Empirical estimates of seat belt effectiveness in two-car collisions

Abstract

The objective of this research is to quantify occupant protection afforded by restraint systems in two-car head-on collisions. An accepted measure of occupant protection is “Restraint System Effectiveness” (RSE). This quantity is defined as the percent reduction in the probability of an occupant injury due to the use of a restraint system. RSE estimates are calculated in the present paper using a previously developed mathematical model of occupant injuries in two-car collisions and obtaining parameter estimates from two data sets. The two sets of data used for estimating RSE are: New York police accident data and data collected by various multidisciplinary accident investigation teams. Both these data sets have been reported in the literature. The RSE estimates obtained from the two data sets range from 29 to 40%. These estimates are shown to increase as the mass of the occupant's vehicle increases, but decrease as the mass of the other colliding vehicle increases. On the average, the use of a restraint system can offset the decrease in occupant protection that would result from a 40% reduction in car weight. Consequently, the use of restraint systems to reduce occupant risk is estimated to be even more important as the number of small vehicles in the U.S. vehicle fleet increases.