Vehicle rollover crashes are a serious public health concern in the United States with far reaching socioeconomic impacts. Rollovers occur in only 3% of crashes, but account for a disproportionately high number of the serious traffic injuries and fatalities, 30% in 2008 (NHTSA, 2010). Currently in the United States, the only federal motor vehicle safety standard (FMVSS 216) governing vehicle crashworthiness for rollover is based on a quasi-static roof crush test that requires each vehicle to have a minimum strength-to-weight ratio (SWR>3.0). FMVSS 216 has been criticized for not accurately representing vehicle structural response to real-world dynamic rollover crashes, as well as for the choice of peak SWR as the evaluation metric. This thesis is a collection of three studies aimed at evaluating the relationships between force, energy, and deformation in rollover crashes. It is hypothesized that for a vehicle in a single roof-to-ground rollover impact that greater total kinetic energy at touchdown will result in larger structural deformation, and that an increase in the peak SWR of the vehicle will not reduce deformation if it is not accompanied by a similar increase in energy absorption.

In Study I, the dynamic force-deformation response of a vehicle subjected to an inverted drop test displayed comparable stiffness, with limited rate dependence, to the quasi-static evaluation on the same vehicle. An on-board optical system was validated against independent deformation measurement from string potentiometers for tracking dynamic high-rate, 3D, multi-point roof deformations across a large area of vehicle structure. In Study II, the effects of variations in touchdown conditions on vehicle kinematic and deformation response were characterized through two full-scale dynamic rollover tests using replicate mid-sized SUV’s on the DRoTS. The two tests sustained similar peak roof deformations while the test with less total kinetic energy produced a larger peak reaction force. Contrary to the results of previous finite element (FE) studies (Parent, 2011), the vehicle with the larger drop height did not sustain more deformation. In Study III, the effects of variations in the distribution of initial kinetic energy (vertical, rotational, translation al) on the force and deformation response of the vehicle roof structure in a rollover were characterized through FE simulations. Eight single roof-to-ground impact simulations were performed using a FE model of a mid-sized SUV in LS-DYNA, with the three kinetic energies each varied between a higher and lower value. Only vertical kinetic energy was found to be a significant predictor of peak roof deformation (p=0.002), while none of the kinetic energies were a significant predictor of peak reaction force. Further, peak reaction force was not a significant predictor of peak roof deformation.

From the results of the three studies, it was hypothesized that ranking vehicles solely by peak SWR is not an optimal assessment of roof strength, and a more appropriate metric would be based on energy absorption. Two additional simulations were performed on a reinforced version of the FE vehicle which demonstrated that an 11% increase in the normalized energy absorption of the structure reduced deformation in a rollover crash simulation by 5%, whereas a 20% increase to peak SWR had no effect. This lead to the conclusion that an increase in the peak SWR of a vehicle will not reduce deformation if it is not accompanied by a similar increase in energy absorption. Therefore, it is recommended that an energy absorption criterion be added to FMVSS 216.