M. Pachayappan

Dr. R. Panneerselvam

Abstract

This paper compares the solution accuracy of three different clustering algorithms, viz. Rank Order Clustering algorithm, Agglomerative clustering algorithm and Principle Component Analysis applied to machine-component cell formation problem of cellular manufacturing system. In this paper, two different measures, viz, grouping efficiency and grouping efficacy are used to compare the solutions of the algorithms, At the end, a complete factorial experiment had been designed to compare the solution accuracy of the three algorithms using Problem Type as Factor A and Algorithm as Factor B. Based on the analysis, it found that there is no significant difference between the algorithms in terms of either grouping efficiency or grouping efficacy, But, based on the mean values of each of the grouping measures, it is found that the agglomerative clustering algorithm performs better grouping efficiency and principle component analysis performs better for grouping efficacy.

Keywords- Cellular manufacturing system, similarity coefficient, grouping efficiency, grouping efficacy and cell formation problem.