| description abstract | Traffic congestion monitoring is a long-term concern in urban areas. However, due to the complex structure of urban road networks and large amounts of traffic data, it is necessary to find an efficient way to identify traffic congestion in urban areas. In the big data era, more and more researchers are using traffic data to model traffic road networks and to identify traffic dynamics. Through the grid mapping method, this paper proposes an efficient abstraction approach to simplify the structure of a road network and then to identify urban traffic congestion. Based on the probe vehicle trajectory data, the intersection nodes between trajectories and grid boundaries are clustered through the method of density-based spatial clustering of applications with noise (DBSCAN). Then, a new traffic performance index is established by the principal component analysis (PCA) method based on the traffic characteristics in the node network. With the case study in Beijing, the proposed method effectively identifies urban traffic congestion in spatial and temporal dimensions. The proposed method is map-independent because it is only based on the probe vehicle data without a digital map. The method is highly efficient for a large urban road network in practice because all the calculations are basic operations based on the cells. Moreover, the proposed method can distinguish the expressway and the frontage roads. The mean absolute error (MAE) is about 10 km/h and the root-mean-square error (RMSE) is lower than 14 km/h. This method is expected to provide valuable spatiotemporal information for traffic engineers and managerial personnel to identify and relieve the traffic congestion problem. | |
