Traffic Flow Analysis in Mixed Environments with Autonomous and Human-Driven Vehicles

Abstract

Autonomous vehicles (AVs) have the potential to improve traffic efficiency, safety, and environmental sustainability; however, their impacts under mixed traffic conditions remain unclear. This study investigates the effects of cautious autonomous vehicles on traffic performance at different MPRs(market penetration rates) in a mixed environment with human-driven vehicles. A microscopic traffic simulation was developed using VISSIM and applied and applied to the Harbour Toll Road in Jakarta, Indonesia with driving behavior parameters calibrated to reflect realistic human-driven and autonomous vehicle characteristics. Several scenarios were analyzed, ranging from fully human-driven traffic to fully autonomous traffic, with AV MPR in 10% increments. Traffic performance was evaluated using level of service (LOS), average speed, traffic volume, vehicle delay, queue length, fuel consumption, and emissions. The results show that at low AV penetration levels, mixed traffic conditions can initially worsen performance due to conservative autonomous driving behavior and interaction conflicts with human drivers. In these scenarios, delays, congestion, and emissions increase compared to conventional traffic. As the AV MPR increases, traffic performance improves significantly. At penetration levels above 90%, cautious autonomous vehicles substantially enhance level of service, increase average speed and traffic throughput, and reduce vehicle delays and queue lengths. Environmental benefits are also observed at high penetration levels, with notable reductions in emissions and fuel consumption. These findings indicate that the positive impacts of autonomous vehicles strongly depend on high adoption rates and highlight the importance of supportive infrastructure and traffic management strategies during the transition to autonomous mobility.

Keywords: Autonomous Vehicles, Mixed Traffic Flow, Market Penetration Rate, Microscopic Simulation

DOI: 10.54941/ahfe1007290