N. Huan, E. Yao, Y. Yang, T. Lu

Pages: 21-34

Abstract
With the rapid development of freeway in many countries, the road network provides multiple route choices to meet drivers’ inter-city travel demand, which also pose challenges to the traffic management, such as the estimation of traffic flow pattern, performance evaluation of tolling policy, etc. Thus, it is essential to study drivers’ route choice behavior under the real condition of decision-making. In this paper, a route choice model for passenger car drivers in freeway network was established for characterizing the impact of various factors including toll fee on route choice behavior. First, a stated preference (SP) survey was conducted to investigate drivers’ route choice preferences under different scenarios. Second, a multinomial logit (MNL) model, considering three types of utility variables: route attributes, trip characteristics and personal attributes, was formulated and calibrated using the SP survey data. The reliability of estimated coefficients in utility function were further discussed. Finally, a sensitivity analysis was performed to explain how the toll rate changes affect the share rate among candidate routes for four main types of drivers. The results showed that freeway toll rate has a greater influence on the long-distance travel while a weaker impact on short-distance travel. When the toll rate rises, some drivers with lower income are more likely to transfer from freeways to national or provincial roads, while drivers who travel for business purposes are less sensitive to the change of toll rate.
Keywords: freeway; toll rate; route choice behavior; MNL model