Abstract: The construction of rational air freight network is one of the important means to improve the competitiveness for express enterprises. Hub-spoke air freight network is an effective model to integrate air freight resources, to improve efficiency of resource utilization, and to increase the competitiveness of express enterprises. This paper establishes a Hub-Spoke air express model from three aspects, i.e. hub location decisions, assignment relationship decisions and hub number decisions, with the goal to achieve the lowest air freight network cost and shortest express processing time. Based on this, this paper chooses genetic algorithm to compute the costs of networks with different number and locations of hubs. Gravity model is applied to the determination of assignment relationship, and super-efficiency DEA model is applied to decide the number of hubs. Finally, the model and algorithm are verified to be very effective in the real case of SF express company, which includes 17 nodes in the network and is going to be extended rapidly. The model and algorithm prove to be useful for express companies to establish an efficient and responsive Hub-Spoke network to gain competitiveness.

Key words: logistics management, air express network, “hub-spoke”, genetic algorithm, super-efficiency DEA model

摘要： 快递业竞争激烈,构建高效合理的航空货运网络是快递企业提高竞争力的重要手段。“枢纽—辐射”式航空货运网络是整合航空快递资源、提高航空快递资源利用效率、提高快递企业竞争力的有效模式。本文以降低航空快递网络成本、加快航空快递处理时间为目标,从航空快递网络枢纽的选取、指派关系的确定、枢纽个数的选择三个方面研究了航空快递网络模型建立问题,选用遗传算法求解不同枢纽个数下航空快递网络的运输成本,并据此进行枢纽的选取,运用重力模型法进行指派关系的确定,在此基础上运用超效率DEA模型确定枢纽个数。接着,以包含17个节点的顺丰航空快递网络的规划为例,对本文所提出的模型和算法进行了验证,验证结果证实了模型的合理性。本文的研究为快递企业构建航空货运网络提供了科学实用的方法,该方法的使用可以降低航空货运成本,提高效率,从而提高快递企业的竞争力。

关键词: 物流管理, 航空快递网络, “枢纽—辐射”, 遗传算法, 超效率DEA模型