Abstract
Design of experiment and correlation analysis for solving cell formation problem with alternative routings. Cell formation (CF) problem involves grouping parts into families and machines into cells. The CIF problem becomes extra complicated if parts are assigned with alternative routings. In this context, we propose a new two-phase approach. The first phase is based on Designed of Experiments. The objective is to realize the best alternative routing for each part in term of CIF performance criteria. In addition, the result goes across and beyond the previous CIF procedures by providing a sensitivity analysis aspect. A new performance criterion which is based on the percentage of conservation of incidence matrix data is proposed and validated. The second phase consists in applying CIF based on correlation analysis to cluster the machines and parts. In order to illustrate the two phases of the proposed approach, two numerical examples are provided. Furthermore, the results of a comparative study based on multiple performance criteria show that the present approach is valid.