A Comparative Study on Predicting Methods for Cell Formation Problem

Abstract

The role of Feasibility Assessment (FA) is critical for Cellular Manufacturing (CM) that based on Group Technology (GT) concept, particularly during the design phase. The results may be used in decision making process to decide whether or not job shop may be converted into CM. Through FA some procedures are followed to identify the right predicted cells number that leads to create effective CM. In this article two methods of predicting a number of cells have been examined. The first method is supported by the number of machines in the incidence matrix and the pre-determinable maximum number of machines in each machine cell. The second method was based on Rogers and Tanimoto Similarity Coefficient and Eigenvalues of the Similarity Coefficient Matrix. Ten data sets selected from literatures are used to test these two methods. The results show that the first method is less accurate, depends on trial and error, and has multi values where most of those values are not integers. On the contrary, the second method is very accurate, has only one solution which is an integer value. Therefore it may be considered as a more efficient method. Furthermore, this study suggested an efficient and logical sequence of these two methods in order to identify the performance of the existing methods that are used in solving cell formation problems.A Comparative Study on Predicting Methodsfor Cell Formation Problem