Generation of Look-Up Tables for Dynamic Job Shop Scheduling Decision Support Tool

2nd International Manufacturing Engineering Conference; 3rd Asia-Pacific Conference on Manufacturing Systems

Oktaviandri, Muchamad^1,3 ; Hassan, Adnan^1 ; Shaharoun, Awaluddin Mohd^2

Faculty of Manufacturing Engineering, Universiti Malaysia Pahang, 26600, Malaysia^1

Faculty of Mechanical Engineering, Universiti Teknologi Malaysia, 81300, Malaysia^2

Fakultas Teknologi Industri, Universitas Bung Hatta, Padang

25143, Indonesia^3

关键词: Artificial intelligent;    Decision support tools;    Discrete-event simulation model;    Dynamic job shop scheduling;    Job shop scheduling problems;    Performance measure;    Scheduling techniques;    Stochastic processing time;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/114/1/012067/pdf
DOI  :  10.1088/1757-899X/114/1/012067

来源: IOP

【 摘 要 】

Majority of existing scheduling techniques are based on static demand and deterministic processing time, while most job shop scheduling problem are concerned with dynamic demand and stochastic processing time. As a consequence, the solutions obtained from the traditional scheduling technique are ineffective wherever changes occur to the system. Therefore, this research intends to develop a decision support tool (DST) based on promising artificial intelligent that is able to accommodate the dynamics that regularly occur in job shop scheduling problem. The DST was designed through three phases, i.e. (i) the look-up table generation, (ii) inverse model development and (iii) integration of DST components. This paper reports the generation of look-up tables for various scenarios as a part in development of the DST. A discrete event simulation model was used to compare the performance among SPT, EDD, FCFS, S/OPN and Slack rules; the best performances measures (mean flow time, mean tardiness and mean lateness) and the job order requirement (inter-arrival time, due dates tightness and setup time ratio) which were compiled into look-up tables. The well-known 6/6/J/Cmax Problem from Muth and Thompson (1963) was used as a case study. In the future, the performance measure of various scheduling scenarios and the job order requirement will be mapped using ANN inverse model.