On the development of a theoretically based, statistically justified, thoracic injury criterion

Advanced crash test dummies are being designed with multiple deflection measuring capabilities in the thorax to allow better characterization of the chest’s local response to impact and thus, better optimization of performance from systems such as belt/bag combinations or four-point restraints. Historically, the injury criteria used to interpret thoracic impact responses were derived empirically from simple parameters, such as peak acceleration and deflection, that were extracted from available experimental data. This study takes a different approach. It combines the vast knowledge of structural responses and interactions embodied in our finite element modeling technologies in a model of the thorax, validates the model’s impact response capability by mimicking the experiments by Kroell et al. (1972) and demonstrates that applying the measured x-y deformation-time histories of two points on the anterior chest wall of a dummy to the model, is sufficient for the model to accurately reproduce the complete two-dimensional deformation shape of the entire thoracic slice. This then allows the potential for the prediction of injury to be made on the basis of local stresses and strains occurring throughout the entire slice over time. Discussion of the development and validation of the slice model concept of post processing of dynamic dummy response output using FE models, verification of minimum necessary dummy inputs, and validation of predictive capabilities are all presented.