Pedestrians are regarded as an extremely vulnerable and high-risk group of road users since they are unprotected in vehicle impacts. More than 1.17 million people throughout the world are killed in road traffic accidents each year. Where, about 65% of deaths involve pedestrians. The head injuries in vehicle-pedestrian collisions accounted for about 30% of all reported injuries on different body regions, which often resulted in a fatal consequence. Such injuries can result in disabilities and long-term sequence, which lead to significant social costs. It is therefore important to study the characteristics of pedestrian accidents and understand the head injury mechanism of the pedestrian so as to improve vehicle design for pedestrian protection. The aim of this study is to investigate pedestrian dynamic response and develop head injury risk functions.In order to investigate the effect of pedestrian gait, vehicle front geometry and impact velocity on the dynamic responses of the head, the multi-body dynamic (MBD) models were used to simulate the head responses in vehicle to pedestrian collisions with different vehicle types in terms of head impact point measured with Wrap Around Distance (WAD), head relative velocity and impact angle. A simulation matrix is established using five vehicle types, and two mathematical models of the pedestrians represented a 50th male adult and a 6 year old child as well as seven pedestrian gaits based on typical postures in pedestrian accidents. In order to simulate a large range of impact conditions, four vehicle velocities (30 km/h, 40 km/h, 50 km/h and 60 km/h) are considered for each pedestrian position and vehicle type.A total of 43 passenger car versus pedestrian accidents were selected from In-depth Investigation of Vehicle Accidents in Changsha, China (IVAC) and German In-Depth Accident Study (GIDAS) database for simulation study. According to real-world accident investigation, accident reconstructions were conducted using multi-body system (MBS) pedestrian and car models under MADYMO simulation environment to calculate head impact conditions, in terms of head impact velocity, head position and head orientation. In order to study kinematics of adult pedestrian, relationship curves: head impact time, throw distance, head impact velocity and vehicle impact velocity, were computed and logistic regression models: head impact velocity, resultant angular velocity, HIC value, head contact force and head injuries, were developed based on the results from accident reconstructions.The automobile windshield, with which pedestrians come into frequent contact, has been identified as one of the main contact sources for pedestrian head injuries. In order to investigate the mechanical behavior of windshield laminated glass in the caseof pedestrian head impact, windshield FE models were set up using different combination for the modeling of glass and PVB, with various connection types and two mesh sizes (5 mm and 10 mm). Each windshield model was impacted with a standard adult headform impactor in an LS-DYNA simulation environment, and the results were compared with the experimental data reported in the literatures.In order to assess head injury risks of adult pedestrians, accident reconstructions were carried out by using Hybrid III head model based on the real-world pedestrian accidents. The impact conditions were obtained from the MBS simulation, including head impact velocity, head position and head orientation. They were used to set the initial conditions in a simulation of a Hybrid III FE head model striking a windshield FE model. Logistic regression models, Skull Fracture Correlate (SFC), head linear acceleration, Head Impact Power (HIP), HIC value, resultant angular acceleration and head injuries, were developed to study brain injury risk.{...].