An integrated analytic network process with mixed-integer non-linear programming to supplier selection and order allocation

As the global business environment changed, much more commercial and technological competitions evolved in this environment. Consequently, more attention is devoted to the business partners including suppliers. In this paper, supplier selection with order allocation problem is addressed for strategic items with respect to the materials' characteristics as well as what has been mostly considered in the literature, suppliers' characteristics. The proposed model comprises an analytic network process (ANP) sub-model to qualify suppliers and filter-suitable candidates among the available ones. Additionally, the ANP sub-model is applied to evaluate suppliers regarding their qualitative attributes which cannot be modeled with mathematical tools properly. Moreover, a mixed-integer non-linear sub-model is presented to simultaneously allocate order quantities to the chosen suppliers. The model is customized to the strategic items by introducing supplier switching cost which is a notable cost component for products. Finally, experimental results and sensitivity analyses are presented to validate the proposed model.