Designing cellular manufacturing for next generation production systems

Designing cellular manufacturing systems is still under intensive study and has attained significant attention from academicians and practitioners. The major problem in designing cellular manufacturing systems is cell formation. Relevant design objectives, practical issues, and constraints should be taken into consideration. Although there are several cell formation techniques, more work is needed in the areas of the main design objectives, practical issues, and constraints. Over the last three decades, most of the approaches used in cell formation have been based on the machine-part incidence matrix alone and focus only on one or two practical issues sometimes including design objectives and constraints. The practical issues are processing time, alternative routings (process plan), part demand, production volume rate, machine capacity (reliability), and machine capability (flexibility). Hence, solving the cell formation problem is not a simple task, and it must be done concurrently and incrementally. Until now, there has been no practical cell formation approach. This void will lead to the proposal of a new cell formation strategy, which consists of five main phases to improve the quality of solution. In the first phase, a heuristic approach is used to group machines into machine cells based on the similarity coefficient between machines. The second phase uses another heuristic approach to form parts into part families while selecting the best process plans. Initial manufacturing cells are formed in the third phase. In the fourth phase, manufacturing cells are evaluated by measuring the manufacturing cells' performance. Revising the initial manufacturing cells will be included in the fifth phase by considering trade-offs between minimizing the intercellular moves and capital investments, maximizing the efficiency of clustering, and maximizing machine utilization to evaluate the optimal cell design. The proposed strategy was implemented and demonstrated through a numerical example.