Research On Hub Location And Routing Distribution For Hub-and-Spoke Logistics Network

With the development of the economic globalization, Hub-and-Spoke logistics network has become the effective network structure which could help to collaborate logistics resource, reduce logistics cost and improve operational efficiencies. Hub nodes and links are two important parts of Hub-and-Spoke logistics network which has a great influence. Therefore, research on location and routing distribution for Hub-and-Spoke logistics network is particularly important. This paper focused on hub location and routing distribution for Hub-and-Spoke logistics network. According to the theory of queuing theory, the hub location is optimized and the design of flow distribution under deliver time requirements is investigated which based on minimizing the total cost in the networks. By reducing the waste of resources in the process of the shipments transportation, the operation efficiency of logistics network is improved and network operators and customers to achieve a win-win objective.In this paper the mixed Hub-and-Spoke logistics network for multi hub-multi distribution and multimodal is developed. The flow of spoke nodes is allowed to be assigned to more than one hub node. There are two types of shipments to be selected between the origin city and the destination city, i.e., direct shipments or hub-based shipment. The shipments can travel directly using direct road shipments and the same as the hub-based shipment between hub and spoke, except that it offers the choice between road and rail transportation. Firstly, the paper summarizes the current research situation and problem at home and abroad, and the models and algorithmic techniques for Hub-and-Spoke logistics network are expatiated systematically. And then, basing on summarizing the literatures of the domestic and abroad, combined the analysis of the developing process and current research status of Hub-and-Spoke logistics network, the bi-level programming model is developed. The upper model which based on minimizing the total cost is proposed by using p-hub median approach. It takes the fixed hub operating costs, modal connectivity cost, the transportation costs, waiting time, service time, inter-hub transportation discount factor and service time requirement into account. The lower model is used to achieve waiting time on the basis of the theory of serial queuing system. A tabu search is selected to solve the mathematic formulation by comparison analysis of algorithms. The solution, strategy and implementation process of the algorithms are developed. Using the tabu search to solve the small-scale problem verify the effectiveness of the model and algorithm.Finally, through the actual data of25coals producing and selling places from Beijing, Tianjin and Jin Shan Yu, the validity of model and algorithm are verified with MATLAB software. According to the factorial experiment, the different parameters between discount factor and service time requirement are designed and the different results can be achieved. The conclusion of this paper can be used as a theoretical reference for network operator to manage Hub-and-Spoke logistics network.