Travel time reliability plays a pivotal role in the system efficiency and level of service of transportation networks. Transportation network users are heterogeneous, and they may value travel time reliability differently. The importance level of travel time reliability for different travelers depends upon many factors including the user’s risk acceptance level and trip purpose and departure time. Thus, travelers tend to maximize travel time reliability, in addition to minimizing their travel times. One of the main challenges in transportation planning is the high computational time of traffic simulation tools that consider heterogeneous users and their responses to travel time reliability. Path finding problem constitutes an essential problem in these traffic simulation tools. Therefore, this study presents two heuristic algorithms to improve the computational time of reliable path finding algorithms by reducing the network size of each specific origin and destination pair in stochastic time-dependent networks. The network contraction algorithms, presented in this study, are based on the comparison of optimistic and pessimistic solutions resulting from minimum and maximum travel time realizations of a Monte-Carlo simulation-based approach. The major contribution of the proposed approach is to improve computational efficiency of the stochastic path finding problem, considering travel time correlations and travelers’ heterogeneity, in large-scale applications. Comparing the performance and accuracy of the approach with those of the approach without any network contraction for two large-scale networks demonstrates significant computational improvements and a high accuracy level. Different traffic and demand management strategies have also been used to improve reliability of transportation networks. These strategies, including congestion pricing, have great impacts on users’ reliable path choices. Considering a reliability measure in the travelers’ path choices naturally impacts the congestion pattern, which in turn, affects the outcomes of pricing strategies. Furthermore, congestion pricing alters link travel time distributions in stochastic transportation networks. Therefore, this study finds an equitable pricing scheme that minimizes the total travel time of auto users in a general bimodal network considering heterogeneous users with different values of time and reliability. The main contribution of this proposed approach is accounting for travel time reliability in finding an equitable pricing scheme. This approach is successfully applied to a modified Sioux Falls network to explore the impacts of subsidization strategy, congestion level, and considering travel time reliability on the pricing strategy and its effectiveness. Finally, emerging technologies, such as connected and autonomous vehicle technologies, have attracted the attention of transportation system planners in recent years, as an alternative to improve mobility and reliability of transportation networks. Having a traffic simulation tool that considers the presence of these technologies is essential to estimate their impacts on traffic congestion and travel time reliability. Therefore, this study presents a mesoscopic simulation tool to account for the presence of connected and autonomous vehicles at the network level by incorporating adaptive fundamental diagrams due to the non-uniform distribution of different vehicle types with heterogeneous drivers. This tool is then used to investigate the impacts of a mixed traffic of connected, autonomous, and human-driven vehicles on traffic flow and travel time reliability at the network level. The results show the superiority of connected and autonomous vehicles over regular vehicles in mitigating traffic congestion and improving travel time reliability.
Read
