Identification and Classification of Spatiotemporal Traffic Congestion Based on Floating Car Data

Congestion identification is an important research issue in the field of transportation. How to differentiate the congestion which is frequent from andand which is an incident has become the main problem. Therefore, the purpose of this study is to identify the spatiotemporal traffic congestion of expressway sections based on floating car data and to distinguish the frequent congestion from the occasional congestion. The method is to analyze the spatiotemporal diagram of expressway speed and to extract the features of congestion. According to the congestion frequency and historical average speed, the congestion is classified by the k-means clustering method. The Guangzhou Airport Expressway is an application of this method. From the result, the spatiotemporal traffic congestion is clearly identified. Finally, the frequent, accidental, and occasional congestions are shown separately in the diagram and the causes of congestion are analyzed. Traffic managers can give improving suggestions of road performance based on it.