TRAFFIC ANOMALY DETECTION FOR SURFACE STREET NETWORKS WITH THE MESH-WISED TRAFFIC INDICES ON MACROSCOPIC FUNDAMENTAL DIAGRAM

This paper describes the incident detection method for the surface street network based on the meshwised traffic indices which is called Traffic Scope. The Traffic Scope has been proposed for the visual comprehension of regional traffic situations in terms of the fluidity and the singularity of traffic conditions based on Macroscopic Fundamental Diagram. Through the dozens of case studies with real floating car data in Tokyo, we have so far found the singularity index has high sensitivity to the incident occurrence which will affect the traffic condition on surface street network. As an incident becomes serious, the area size affected by the incident will be larger. This implies us we may estimate the seriousness of the incident by counting the meshes which have higher singularities. In this paper, let us first introduce the concept of the Traffic Scope followed by the investigation on incident detection in Tokyo area.