Are Expected and Observed Effectiveness of Emergency Brake Assist in Preventing Road Injury Accidents Consistent

This paper proposes to estimate and to compare the expected and the observed effectiveness of the Emergency Brake Assist (EBA) in terms of reduction in injury accidents in France. The evaluation of the expected effectiveness of EBA is based on the simulation of the reduction in injuries in non-EBA cars which could result in lower collision speeds resulting themselves in higher mean deceleration, would EBA have been available and applied in those cars. A sample of fatal police reports, for which most of the vehicles involved in an accident, braking distance, collision speed and injuries outcome are available, is used for the simulation. The evaluation of the observed effectiveness of EBA follows a 3-steps process: (1) The identification, in the French National injury accident census, of accident-involved cars for which the determination of whether or not the car was fitted with EBA is possible. A sample of 917 cars involved in injury accidents that occurred from January 2000 to June 2004 was selected. (2) The identification of accident situations for which we can determine whether or not EBA is pertinent. (3) The calculation, via a logistic regression, of the relative risk of being involved in an EBA-pertinent accident for EBA equipped cars versus unequipped cars, divided by the relative risk of being involved in a non EBA-pertinent accident for EBA equipped cars versus unequipped cars. This relative risk is assumed to be the best estimator of EBA effectiveness. Both evaluations result in a good effectiveness of EBA. Furthermore, the rather consistent estimations out coming from expected (-7.5 % of car occupants fatalities, -10 % of pedestrian fatalities) and observed (-11 % of overall injuries) effectiveness of EBA validates the methodology used for the expected effectiveness.