Abstract
Cell formation in manufacturing has been successful for mutually exclusive incidence matrices. Some methods are catered for only mutually exclusive matrices but in reality these methods provide no clue to their success unless the problem is attempted. Cluster Identification Algorithm devised for such matrices was later modified by Boctor ( International Journal of Production Research 29 (6), 1081-1100, 1991) to incorporate the mutually nonexclusive situation. However, in a worst case scenario Boctor's algorithm also failed to identify distinguishable clusters. A new computational simple heuristic for grouping parts is proposed here to overcome these discrepancies. Test results reveal that the heuristic performs well for both exclusive and nonexclusive incidence matrices.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have