Simultaneous Hierarchical Clustering for Cell Formation Problems with Production Information

Abstract

The purpose of this paper is to advance the similarity coefficient method to solve cell formation (CF) problems in two aspects. Firstly, while numerous similarity coefficients have been proposed to incorporate different production factors in literature, a weighted sum formulation is applied to aggregate them into a nonbinary matrix to indicate the dependency strength among machines and parts. This practice allows flexible incorporation of multiple production factors in the resolution of CF problems. Secondly, a twomode similarity coefficient is applied to simultaneously form machine groups and part families based on the classical framework of hierarchical clustering. This practice not only eliminates the sequential process of grouping machines (or parts) first and then assigning parts (or machines), but also improves the quality of solutions. The proposed clustering method has been tested through twelve literature examples. The results demonstrate that the proposed method can at least yield solutions comparable to the solutions obtained by metaheuristics. It can yield better results in some instances, as well.

Grouping technology and a hybrid genetic algorithm‐desirability function approach for optimum design of cellular manufacturing systems

“…Zhu and Li [32] used one classical method: hierarchical cluster analysis to aggregate both machines and parts simultaneously. As one feature of the suggested method, block diagonal forms that show some intermediate sorting of machines and parts have been created without defining the structural criteria (e.g., the number of cells).…”

Retract-Complete Similarity Measure Mathematical Model For Cell Layout Design

“…The quality of the algorithm is measured by using Grouping efficiency and grouping efficacies which are most widely used measures the superiority of cellular manufacturing systems. The proposal in the literature of Yingyu Zhu and Simon [8] used to advance the similarity coefficient method to solve cell formation (CF) problems in two aspects. Firstly, a weighted sum formulation is applied to aggregate them into a non binary matrix to indicate the dependency strength among machines and parts and secondly, a two mode similarity coefficient is applied to simultaneously form machine groups and part families based on the classical framework of hierarchical clustering.…”