TY - JOUR

T1 - A genetic-based algorithm with the optimal partition approach for the cell formation in bi-directional linear flow layout

AU - Chiang, Chih Ping

AU - Lee, Shine-Der

PY - 2004/6/1

Y1 - 2004/6/1

N2 - This paper addresses the joint problem of the cell formation and the intercell layout, in which machine cells are located along a linear flow layout. The objective is to minimize the actual intercell flow cost, instead of the typical measure that optimizes the number of intercell movements. A genetic-based algorithm with optimal partition approach is developed for solving the joint problem. In the proposed approach, the genetic operators are used to generate a pool of solutions, where the solution represents the sequence of machines in the linear layout and each sequence of machines is an individual in the population. The evaluation function, machine cells and its intercell flow cost of each solution are determined by a dynamic programming algorithm, where a sequence of machines is partitioned into several segments or cells, subject to a cell size constraint. In this sense, the genetic algorithm is enhanced with the optimization framework in the proposed approach. The computational efficiency has also been improved since the number of machine cells does not need to be specified in advance in the cell formation approaches. Numerical studies, including 14 data sets adapted from the literature, are performed to demonstrate the viability of the approach.

AB - This paper addresses the joint problem of the cell formation and the intercell layout, in which machine cells are located along a linear flow layout. The objective is to minimize the actual intercell flow cost, instead of the typical measure that optimizes the number of intercell movements. A genetic-based algorithm with optimal partition approach is developed for solving the joint problem. In the proposed approach, the genetic operators are used to generate a pool of solutions, where the solution represents the sequence of machines in the linear layout and each sequence of machines is an individual in the population. The evaluation function, machine cells and its intercell flow cost of each solution are determined by a dynamic programming algorithm, where a sequence of machines is partitioned into several segments or cells, subject to a cell size constraint. In this sense, the genetic algorithm is enhanced with the optimization framework in the proposed approach. The computational efficiency has also been improved since the number of machine cells does not need to be specified in advance in the cell formation approaches. Numerical studies, including 14 data sets adapted from the literature, are performed to demonstrate the viability of the approach.

UR - http://www.scopus.com/inward/record.url?scp=2542444688&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2542444688&partnerID=8YFLogxK

U2 - 10.1080/09511920310001640512

DO - 10.1080/09511920310001640512

M3 - Article

AN - SCOPUS:2542444688

VL - 17

SP - 364

EP - 375

JO - International Journal of Computer Integrated Manufacturing

JF - International Journal of Computer Integrated Manufacturing

SN - 0951-192X

IS - 4

ER -