Significant improvement in engineering productivity and technical competitiveness in the field of air bag restraint system design can be achieved through integration of the two technologies of Computerized Crash Simulation Modeling and Design of Experiments (DOE). The methodology for minimization of required physical crash tests conducted during the development of a new vehicle air bag restraint system is the topic of this paper. This technique is demonstrated through a hypothetical case typical of the development of a new driver air bag restraint system required to meet the Federal Government's targets regarding occupant injury measures (FMVSS 208). The objective of this demonstration is the establishment of an L8 DOE test matrix to be used as a baseline for sled testing. The three system design factors selected for this test matrix are screened from all possible system design factors and are analytically shown to have the most significant effect on the system's FMVSS 208 performance.The process of screening for these factors, which is based on factor average effects and analysis of variance of the system's FMVSS 208 performance using the mathematical crash simulation model, is presented and discussed.Finally, a discussion of the potential cost savings that can be realized from the use of this technique is presented.