A system dynamics approach for assessing the impacts of autonomous vehicles on collision frequency

Abstract

This study employs a System Dynamics (SD) approach to assess the long-term impact of Autonomous Vehicles (AVs) on traffic collisions. By modeling key variables affecting collision frequency, the SD framework represents existing transportation systems and incorporates AV adoption to evaluate changes in collision rates. The models predict the frequency of Property Damage Only (PDO) and Fatal-Injury (FI) collisions at different levels of AV market penetration. Two policy scenarios are examined: incentives for shared AVs and higher occupancy rates for shared AV services. Results indicate significant improvements in traffic safety as AV penetration increases, particularly when coupled with shared AV usage and advanced sensing and communication technologies. The models provide valuable insights into the complex interactions between AVs and collision-related factors. These findings support the development of effective policies to guide the adoption of AV technology, reduce accidents, and enhance overall road safety in the long term.