Evolution of Traffic Flow Analysis under Accidents on Highways Using Temporal Data Mining

Evolution of traffic flow on highways, especially under accidents, is crucial to understanding and reducing the impact of accidents. This paper proposes a model, based on temporal data mining, to describe traffic flow evolution when accidents happen on highways. Time series model was constructed by using Cell Transmission Model to reflect the state of traffic flow by ternary numbers. To overcome the defect of Euclidean distance, which does not consider the linear drift in the time domain, and to reduce the computational cost, Discrete Fourier Transform was implemented to turn the time series from time domain to frequency domain. By clustering analysis, the traffic dynamics can then be studied. A numerical experimentation was carried out and the result showed the effectiveness of the proposed method. The newly developed method will provide a better insight into the evolution of traffic flow on highways and the impact of highway crashes.