Examination of Pedestrian Throw Models in Real-World “Wrap” Vehicular Impact: A Case Study

Modeling of pedestrian throw may be based on regression fitting to existing data (Type I model) or modeling pedestrian kinematics through physical laws (Type II model). Validation of these models is typically limited to controlled laboratory studies as vehicle impact speed is frequently unknown in real-world impact. Additionally, vehicle impact to multiple pedestrians is uncommon and rarely analyzed. A video-recorded July 4, 2020, “wrap” impact between a vehicle and two pedestrians provided the necessary data to explore the performance of pedestrian projection models in the context of a real-world collision. This study examined the output of different pedestrian throw models using this real-world collision. The true throw distance was determined via analysis of captured video, as was the vehicle impact speed. Five different pedestrian throw models were analyzed using measured pedestrian throw distance as input. The impact velocity reconstructed with each model was compared with each other as well as the calculated vehicle impact speed from the video recording. The models showed general consistency with each other, and all models apart from the Han-Brach Mechanism Model demonstrated mean outputs within 10% of the calculated impact velocity. It was determined that Type I models yielded a larger range of vehicle impact velocities by comparison to Type II models, and thus Type II models are recommended by the authors for practical use. Additional insight into the required inputs and usefulness of each model was developed. The results of this case study support the use of pedestrian throw models in the accident reconstruction of real-world impacts.