Abstract
Injury severity in vehicle crashes is influenced by various factors within the pre-crash environment, including the environment, vehicle, driver, and other attributes. To investigate the relationships among these factors and identify the key determinants of injury outcomes, this study employed a graph neural network to model complex interactions and dependencies from a police-reported tabular database. The analysis revealed critical contributors to injury severity, uncovering the relationships among the variables in the pre-crash environment. These findings provide actionable insights for enhancing traffic safety and developing effective injury-prevention strategies.
