Building the Traffic Flow Network with Taxi GPS Trajectories and Its Application to Identify Urban Congestion Areas for Traffic Planning

Traffic congestion is becoming a critical problem in urban traffic planning. Intelligent transportation systems can help expand the capacity of urban roads to alleviate traffic congestion. As a key concept in intelligent transportation systems, urban traffic networks, especially dynamic traffic networks, can serve as potential solutions for traffic congestion, based on the complex network theory. In this paper, we build a traffic flow network model to investigate traffic congestion problems through taxi GPS trajectories. Moreover, to verify the effectiveness of the traffic flow network, an actual case of identifying the congestion areas is considered. The results indicate that the traffic flow network is reliable. Finally, several key problems related to traffic flow networks are discussed. The proposed traffic flow network can provide a methodological reference for traffic planning, especially to solve traffic congestion problems.