Conjoint measurement approach for machine-part grouping in manufacturing cell formation

In order to measure the joint effects of a set of independent variables on the ordering of a dependent variable, conjoint measurement is used. In this paper the underlying principle is applied to clustering machine-part for cellular manufacturing. Assuming machines (parts) as products and parts (machines) as attributes of the products and elements of the incidence matrix as levels of attributes, the utilities of the machine (parts) have been calculated. The method is found to produce consistently good results when applied to a wide range of problems reported in literature.