This paper proposes a moving objects database model, dynamic transportation network based moving objects database (DTNMOD). Besides, a real-time traffic flow statistical analysis method based on moving object trajectories is provided. In DTNMOD, the underlying transportation network is modeled as a dynamic graph so that the state, topology, and traffic parameters of the transportation graph at any time instant can be tracked and queried.Moving objects are modeled as moving graph points which move only within predefined transportation networks.The data model is given as a collection of data types and operations which can be plugged as attribute types into a DBMS to obtain a complete data model and query language. To evaluate the performance of the proposed model,this paper has implemented a prototype system based on PostgreSQL and conducted a series of experiments. The experimental results show that DTNMOD provides good performances in processing region and join queries.