Abstract: The supply chain network growth is analyzed to understand the evolution of supply chain network, and the deficiencies of existing complex network models that are used to describe real supply chain network are pointed out. The initial network is star topology, and the link nodes connected with the new added one are chosen in the local world. The selection of the local world depends on the value of network path between nodes. and the size of the local world increases with time. Then, a new dynamic evolution model is proposed. Model generating algorithm of the supply chain network is put forward. The model considers both the network node removing and the inner and outer growth of the network so that the dynamic evolution of a true network can be described in more detail. Simulation result shows that the supply chain network has small-world, scale-free property.

Key words: supply chain, complex networks, local world, dynamic evolution model, scale-free, small-world, cluster coefficient

摘要： 为了深入理解供需网络的演化规律，研究了已有的复杂网络演化模型刻画供需网络生长过程的不足，提出了以星型网络表示初始网络，在局域世界中选择新增节点的连接节点，局域世界的选取，采用了依据节点之间的网络路径值作为选取局域世界的原则，同时定义了局域世界的规模动态增长，从而建立了复杂供需网络的动态演化模型，并给出了生成模型的算法。该模型在考虑网络动态增长的同时，也考虑到网络内部边的动态演化以及节点的退出，这与现实情况相吻合，因此更细致、真实地刻画了供需网络的动态演化特性。仿真结果表明，该模型所生成的网络模型具有无标度特性和小世界性。

关键词: 供需链, 复杂网络, 局域世界, 动态演化模型, 无标度特性, 小世界性, 聚集性