Competitive Logistics Network Optimization Based On Customer Value

The structure of logistics network and its operating tactics are key factors for determining the efficiency of enterprise’s logistics, and also are important guarantees to achieving the enterprise’s strategy. Competitor and customer as the two major external entities the enterprise must to face, have a significant impact on the enterprise’s business. Logistics network optimization as a part of enterprise’s strategic decisions, will inevitably restricted by the competitor and customer. Logistics network optimization should considers how to meet customer needs, but also should pay more attention to the differences of customer value, so the enterprise’s logistics network structure and operating strategies can grasp the demands of high-value customers more better and achieve their own interests. In addition, competitor in the market can’t be ignored in decision-making. Once the layout of the enterprise’s logistics terminal node is changed, the market competition situation will change, and will lead to counterattack from the existing competitor. To optimizing the logistics network so that the enterprise could occupy competitive advantages even in the competitor’s counterattack is an important issue worthy of study.The optimization of competitive logistics network based on customer value studies the problem mainly from the both aspects of customer value and competitor. It uses bottom-up logic to optimizing the overall logistics network. It makes the enterprise can not only meet customer requirements from the customer’s view, but also consider the repaying brought by the different-value customers in competitive environment from its own point of view. Looking at the text of this article, the major work done as follows:(1) The life cycle stages of customer relationship are elaborated and the meaning of customer value in this article is clarified. The measure method of customer’s potential value is proposed based on the analysis of existing prediction method. The customer are subdivided into VIP customer, strategic customer, potential customer and underper-formance customer according to the difference of customer’s current and potential value. The customer’s transfer trends between different types are discussed. The strategies for different types of customers are discussed in order to maintaining high-value customers and promoting the realization of the customer’s current and potential value.(2) The targets are set for different types of high-value customers in optimization of logistics network base on the different customer value and the corresponding service goals. A bi-level programming model of single-product logistics network in static competitive environment is established using bottom-up optimization logic, which considers the layout of competitor’s logistics network, the antinomy between logistics cost and service level, the customers’ accepting product prices and other factors. In the model, the customer demands are flexible. An iterated algorithm is designed by the parallel selection genetic algorithm and the software of Lingo aimed to the characteristics of this model. Finally, a case is analyzed; the decision-making data of the enterprise and competitors are compared, and the characteristics and advantages of enterprise’s decision-making are analyzed.(3) The competition in market is dynamic. The adjustments of enterprise’s logistics network or operating tactics will lead to relative adjustments of existing competitor. Aimed at the differences of customer value and the dynamic reactions of competitor, an optimizing model of single-product logistics terminal facilities is established firstly, which considers the influences of product pricing, service level, distance, communication convenient and construction level of terminal facility on competing the demands of high-value customer. In the model, the enterprise acts as the leader, the competitor acts as the follower, and the customer demands are flexible. In order to achieve overall optimization of the logistics network, the optimization model of the upstream logistics network is established, which considers the factors of production cost, transportation cost, operating cost of distribution center, time, capacity constraint and transportation mode. An iterative algorithm is designed to solve the problem based on the elitist and fast non-dominated sorting genetic algorithm (NSGA-II) according to the features of the model. Finally, a case is analyzed and the decision-making data are compared, the results show that, even in the dynamic reactions of competitor, the enterprise gains advantages on both of the sales and the profits.(4) Due to market competition and customer growth, the customer’s value at different times will changes, and the service level, product price and other factors also have continuing influences in competing for high-value customer’s demands. Aimed to the dynamic changes of customer value and competition, and the continuing impact of service level and product price on attracting customer value as well, the optimization model of single-product multi-cycle logistics terminal facilities is established in the dynamic competitive environment. What’s more, the optimization model of the upstream logistics network is established. An iterative algorithm is designed to solve the problem based on NSGA-II, but the arithmetic crossover is used affter the segment-crossover in order to improve the search performance of NSGA-II. Finally, a case is analyzed and the decision-making data are compared. The results show that the enterprise grasps advantages both on profits and sales since it considers the dynamic changes of customer value in the optimization.(5) Aimed to the situation of most of the logistics network are used to distribute a variety of products as well as customer demands are uncertainty, the logistics terminal facilities and upstream logistics network optimization model are established in dynamic competitive environment based on the dynamic changes of customer value, with multi-product, multi-cycle and stochastic customer demands. For the characteristics of the model, the improved NSGA-Ⅱ algorithm and adaptive genetic-annealing hybrid algorithm are used to solve the case. The decision-making data of the improved algorithm and the original algorithm are compared. The results show that the improved algorithm makes enterprise gains more sales, but also makes competitor find more profitable decision-making scheme.