Is vehicle automation enough to prevent crashes? Role of traffic operations in automated driving environments for traffic safety.

Automated driving systems (ADSs) are expected to prevent traffic accidents caused by driver carelessness on freeways. There is no doubt regarding this safety benefit if all vehicles in the transportation system were equipped with ADSs; however, it is implausible to expect that ADSs will reach 100% market penetration rate (MPR) in the near future. Therefore, the following question arises: 'Can ADSs, which consider only situations in the vicinity of an equipped vehicle, really contribute to a significant reduction in traffic accidents?' To address this issue, the interactions between equipped and unequipped vehicles must be investigated, which is the purpose of this study. This study evaluated traffic safety at different MPRs based on a proposed index to represent the overall rear-end crash risk of the traffic stream. Two approaches were evaluated for adjusting longitudinal vehicle maneuvers: vehicle safety-based maneuvering (VSM), which considers the crash risk of an equipped vehicle and its neighboring vehicles, and traffic safety-based maneuvering (TSM), which considers the overall crash risk in the traffic stream. TSM assumes that traffic operational agencies are able to monitor all the vehicles and to intervene in vehicle maneuvering. An optimization process, which attempts to obtain vehicle maneuvering control parameters to minimize the overall crash risk, is integrated into the proposed evaluation framework. The main purpose of employing the optimization process for vehicle maneuvering in this study is to identify opportunities to improve traffic safety through effective traffic management rather than developing a vehicle control algorithm that can be implemented in practice. The microscopic traffic simulator VISSIM was used to simulate the freeway traffic stream and to conduct systematic evaluations based on the proposed methodology. Both TSM and VSM achieved significant reductions in the potential for rear-end crashes. However, TSM obtained much greater reductions when the MPR was greater than 50%. This study should inspire transportation researchers and engineers to develop effective traffic operations strategies for automated driving environments.