A Study of Vehicle Characteristics of Signalized Intersection based on Queuing Theory信号交叉研究.pdf

WSEAS TRANSACTIONS on MATHEMATICS Lian Xue, Minghui Wu, Hui Yan A Study of Vehicle Characteristics of Signalized Intersection based on Queuing Theory LIAN XUE, MINGHUI WU*, HUI YAN School of Computer and Computing Science Zhejiang University City College Huzhou Street 51, Hangzhou 310015 P.R. CHINA *corresponding author email: wmhmathzju@ Abstract: - This paper uses queuing theory to study the features of the vehicle’s arriving and passing of the signalized intersection. The first part of this paper presents the background, purpose and meaning. In the second part, the arrival process and serving process are studied, especially the delay process is mainly analyzed including the dispersion vehicle headway time distribution and vehicle delay at signalized intersections is analyzed. In the third part, time series, cubic spline difference and gray prediction, RBF neural network are used to get the number of vehicles arrived outside of next 18 cycles. Based on the queuing theory, the forth part introduces a predict model on the vehicle’s arrival. It predicts a certain intersections’ average length, average waiting time and average length of stay in the next 18 cycles. The fifth part optimizes the intersection signal timing and gets the best signal cycle time. Scientific model and more complete analysis are founded in the paper. These methods have a good reference value to the relevant departments. Key-words : - signalized intersection, vehicle characteristics, queuing theory, RBF neural netw