A transportation mode identification method based on mobile phone positioning

Transportation information is the foundation of traffic planning and management. Given the limitations of traditional acquisition method in quantity, coverage, and accuracy, the use of mobile phone positioning technology to obtain transportation information and conduct identification has good application prospects. With regard to computation complexity, excessive processing work, and low accuracy, this study has analysed the temporal and spatial features of three regular transportation modes, namely, bus transportation, driving transportation, and walking transportation (going to and from work on work days). This study has also proposed an identification method for the transportation modes based on the matching of temporal and spatial features. We have recruited volunteers in Guangzhou to conduct transportation acquisition for two weeks (10 work days) and used this method to perform transportation mode identification. Results show that the accuracy of bus transportation identification is 90.98%, that of driving transportation identification is 90.63%, and that of walking transportation identification is 88.60%. Therefore, the method has high accuracy. This study has demonstrated that the temporal and spatial matching method can accurately identify the three regular transportation modes and provide basic reference and decision-making support to traffic planning and management departments. Thus, the proposed method can be applied and promoted.