Abstract: To explore the evolution of car supply chain system,a multi-agent modeling method was used to simulate and analyze data from the perspective of a complex adaptive system.The Agent behavior characteristics and overall structure of car supply chain were researched,and a four-layered supply chain complex system model with manufacturers as the core was established on AnyLogic7 platform.The simulation for nonlinear characteristic and evolution rules of a car supply chain was analyzed.The research showed that the system exhibited the typical characteristics of a complex adaptive system with chaotic motion;its evolution experienced formation,growth and maturity over three stages,and the focuses for supply chain point of different stages were also different.This research had  practical guiding significance for supply chain management.

Key words: multi-Agent, car supply chain, complex adaptive system, evolution rule

摘要： 为探索轿车供应链网络的演化规律,从复杂自适应系统的角度出发,运用多Agent建模仿真方法进行研究。分析了轿车供应链各Agent的行为特征及网络总体结构;利用AnyLogic7平台构建了一个以多个制造商为核心的四级轿车供应链网络仿真模型,并对轿车供应链系统的非线性特性、各Agent和网络整体的演化规律进行了仿真分析。研究表明,该系统具有复杂自适应系统的典型特征,其运动呈现混沌状态;该系统的演化经历了形成、成长、成熟三个阶段,不同阶段供应链上节点的竞争策略重点不同。该研究对供应链管理具有实际的指导意义。

关键词: 多智能体, 轿车供应链, 复杂自适应系统, 演化规律