A COMBINED SYSTEM OF STATIC TRAFFIC ASSIGNMENT AND ANN DELAY MODEL: SOLVING THE CONVERGENCE OF ITERATIVE OPTIMIZATION OF NON-MONOTONIC TRAFFIC ASSIGNMENT

Transportation planning models, which estimate traffic volumes on transportation network links, are often unable to realistically consider travel time delays at intersections. Introducing signal controls in models often result in significant and unstable changes in network attributes including the monotonicity of link travel time, which, in turn, leads to inability of planning models to arrive at a network solution based on travel costs that are consistent with the intersection delays due to signal controls. Simultaneous optimization of traffic routing and signal controls has not been accomplished in real-world applications of traffic assignment. A delay model dealing with five major types of intersections has been developed using artificial neural networks (ANN). The delay estimates by the ANN delay model have satisfactory percentage root-mean-squared errors (%RMSE). A combined system has also been developed that includes the ANN delay model and a user-equilibrium (UE) traffic assignment. The combined system employs the Frank-Wolfe method to achieve a convergent solution, although the global optimum may not be guaranteed.

Dr. Zhen Ding is currently struggling as a transportation analyst in Dallas, Texas of the U.S. He managed his Ph.D in transportation engineering at Florida International University, after his master study in University of Toledo, Ohio of the U.S. He cherishes his undergraduate study in Zhejiang University. China is his motherland.