文章编号： 1672-5328（2019）01-0096-06

吴若乾，周勇，陈振武
(深圳市城市交通规划设计研究中心有限公司，深圳市交通信息与交通工程重点实验室，广东深圳518021)

摘要： 已有的城市交通网络瓶颈识别研究往往关注于路网结构特征或是路段交通流的属性，而较少 从整体网络层面研究交通流的组织变化情况。将统计物理学中的渗流理论引入城市交通研究，从整 体网络的交通连通性角度出发，探究交通网络中的全局连通交通流崩溃为局部连通交通流的动态组 织变化过程。整个过程类似于统计物理的渗流相变过程，因此也被称为交通渗流。基于深圳市的城 市交通数据，对实际道路网络进行抽象建模，利用实时浮动车数据构建动态交通流网络研究交通渗 流现象。通过对交通渗流的相变临界过程进行探究，识别对于维持交通流全局连通性具有关键作用 的交通瓶颈。该研究方法从复杂网络和统计物理学的角度为城市交通研究提供了新视角，并为识别 城市交通的瓶颈路段和治理城市交通拥堵提供参考。

关键词： 交通连通性；交通渗流；交通拥堵；交通瓶颈；深圳市

中图分类号： U491.1+12

文献标识码：A

Identifying Urban Traffic Bottlenecks with Percolation Theory

Wu Ruoqian, Zhou Yong, Chen Zhenwu
(Shenzhen Urban Transport Planning Center Co., Ltd., Shenzhen Key Laboratory for Traffic Information and Traffic Engineering, Shenzhen Guangdong 518021, China)

Abstract： Existing urban traffic studies often focus on the characteristics of road network structure or the traffic flow of road segment. Little attention has been given to the traffic flow performance of overall network. This paper introduces how to use the percolation theory commonly applied in statistical physics to the urban transportation study. That is to study the connection between the dynamics of network system breakdown and localized traffic flow congestions based on the integrated network system perspective. This working process is similar to the percolation phase in statistical physics, thus it is called traffic percolation. Taking Shenzhen as a case study, a network model is formulized based on actual city roadway network, which is used to investigate the phenomenon of traffic percolation with the data from floating car. By exploring the critical process of phase change of traffic percolation, the bottlenecks, which play key roles in maintaining the overall network connectivity, are identified. This research method provides a new perspective in studying urban traffic from the perspective of complex network and statistical physics. it also provides a reference for identifying the bottleneck sections and alleviating urban traffic congestion.

Keywords： traffic connectivity; traffic percolation; traffic congestion; traffic bottlenecks; Shenzhen