Search Results

This paper contains a review of methods for deriving risk curves from biomechanical data obtained from impact experiments on human surrogates. It covers many of the problems and pitfalls of obtaining realistic human risk curves from impact experiments. The strength and weakness of both parametric and non-parametric methods are evaluated. The limitations of standard analysis of censored impact test data are presented. Methods are given for determining risk curves from both doubly censored data and data obtained from impacts to body regions in which there are more than one mechanism of injury. A detailed set of examples is presented in which different experimental data are analyzed using the Consistent Threshold method and the logistic approach. Finally risk curves for published data are presented for the femur, head, thorax, and neck.

Two barriers commonly used to evaluate the response of a vehicle in a frontal impact are the rigid barrier and the offset deformable barrier, each produces different deformation patterns. One possible cause of the difference is that an impact into a rigid barrier generates significantly greater stress waves than impacts in the real world resulting in final deformation patterns that are different from those seen in the field. To evaluate this hypothesis, models of two types of rails, one for a truck design and one for a passenger vehicle design undergoes two different types of impacts. Both rails are analyzed using an explicit dynamic finite element code. Results show that the energy perturbation along the rail depends on the barrier type and that the early phase of wave propagation has very little effect on the final deformation pattern of both rails.

Cadaver data from experiments and data from field studies collected for the purposes of risk analysis are almost always censored. If the form of the underlying distribution is known, then the best method of analysis is to estimate the parameters using a maximum likelihood approach, but if it is not known, the best method is a non-parametric approach. The Consistent Threshold Estimate, introduced in this paper, is a method to estimate the underlying distribution that is both non-parametric and a maximum likelihood estimate. In this paper, we will use it to estimate threshold HIC values for skull fracture or tissue damage, but it can be used for any application that has censored data. In addition to mathematically defining the Consistent Threshold estimate, a simple method to compute it for doubly censored data is given and it is compared to other estimators by means of Monte Carlo Tests. The Consistent Threshold estimate is then applied to experimental head impact to cadaver data.

The development of the Head Injury Risk Curve (HIRC) and the Skull Fracture Risk Curves (SFRC) that were proposed by Prasad and Mertz for the adult driving population are reviewed, and the problem with using the Maximum Likelihood method to analyze the cadaver 15ms HIC data is discussed. The cadaver data base for skull fracture is expanded by including non-fracture 15ms HIC values for a number of cadavers which had skull fractures at higher impact severities and by including cadaver test results of Ono and Tarriere. This expanded data set was analyzed using the Mertz/Weber method and a new method called “Certainty Grouping”. An updated version of the Skull Fracture Risk Curve (SFRC) is proposed. The efficacy of this revised curve is demonstrated by comparing its predictions to the results of simulated fracture impacts using a finite element model of the head. The HIRC was not changed since no additional brain damage data were analyzed.

The interaction ILLEGIBLEf the chest of the Hybrid III dummy with the air bag restrILLEGIBLEt system during a crash is complex. Forces applied to one ILLEGIBLEmponent of the dummy can generate an unexpected response in a distal part. Motion, both linear and angular, of the pelvis during impact can create an enigmatic spike in the acceleration of the chest. Because significant changes in the chest acceleration response can affect the development of an airbag system, this pelvis-chest interaction is cause for concern. The factors that appear to affect the chest acceleration spike as a result of the pelvis-chest interaction are: the mass moment of inertia of the pelvis, the interaction of the pelvis with the femur, the characteristic of the lumbar spine, and the differential velocity of the pelvis with respect to the chest.