Modeling and Forecasting the Urban Volume Using Stochastic Differential Equations

Traffic flow prediction can be used for the management of traffic control systems and can be applied toward improving traffic light split times at intersections. In this paper, we developed a methodology for the short-term prediction of traffic flow using the stochastic differential equation (SDE). Since the current volume depends on the previous short-term volume and time, we used the Hull-White model. With the proposed method, a flexible short-term prediction of volume is suggested. It is possible to simulate traffic conditions easily and also detect incidents precisely. This method is applied in Tehran´s highways, and the obtained results are compared with previous artworks. Our results offered a better fit to the traffic volume.