Supplier Selection in Closed Loop logistics network design using Monte Carlo simulation with Robust-Fuzzy Approach

Globalization of economic activities along with the rapid growth of technology as well as limited resources have placed companies in a tight competition. Among the competitive advantages for companies is to make activities such as the supply chain more efficient and effective. Since suppliers exert a fundamental influence on the success or failure of a company, it is known as a strategic task. Considering the importance of supplier selection, in this paper a reverse logistics network model is designed for supplier selection under uncertainty. The objective functions of the designed model include minimizing the total cost, the total number of defective parts, the timely delivery of all parts to the customer, and the hazardous environmental factors associated with suppliers. In order to be closer to reality, parameters such as demand, transportation cost, product production costs, and product purchase price are considered uncertainty, and robust-fuzzy approach is used to deal with uncertainty parameters in modeling. Finally, in order to avoid weighting in multi-objective model decision making, Monte Carlo simulation has been developed to determine the total number of Pareto solutions from the presented model. The results of the evaluation of the mathematical model with the robust-fuzzy approach show that as the penalty coefficients of the objective function increase, the cost of the total supply chain network increases, but its standard deviation decreases. This issue shows the high capability of the robust method in controlling the uncertainty model of the problem.