A preliminary study of a fuzzy clustering and assignment problem-based cell formation algorithm

The ideal situation of cellular manufacturing is achieved when each part-cluster together with the machine-cluster for the manufacturing cell is independent from the rest of the manufacturing cell; that is exceptional elements are non existent. However, such a situation is rarely achieved. In this paper, a cell formation algorithm that identifies part-clusters and machine-clusters separately is introduced. The algorithm is based on fuzzy clustering and the assignment technique. Computational results show this new algorithm can give superior performance, especially for higher density machine-part incidence matrices.