aimed to continue the research using shifting bottleneck heuristic. The result successfully produces a feasible schedule for the same amount of jobs with the makespan of 734, which is 8% smaller makespan than the one produced using non-delay algorithm.