Growing hierarchical self-organizing map computation approach for clustering in cellular manufacturing

This article focuses on approach that provides visualization of machine–part clustering in cellular manufacturing system based on sequence of operation. We propose a novel cell formation approach, namely the growing hierarchical self-organizing map (GHSOM), for dealing with 14 benchmark problems from literature. The performance of the proposed algorithm is tested with the problem data sets and the results are compared using the group technology efficiency (GTE) and computational time with the existing traditional clustering algorithms. It is found that the proposed algorithm resulted in an increase in GTE in most of the problem data sets, and the outputs of cell formation are either superior or same as existing methods. The outputs of the experiments conducted in this research lead us to the conclusion that the GHSOM is a promising alternative cell formation algorithm owing to its adaptive architecture and the ability to expose the hierarchical structure of data.