The invention provides a city road segment classification method based on traffic characteristics. By acquiring the GPS data of a floating vehicle, a map matching algorithm and a data statistical method are combined, the GPS data of the floating vehicle is effectively processed to obtain a speed time sequence of each research road segment, Pearson correlation coefficients between each two speed time sequence of the research road segments are calculated to screen the speed time sequences to obtain a speed feature time sequence, the research road segment is simply and preliminarily classified toa smooth road segment, a continuous congested road segment and a complex road segment according to the features of the speed feature time sequences, the dynamic time sorting distance and the statistics features of the sequences are combined to calculate the similarity between the speed feature time sequences, all the complex road segments are classified again based on the hierarchical clusteringalgorithm. The method can obtain the city road segment speed distribution mode to allow traffic management departments to perform management for road segments in different speed distribution modes.