Urban transit system planning in large cities requires information about demands on passenger transfers. These demands are commonly represented as an origin-destination matrix (O-D matrix) that contains aggregated data of passenger turnover and traffic. Information, needed for these representations to be constructed, is usually collected using analytical models and natural surveys. Unfortunately, these methods do not provide necessary precision for high quality transport system management and are also quite expensive and hard to implement. Yet it shall be considered that modern public transport is equipped with special sensors, allowing to track the trajectory and journey time of every single kind of public transport, and validators are fixing the exact time of any passenger paying for his trip. Using modern information technologies in the city public transport today makes possible to easily collect all the necessary information to provide correspondences’ matrices, passenger turnover and traffic with any detalization (micro and macro levels included) by proceeding data, contained within the systems of public transport management and electronic trip payment control. Using information about payment time and information about vehicle movement it is possible to determine the exact time and place of passenger's trip starting and ending. This data analysis allows to combine several trips into a correspondence - chain of related trips, one passenger did in a row. This information not only allows to combine the origin-destination matrix, passenger turnover and traffic flow, but also provide dependences between any single correspondence and all the trips, that are included within. In our article we are describing our experience of building such a system using the public transport information in St. Petersburg, Russian Federation.