A multi-level risk framework for driving safety assessment based on vehicle trajectory

Few existing research studies have explored the re-lationship of road section level, local area level and ve-hicle level risks within the highway traffic safety system, which can be important to the formation of an effective risk event prediction. This paper proposes a framework of multi-level risks described by a set of carefully select-ed or designed indicators. The interrelationship among these latent multi-level risks and their observable indica-tors are explored based on vehicle trajectory data using the structural equation model (SEM). The results show that there exists significant positive correlation between the latent risk constructs that each have adequate con-vergent validity, and it is difficult to completely separate the local traffic level risk from both the road section level risk and vehicle level risk. The local and road level in-dicators are also found to be of more importance when risk prediction time gets earlier based on feature impor-tance scoring of the LightGBM. The proposed conceptual multi-level indicator based latent risk framework gener-ally fits with the observed results and emphasises the im-portance of including multi-level indicators for risk event prediction in the future.