A physical model capable of credible simulation of human trauma is badly needed in automobile safety research. Development of such a model has been limited, not only by lack of biological and engineering data, but by lack of a coherent, sequential plan of attack on the problem. Such a plan, based upon the systems approach, is offered. The plan has two major phases. The first phase includes qualitative analysis, hypothetic modelling, and program planning; the second provides for emperic research, system modlling, and design, fabrication, and test of a physical model.