Stability of THUMS Pedestrian Model and Its Initial Trauma Response against a Real-Life Accident

With dramatically rapid development of computing and modelling technology, occupant and pedestrian safety models went through the development of crash test dummies and multi-body mathematical dynamic modelling to finite element pedestrian human model (THUMS 4.0). THUMS 4.0 is a state of art human model which includes a skeleton structure, as well as internal organs and soft tissues, which makes it a suitable candidate to analyse accident trauma. This paper aims at investigating the stability of the THUMS pedestrian model as well as performing an initial injury comparison against a set of real-life pedestrian collisions. The study will use a dataset of fatal accidents which have been recorded by West Midland Police (UK). The paper will initially review a typical real-life pedestrian collision and focus on throw-away distance, vehicle impact speed, pedestrian injury and trauma autopsy results collected and recorded from the accident. Based on the loadcase parameters collected, a real-life accident will be reconstructed and simulated. During the simulation process, selected responses of the THUMS pedestrian human model will be closely monitored, including throw-away distance, contact impact forces and accelerations, as means of capturing injury and trauma pattern as well as strains and displacements on impacting vehicle structure. Injury and trauma pattern will be investigated against autopsy results and will be used to conclude on the current trauma capability predictions of the THUMS human pedestrian model.