An efficient solution to the system optimum dynamic traffic assignment problems based on Lagrangian relaxation

The study of dynamic route guidance model over a general road network has been deployed for some time in intelligent transportation systems (ITS) field. The paper presents the general formulation of the system optimal dynamic traffic assignment (DTA) problems, and enunciates the discrete form model. For the sake of computational convenience, the simplification of the model is explored according to the characteristics of realistic traffic system. Then the algorithm base on Lagrangian relaxation and quasi-Newton iteration is given to solve the simplified problem. Simulation is carried out to show the practicability and efficiency of the algorithm. Moreover some analysis on the simulation and guidance for future studies are presented at the end.