A neuro-genetic approach to design and planning of a manufacturing cell

• Published in: Journal of intelligent manufacturing Vol. 16; no. 4-5; pp. 453 - 462
• Main Authors: Cakar, Tarik, Yildirim, Mehmet Bayram, Barut, Mehmet
• Format: Journal Article
• Language: English
• Published: London Springer Nature B.V 01.10.2005
• Subjects: Design; Employment; Genetic algorithms; Heuristic; Job shops; Machine learning; Manufacturing; Manufacturing cells; Mathematical programming; Neural networks; Product design; Resource allocation; Resource availability; Scheduling; Simulation; Studies; Work stations
• ISSN: 0956-5515; 1572-8145
• DOI: 10.1007/s10845-005-1657-2

In this paper, we propose a neuro-genetic decision support system coupled with simulation to design a job shop manufacturing system by achieving predetermined values of targeted performance measures such as flow time, number of tardy jobs, total tardiness and machine utilization at each work center. When a manufacturing system is designed, the management has to make decisions on the availability of resources or capacity, in our setting, the number of identical machines in each work station and the dispatching rule to be utilized in the shop floor to achieve performance values desired. Four different priority rules are used as Earliest due date (EDD), Shortest Processing Time (SPT), Critical ratio (CR) and First Come First Serve (FCFS). In reaching the final decision, design alternatives obtained from the proposed system are evaluated in terms of performance measures. An illustrative example is provided to explain the procedure. [PUBLICATION ABSTRACT]