Automotive vehicle crash safety requirements have steadily become more stringent over the last decade. Automobiles of tomorrow have to comply with a host of requirements in various crash modes in order to be considered roadworthy. In the first section of the paper, the current major requirements, some important requirements that are imminent, and desirable requirements in the near future are briefly discussed.Until recently, crash requirements have been focused mostly on the vehicle structure rather than the occupant protection, with the exception of frontal crash. Scarcity of in-depth interpretation of accident data, lack of biofidelic injury assessment devices (“crash dummies”) and the necessity for test repeatability had kept the testing procedures simple. Often, crash testing involved statically loading the vehicle to measure the structural strength, without consideration of the dynamic behavior of the structure or the structure/occupant interaction. But the technological gains in testing/instrumentation, advances in biomechanics leading to a better understanding of the injury mechanisms, availability of crash dummies, and the growing “safety” consciousness of the consumers have made the requirements increasingly more complex but realistic. With the increase in the complexity of the requirements, the complexity of the computer simulations has also increased. To simulate a crash test now requires the coupled simulation of the vehicle structure, barrier structure, restraint systems and the occupant surrogate. These advances, especially with respect to finite element methodologies, are discussed in the second section.In the third section, a case study of a passive restraint is presented. Energy-absorbing knee bolsters are becoming popular along with airbags. In particular, the design considerations for knee bolsters are discussed. It is shown how transient, nonlinear finite element methods can be effectively utilized to design knee bolsters. Comparison of simulation results with those of sled tests are also presented.