Sensitivity Analysis of Simulated Postimpact Vehicle Motion Using Design of Experiments (DOE)

An important component of the process of the reconstruction of a vehicle crash involves the modeling of the motion of the vehicle(s) before and after a collision. Depending on the conditions, this motion might be modeled using a vehicle dynamics simulation program. In the simulated dynamics of vehicle motion, the tire forces are the predominant means by which the path of the vehicle is determined, with aerodynamic loads being the other force acting on the vehicle. Recent literature on this topic investigated the effect of the steer angle of the front wheels on the postimpact trajectory of a light vehicle for a large initial angular velocity. This paper looks more broadly at the modeling of light vehicle postimpact motion using vehicle dynamics simulation but for a wider range of factors.

Design of experiments (DOE) is used to rank the effect of various physical factors of vehicle postimpact motion. The response variable used in the DOE analysis uses the rest position of the vehicle (characterized by the x and y coordinates of the CG and the vehicle heading, θ) for a given combination of factor changes. The results of the study show that in the four different designs that were conducted, a trend in the response was consistent. The single factor that consistently appeared in the various DOE analyses (with various factor combinations) was the tire-to-roadway frictional drag coefficient. Various other factors and 2-factor combinations were also found to be significant. Some of the significant factors are not intuitively obvious, such as aerodynamic drag.