A calculation method of OD matrix in multi-modal transit network based on traffic big data

Traditional method to obtain public transport OD matrix by massive sample surveys cost a lot of resources and cannot reflect the actual travel demand. Many researchers have put forward various methods to estimate public transport OD matrix based on limited public traffic data. In order to take advantage of the big data and obtain the public transport OD matrix in city scale, this paper proposed a method to calculate the route OD matrix in multi-modal transit network. Firstly, the boarding stations of the passengers can be obtained by matching the IC card data with bus stop data recorded in the process of bus running. Then, for propose of analyzing the destination of the passengers, the IC card records in a day are divided into three categories: 1) only one IC card record, more IC card records of the passenger should be involved, 2) two IC card records, if they have the same route, the boarding station of next record can be designated as the destination simply. If not, the route information should be involved, and 3) more than two IC card records, space-time information and the IC card data by subway are all should be considered to determine which station is the destination. Finally, the route OD matrix is obtained. The method had been implemented to compute the route OD matrix using traffic big data collected in Guangzhou, China. The results show that the proposed method can accurately calculate the public transport route OD matrix of whole city without traffic surveys, and it is naturally more reliable than other OD estimation methods.