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

The hub-and-spoke logistics network has a wide application within the design of supply chain, tourism routing, air transportation network, postal delivery network, etc. Today Chinese logistics industry was rapidly developing, small size as well as dispersed distribution, so the hub-and-spoke logistics network would do great help to collaborate logistics resource, improve efficiency and reduce cost for Chinese logistics industry. Compared to the nonstop logistics network, the hub-and-spoke logistics network has the advantages of producing scale merit, collaborating logistics resources, developing the third party logistics, reducing conveyances and so on, while it also brings the disadvantages of bypass cost and node cost, etc. To exert the advantages, avoid the disadvantages and popularize the application of hub-and-spoke logistics network, this dissertation concentrages on location and routing optimization, and studies the problems of multi-allocation hub maximal covering, cost & route optimization for hub-and-spoke network, network design with multiple cost functions or node cost fundamentally with some innovative perspectives, and demonstrates the empirical research based on sereral airline transportation networks and postal delivery networks.Firstly, the thesis introduces the definitions and characters of hub as well as hub-and-spoke logistics network, explains the background and the significance of choosing this topic, summarizes the theories of hub location and relative optimization problems and proposes the research objection and research route.Secondly, the thesis introduces the model of multi-allocation hub median problem, which approached by tabu search and applied to one case of Chinese hub-and-spoke airline network.Thirdly, the thesis presents an improved model for multi-allocation hub maximal covering problem, provides genetic algorithm, tabu search and path relinking approach for the problem. Both the model and the heuristics are experimented on AP data set. The model and heuristics are also applied to one case of Chinese hub-and-spoke airline network.Fourthly, the thesis presents an improved model for multi-allocation hub set covering problem, and provides the heuristics of genetic algorithm and scatter search for the problem. Both the model and the heuristics are experimented on AP data set. We evaluate the design result of multi-allocation hub set covering problem by AP2-11 network.Fifthly, according to the disadvantages of multi-allocation hub median problem and hub covering problem, the thesis presents the cost & route optimization problem for hub-and-spoke network (CROPHN) by a new hypothesis. We establish a mixed-inter model and a greedy approach for the CROPHN. Both the model and the algorithm are experimented on AP data set. The design result of the CROPHN is evaluated by AP2-11 network.Sixthly, due to the shortcoming that the CROPHN only concerns two cost functions between cost and flow, the thesis presents the problem of network design with multiple cost functions (HNDMCF). We establish a mixed-inter model and a greedy approach for the CROPHN. Both the model and the algorithm are experimented on AP data set. We compare design result of the CROPHN and HNDMCF by AP2-11 network.Seventhly, due to the shortcoming of node cost, the thesis discusses the problem of hub-and-spoke network design with nodes cost. The empirical analysis for Chinese hub-and-spoke airline network with node cost is presented by the model and algorithm in paper.Finally, the thesis gives conclusion to the contents and the innovative achievements of the research, and presents the future scope, the purpose and the prospect of this topic in further studies.