Validation of a child FE model is a great challenge due to the lack of sufficient data for children. In their previous study, the authors have developed a methodology for validating a child FE model by reconstructing pedestrian accidents and comparing predicted and observed injuries. However, the study reconstructed only one accident case and more validation cases were needed for enhanced confidence of the estimated material property. The current study therefore reconstructed two additional child pedestrian accident cases. In addition, published 3–point bending test results of child long bones were also used for quantitative assessment of the material property.
The accident cases were taken from the PCDS and CIREN databases. Based on the information from the accident data, a multi–body simulation and optimization procedure were used to identify impact conditions. The estimated impact conditions were used to simulate the accident cases using the FE model with failure criteria. The predicted injuries using various sets of material parameters representing individual variation were compared with observed injuries. Published 3–point bending tests for child femora and tibiae were simulated using the models for these bones. The upper and lower bounds of the estimated individual variation of the material parameters were applied to the models to obtain predicted response corridors and to compare with the test data.
The results of the FE accident reconstruction showed that the observed lower limb fracture patterns were reproduced within the estimated individual variation of material property. It was also found that most of the 3–point bending test results fell within the predicted force–deflection response corridors. Combining these two results, the accuracy of the estimated average material parameters and their individual variations were quantitatively assessed.