Numerical reconstruction of injuries in a real world minivan-to-pedestrian collision

The purpose of this study was to evaluate the capability of the Total Human Model for Safety (THUMS) - pedestrian model in predicting pedestrian injuries, and to investigate pedestrian injury mechanisms in minivan collisions via numerical reconstruction of a real world minivan-to-pedestrian impact case.

A typical minivan-to-pedestrian collision case was selected from the In-depth Investigation of car Accidents in Changsha (IVAC) database. The THUMS middle-size adult male FE model and a minivan front FE model were then employed to represent the case participants and injuries to the pedestrian's lower limb, thorax and head were reconstructed. Finally, the capability of the THUMS model in predicting pedestrian injuries and pedestrian injury mechanisms in minivan collisions were analyzed through comparisons between predictions and the accident data.

The results show that the THUMS has a good capability in predicting pedestrian thorax injuries, but a lower prediction of leg bending moment and brain strain. The extra bull bar concentrates crash load to pedestrian's leg and raises tibia/fibula fracture risk, thorax injuries in the struck side are mainly from direct contact at the lower chest level, lung injury in the non-struck side could be caused by inertia force from the heart. Rotational acceleration shows good match with brain strain and could be the key mechanism for concussion.

The results suggest that further improvement in biofidelity of the THUMS model is still needed. The findings also offer basic understanding on pedestrian injury mechanisms in minivan collisions.