This paper presents the results of some researches in the USA to seek definite associations between antilock braking systems (ABS) and risk of involvement in two-vehicle crashes. By concentrating on such crashes, the study could address the following questions: (1) How does an ABS affect a vehicle's risk of colliding into a vehicle that it is following? and (2) How does an ABS affect a vehicle's risk of being hit in the rear? In the first set of calculations, the influence of ABS on the Front-to-Rear Ratio, R (ratio of front to rear impacts) was determined on wet and dry roads. In the second set of calculations, an attempt was made to estimate the probability of front and rear impacts, by normalising with respect to side impacts, which are less likely to be influenced by ABS. Data from five states of the USA were used. The study used seven General Motors car models that had ABS as standard for 1992 but not for 1991. For driving on wet roads, ABS was estimated to reduce the value of R by 48%; the standard error of this estimate was about 6%. Assuming that accident risk can be estimated from side impacts, it was estimated that ABS reduces the probability of front impact by 32%, but increases probability of rear impact by 30%; these estimates have respective standard errors 8% and 14%. No significant dry road effects were found.
Abstract