Prior studies indicate that a majority of Hybrid III dummy models are validated over a limited range of loading velocities in accordance with the specification of CFR 49 Part 572. The shortcoming is that the dummy model response, based on validation at regulatory velocities, may not correlate well with experiments when loaded at different velocities. The fidelity of models at an extended range of velocities is important, as in car crash tests dummies are frequently exposed to a variety of loading conditions in terms of loading type and loading velocity, which are differing from that of the Hybrid III standard certification tests.

In this study, a finite element model of Hybrid III 50th percentile dummy with high-fidelity response is developed using the non-linear finite element code PAM-CRASH. The methodology implemented for the model development is presented, with particular focus on material calibration and validation of the model against experimental data at different structure levels (component level, sub-system level, and system level), under a wide range of loading velocities. In addition to compliance with the typical certification requirements, the developed model has reasonable correlations with the physical dummy for a series of loading conditions. The model response has proven to be robust and reliable while maintaining computational efficiency, showing good potential to be used for accurate prediction of occupant injury numbers in crash simulation.