On solving the problem of manufacturing cell formation using decision factors in a genetic algorithm

The main purpose of group technology in a manufacturing context is the design and creation of production cells in which each cell specializes in producing one family of components (i.e. parts). This is achieved through clustering parts and machines based on the quantity of similar operations. The main problem of switching between a traditional manufacturing layout to a Group Technology-based layout is grouping parts into families. As result of an exhaustive survey of previous contributions, a revised and improved genetic algorithm is proposed to solve the cell formation problem. This algorithm includes a part-allocation scheme that assigns parts to machine cells by means of decision factors. These factors quantify each part contribution to the global solution efficiency. Other feature that distinguishes this algorithm is the retention of good characteristics from previous methodologies such as the chromosome codification and removes the burden of local heuristics used to allocate pieces to machines. In order to validate this proposal, a total of 28 problems were selected from the scientific literature and solved through the proposed algorithm.