modeling suburban freeway travel variability considering connected and autonomous vehicles

This study examines how coexistence of connected autonomous vehicles (cavs) and human-driven vehicles (hdvs) can affect travel time in the tehran-karaj suburban freeway during peak hours. employing machine learning algorithms and monte carlo simulation, researchers analyzed the effects of varying cav penetration rates (prs) on traffic flow patterns and travel time uncertainties. building on existing research, the study demonstrates significant travel time reductions offered by cavs, providing substantial benefits to road users. the research uniquely addresses the uncertainty inherent in the transition to cavs by employing monte-carlo simulation and probabilistic modeling to forecast potential impacts on travel times in mixed traffic states. during the morning peak, the study found that low cav prs could potentially increase travel time by up to 53%. medium and high prs exhibited more variable effects, with fluctuations ranging from – 11 to 26% and – 6 to 7%, respectively. in contrast, the evening peak saw decreases in travel time for medium and high cav prs, with maximum reductions reaching 30% and 42%, respectively. these findings contribute valuable insights to the ongoing discourse on integrating cavs into mainstream transportation networks, highlighting the complex interactions between cav penetration rates and travel time variability across different traffic conditions.