Abstract: Reasonable part grouping is one of the key factors to increase productivity. A case study on part grouping is conducted for a panel furniture factory at Dezhou city. By using process analysis, a similarity matrix for the parts is built. Then, an equivalent matrix is constructed by transfer closure method such that part grouping can be done. With Matlab, an algorithm is presented to solve the complex part grouping problem. By this method, the number of part groups obtained is dependent on the threshold value. As the threshold value increases, the number of part groups increases and the average similarity index increases accordingly. Thus, if a furniture company is small, the threshold value should be set to be smaller than that for a larger company so as to obtain a small number of part groups, while for a large company the threshold value should be set to be relatively larger.

Key words: process analysis, group technology, fuzzy clustering