This paper describes a study to develop and validate methodology for simulating human infant head impact using the Hertz contact model. The study had two objectives. The first was to simulate Aprica 2.4 dummy head -rigid plate impact using the Hertz contact model to estimate head response variables. Model estimates were then compared with corresponding test variables. The second objective based on success of the first, was to evaluate the feasibility of using Hertz contact model to simulate human pediatric head – rigid plate impact at contact velocities ranging from 1.7 m/s to 6.26 m/s.

During objective 1 of this study, known geometric and material properties of the Aprica dummy head and steel plate were used as the Hertz model parameters. Model estimates of peak acceleration, peak head compression, pulse width, and time to peak displacement were compared with corresponding test data for contact velocity of 2.3 m/s. Percentage differences in response variable

s were:​ peak acceleration – 2.5;​ peak head compression – 1.3;​ pulse width – 6.1;​ and time to peak compression - 2.2. During objective 2 of this study, human head impacts were divided into 4 age groups – neonate (under 1-month), 5- months, 9-months and 11-months. Objective 2 was divided into 2 stages – Model building and Model validation. In the Model Building stage, a method was developed to estimate Hertz contact model parameters using human 30 cm drop test data. In the Model Validation stage, the model was used to estimate head response variables for 15 cm, 30 cm and 2 m drops for all four age groups and compared with human test data. Model estimates for peak head acceleration of neonates in 15 cm and 30 cm drop tests differed from average test peak head acceleration by 11%, and 13% respectively. Neonate estimated pulse widths for the same drop heights differed from test average by 0% and 1%. Maximum and minimum differences for 5-, 9-, and 11-month infant model estimates from average test values in 15 cm and 30 cm drops were:​ 13.47% and 0.44% for peak acceleration and 6.68% and 0.03% for pulse width. Simulation results of 2 m drops of 5-month, 9-month, and 11-month old heads indicated that estimated head Jerk (rate of change of acceleration) was very close to human test results. Since the pediatric heads sustained fractures, it was not possible to compare peak accelerations.

The model reproduced, very closely, the static force-deformation curve for 5-month old but provided poor estimates for some neonates. Model reproduced finite element model results for 30 cm drop test for 5-month old head on to concrete and hard foam. The proposed model and methodology provide a simple procedure to estimate pediatric head acceleration, head deformation and pulse width for contact velocities ranging from quasi-static to 6.3 m/s onto rigid and soft surfaces.