Multi-criteria decision making approach based on immune co-evolutionary algorithm with application to garment matching problem

In this paper, we propose a co-evolutionary immune algorithm for the multi-criteria decision making (MCDM) model, and use the model to solve the large scale garment matching problem. Size fitting problem is a main obstacle to large scale garment sales and online sales because it is difficult to find the fit garments by the general size information. This study regards the fit garment matching problem as a MCDM model with the constraints of size satisfaction. An immune co-evolutionary algorithm is used to search the fit garments from the candidate garments in the stock. The garments in the stock are taken as antibodies and the customer request as antigen. The concepts of ideal garment and loose garment are virtual garment to model the customer request. Two affinity measures including dominance affinity and distance affinity are defined to represent the similarity of antibody to antibody and that of antibody to antigen. Correspondingly, the fit garments are chosen by two methods: the Pareto optimal garments by the MCDM solving algorithm, and the optimal garments with the minimal distance affinity to the ideal garments. An evaluation model on garments is proposed to evaluate the fit garments by the affinity measures. Based on the experiment data from the effective study of detail factors of female trousers, the proposed model and algorithm demonstrate to be a feasible and effective attempt aiming at a valuable problem and provide the key tool for garment store sale system or online garment order system to support accurate garment size matching.