Evaluating urban network efficiency and safety impacts of connected and autonomous vehicles in complex city environments

Abstract

The rapid advancements in autonomous vehicle technology, particularly in ride-sharing services within urban networks, emphasize the critical need for comprehensive research on their impact, especially as CAV (connected and autonomous vehicle) operators target new markets. This study addresses the pressing gap in evidence regarding the effects of connected and autonomous vehicles (CAVs) on urban infrastructure, a concern for communities that lack a thorough understanding of how these technologies will influence traffic on their network. By modeling a 13 sq. km network of downtown San Jose using the VISSIM microscopic traffic simulation tool, this research assesses both the operational and safety performance of the network at varying market penetration rates for the CAVs. The key evaluation metrics include average travel times, delays, and speeds, alongside surrogate safety assessments to quantify simulated conflict types.

Notably, the findings indicate significant improvements in roadway performance and safety correlating with increased CAV penetration, with average stop delays and overall vehicle delays decreasing by up to 11% and 7%, respectively. However, the maximum platoon size did not significantly enhance these benefits. This phenomenon may be attributed to the inherent complexities of urban networks, which present numerous interruptions, such as traffic signals and multimodal traffic accessing the network from several points. Based on the Surrogate Safety Assessment Model (SSAM) conducted, while the number of critical crossing conflicts decreased, a rise in lane-change and rear-end near misses was identified with increasing CAV penetration rates. As communities consider allowing CAV operations, this research not only highlights the potential advantages of integrating CAVs into urban traffic systems but also emphasizes the necessity for informed planning to harness their full benefits while mitigating potential challenges. Ultimately, understanding the dynamic interactions between CAVs and human-driven vehicles (HVs) and other multimodal traffic is essential for developing effective strategies that promote sustainable and efficient urban mobility.