Research

Dynamical organization and bottlenecks identification of city traffic
With the development of urban transportation systems towards multilayer of complex networks, the traffic state in cities comes from the interaction between dynamical competitions of traffic flows and the topology evolution of transportation networks. Considering the explosion and heterogeneous distribution of population, the improvement of traffic utility has become an important research topic in the management of future city. Understanding the dynamical organization of city traffic is one of the fundamental scientific questions, whose typical phenomenon is the alternate transition between free flow and congestion.
While the research on the alternate transition between free flow and congestion in highways has received much attention, this issue is rarely studied on the scale of city network due to the complex network feature of city traffic. Especially, isolated local flows will form during the global congestion, which can integrate into global flows in the state of free flow. This dynamical integration of local flows into global flow has not been fully understood due to the lack of real data and relevant methods.
To investigate the above question, we have collected the real-time traffic data of the Liuliqiao region of Beijing, and based on the definition of functional networks, explored the evolution of city traffic flow. We found the percolation process in the realistic organization of city traffic [1]. In this type of organization, there exist many high-velocity local clusters connected by directed paths. Links with lowest velocity along these paths (called “bottlenecks”) determine the organization efficiency of city traffic. Furthermore, we found that the bottlenecks evolve with time according to traffic dynamics in a day, which are different from structural bottleneck links found by traditional network analysis. Small improvement of the identified bottlenecks can significantly improve the global traffic.
The above findings can help to understand the dynamical organization feature of city traffic. Based on the traffic data and percolation theory, our dynamical analysis method can evaluate the real-time reliability of transportation network systems. Based on the criticality of network percolation, we have presented a strategy to identify bottlenecks and improve the global utility, which may help to optimize the free state of city traffic and support the dynamical management of city traffic. Until now, the research on the management of city traffic is still full of challenge and needs further efforts of exploration. In the meanwhile, we believe that based on the big-data technology, research of city traffic reliability will support the development of intelligent transportation.


Daqing Li,associate professor, school of reliability and systems engineering, Beihang University, E-mail: daqingl@buaa.edu.cn or li.daqing.biu@gmail.com

Reference
[1] Daqing Li, Bowen Fu, Yunpeng Wang, Guangquan Lu, Yehiel Berezin, H. Eugene Stanley and Shlomo Havlin. Percolation transition in dynamical traffic network with evolving critical bottlenecks. PNAS 112(3):669-672 (2015)





Fig. 1 Percolation process in city traffic and its critical value qc





Fig. 2 Bottlenecks evolution and the corresponding traffic organization