Research On Multi-Objective Order Allocation Model Under Centralized Procurement

Purchasing management is an important part of supply chain management. The costs of procurement affect directly the profit of the enterprise. Reduce procurement costs is conducive to improving the competitiveness of the entire supply chain. Centralized procurement can form a scale advantage and reduce the purchasing cost. It is a kind of procurement mode which is widely adopted by the group enterprises. Considering the suppliers’ supply capacity and demand uncertainty, enterprises generally maintain multiple identical or similar vendors for the same kind of material. Order allocation can lead to competition among suppliers, which can help enterprises to obtain the products with good quality and low price. Therefore, it is of great practical significance to study the order allocation problem in the centralized procurement mode.In this paper, we study the order allocation model of centralized procurement and decentralized delivery of the group companies with multiple factories to purchase the same kind of material from multiple suppliers. First, the present status and existing problems of domestic and foreign research on the vendor selection and the distribution of purchasing quantity are analyzed, and the principle of Non-dominated Sorting Genetic Algorithm with elite strategy (NSGA-II) is introduced. Second, in the situation of plants’uncertain demand and total quantity discount, a multi-objective optimization mathematical model is developed based on overall cost, quality and delivery lead time. A modified non-dominated sorting genetic algorithm-II is devised to solve the multi-objective optimization problem. In order to increase searching efficiency of the algorithm, the search space finite method is used to deal with the problem constraints. When the gene value after crossover or mutation beyond the scope of the decision variables, it is set to the corresponding extremum, to ensure that the corresponding solution for the new individual is a feasible solution. Finally, a numerical example is used to verify the feasibility and effectiveness of the model and the algorithm.