The accuracy of the Simulation Model of Automobile Collisions (SMAC) computer program was evaluated in terms of its ability to predict the correct paths and damage profiles for vehicles involved in a crash. A comparison of the results from SMAC and EDSMAC were presented along with measured results from twelve staged collisions. Statistical analysis of those results revealed the average path error was 25 to 29 percent and the average damage profile error was 109 to 287 percent. A procedure was presented for improving the match between simulated and measured paths. After using this procedure, the average path error was reduced to -2 to 7 percent, and the average damage profile error was 54 to 186 percent. Collision Deformation Classification (CDC) predictions were very good. Damage profile errors, which did not reduce the program's overall effectiveness, were the result of the way the program computes inter-vehicle forces, leading to a recommendation that the algorithm be reformulated to include an initial force coefficient. The findings also led to a discussion of how the term 'accuracy' was defined for crash simulation programs. (A)
Abstract