Development of Flowshop Scheduling Simulator

Abstract

Scheduling or production scheduling is an important activity in production process. Production scheduling is an activity that determine the efficiency of production process. With a good production scheduling, the production process will be running smooth and benefit in cost. With the growing scale in number of jobs and machines on production floor, manual scheduling is irrelevant these days. So a software is required to do a scheduling activity and simulating the result. Scheduling is an activity of allocating resources or machines to run some tasks in specific time range[1]. The result of scheduling is a sequence of jobs that used for reference in determining on how to run a production process. There are several criteria in scheduling process, one of them is minimizing make span [2] or the total amount of time to process all the jobs. The production process has two major flow type, they are flowshop and jobshop.
Flowshop production flow is a production flow type which a sequence of jobs processed on an array of machine. In flowshop production flow every job on the sequence will be processed on every machine on the machine’s array, while on jobshop, every job is processed based on the requirement and characteristic of the job. Scheduling in flowshop production flow can be done using many methods. The flow shop can be done using Nawaz Enscore Ham (NEH)[3] method, Palmer[4] method, and Gupta[5] method. Those method prioritizing a job with biggest total of processing time to be processed first.
The previous paper [6] compare the performance of these three methods. In this paper, we develop the flowshop scheduling simulator to make it easy to visualize the running process of these algorithm. The software simulates the result of flowshop scheduling process in a form of gantt chart animation. Gantt chart is used for simulating the flow of production scheduling process. By using simulation, user can see the whole scheduling planning, estimate the completion time of each job, see the total processing time or makespan, and know the tardiness of each job. Similar with the previous paper [6], NEH algorithm constantly works best compare with Palmer and Gupta algorithm in various types of problems.