Supplier selection is an essential task within the purchasing function. A well-selected set of suppliers makes a strategic difference to an organization's ability to reduce costs and improve the quality of its end products. This realization drives the search for new and better ways to evaluate and select suppliers.; First, this research presents a three-phase methodology that integrates the various steps of the supplier selection process. This helps decision makers reduce a base of potential suppliers to a manageable number and make the final selection and order quantity allocation by means of multi-criteria techniques, such as the ideal solution approach, analytical hierarchy process (AHP), and goal programming. The first two, respectively, are used to reduce a large number of potential suppliers. The last one is used to decide the final order allocation. For illustrative purposes this three-phase methodology was applied to a manufacturing facility located in Tijuana, Mexico.; Second, this research considers the importance of inventory management in determining the optimal order quantity from selected suppliers. Two mixed integer nonlinear programming models are proposed to obtain optimal inventory policies that simultaneously determine how much, how often, and from which suppliers to order. They minimize the setup, holding, and purchasing costs per time unit under suppliers' capacity and quality constraints. The first model allows independent order quantities for each supplier and multiple orders from selected suppliers within an order cycle. This model outperforms an existing model in the literature. The second model restricts all order quantities to be of equal size, as required in a multi-stage [supply chain] inventory model. A closed-form solution is derived for the second model to determine the optimal inventory policy for the case when two potential suppliers are considered. Both proposed models allow the user to control the length of the order cycle time to streamline the inventory management process.; Next, the two optimization models discussed in the previous paragraph are extended to consider transportation cost. This consideration is important because it has been repeatedly overlooked in supplier selection literature. Since they are neither continuous nor convex, LTL transportation freight rates are approximated using either a linear or a power function to obtain near-optimal inventory policies. To obtain optimal policies for small to medium-size problems, actual LTL transportation costs are modeled with a piecewise linear function using binary variables. In the numerical example illustrated, the total cost per time unit obtained using the power function to estimate actual freight rates was only 1.4% greater than the optimal total cost per time unit.; Finally, given the prevalence of both supplier selection and inventory control problems in supply chain management, this research addresses these problems simultaneously by developing a mathematical model for an N-stage serial system. The model determines an optimal inventory policy that coordinates the different stages of the system while allocating orders to selected suppliers in Stage 1. A lower bound on the optimal total cost per time unit is obtained and a 98% effective power-of-two inventory policy is derived.
展开▼
