Supplier selection: an analytic network process and imprecise goal programming model integrating the decision-maker's preferences

The supplier selection decision is very difficult and complex for the supply function manager. Supplier selection is considered as a multi-criteria problem that includes both quantitative and qualitative criteria. Moreover, the purchasing decision-maker (DM) considers several conflict objectives simultaneously such as maximising the total value of purchases, minimising the total cost of purchases and minimising the total rate of defective products. In this paper, we propose a new approach for selecting suppliers based on the combination of the imprecise goal programming (IGP) and the analytic network process (ANP). All the goals considered are interval-valued. The concept of satisfaction functions will be used to integrate explicitly the DM's preferences. This approach allows us to respond to two key strategic questions: Which suppliers are the best? and How much should be purchased from each selected supplier? To answer the first question, we used the ANP method to evaluate suppliers. Then, the weights of suppliers obtained by the ANP are included in the IGP to determine the quantity purchased from each supplier. A numerical example illustrates the proposed approach.