Traffic state evaluation and intersection-movement-based incidents detection of expressway network

Currently, the national expressway has more than 120,000 kilometers, ranking first in the world. Unfortunately, the attendant problems such as the traffic congestion, traffic accidents, environmental pollution, and other negative effects also increased rapidly almost day by day. In this article, the mesoscopic traffic flow simulation model has been improved by inputting the expressway data. The estimated method of the expressway network running state is proposed, which takes approach proportion, or choice probability of intersection movements, as input information. Upon the designed mesoscopic traffic simulation process, the traffic accidents detection method is put forward under the condition of lacking the real-time data on road. The real expressway network of Shandong province and the toll data are used to test the validity of the proposed method. The results show that the proposed model and method can be more accurate in estimating the real-time network running state and the relevant macroparameters, such as the traffic flow speed, the averaged density, and flow. It is also feasible and effective to detect traffic accidents on road sections of the network. The proposed model and method can help expressway management department publish information, lay detectors, and conceal accident cover-up by the subordinate departments.