A comparison of the performance of artificial intelligence techniques for optimizing the number of kanbans
- PDF / 214,441 Bytes
- 8 Pages / 595 x 794 pts Page_size
- 91 Downloads / 173 Views
#2002 Operational Research Society Ltd. All rights reserved. 0160-5682/02 $15.00 www.palgrave-journals.com/jors
A comparison of the performance of artificial intelligence techniques for optimizing the number of kanbans C Alabas, F Altiparmak and B Dengiz* Gazi University, Maltepe, Ankara, Turkey This paper discusses the use of modern heuristic techniques coupled with a simulation model of a Just in Time system to find the optimum number of kanbans while minimizing cost. Three simulation search heuristic procedures based on Genetic Algorithms, Simulated Annealing, and Tabu Search are developed and compared both with respect to the best results achieved by each algorithm in a limited time span and their speed of convergence to the results. In addition, a Neural Network metamodel is developed and compared with the heuristic procedures according to the best results. The results indicate that Tabu Search performs better than the other heuristics and Neural Network metamodel in terms of computational effort. Journal of the Operational Research Society (2002) 53, 907–914. doi:10.1057/palgrave.jors.2601395 Keywords: kanban; Just in Time system; kanban-controlled system; simulation optimization; genetic algorithms; simulated annealing; tabu search; neural network
Introduction Due to manufacturing company’s interest, researchers started investigating the Just in Time (JIT) philosophy and much work has been done to find the number of kanbans required in a JIT system. In general, a JIT system, if implemented properly, will result in increased productivity, reduced work-in-process (WIP), and higher product quality. Kanban developed at Toyota Motor Company means signboard in Japanese. Usually, a kanban is a printed card that carries the name of the part and other relevant information. Every part that moves through the production sequence has an accompanying kanban. Therefore, a kanban-controlled system as a means of production activity control to achieve the goals of JIT is used to direct materials to workstations and pass information as to what and how much to produce.1 The number of kanbans allocated to each product type that affects the desired performance level is an important decision problem. Wang and Wang,1 Kimura and Terada,2 Rees et al,3 Bitran and Chang,4 Deleersnyder et al,5 Monden,6 Askin et al,7 and Fukukawa and Sung-Chan8 studied kanban-controlled systems and presented many mathematical approaches to formulate setting the number of kanbans considering different system configurations and objective functions. When the system under investigation is complex, as is often the case in manufacturing environments, analytical solutions of these systems become intractable. Because of *Correspondence: B Dengiz, Department of Industrial Engineering, Gazi University, Celal Bayar Bulvari, Maltepe 06570, Ankara, Turkey. E-mail: [email protected]
the complex stochastic characteristic of the systems, simulation is used as a tool to analyse them. However, the major drawback of simulation for practical applications is that it is c
Data Loading...
