Reconsidering accident causation analysis and evaluating the safety benefits of technologies: final results of the TRACE project

The objectives of the EU-funded project TRACE (TRaffic Accident Causation in Europe, 2006-2008) are the up-dating of the etiology of road accidents and the assessment of the safety benefits of promising technology-based solutions. The analyses are based on available, reliable and accessible existing databases (access to which has been greatly facilitated by a number of partners highly experienced in safety analysis, coming from 8 different countries and having access to different kinds of databases, in-depth or regional or national statistics in their own country). Apart from considerable improvements in the methodologies applicable to accident research in the field of human factors, statistics and epidemiology, allowing a better understanding of the crash generating issues, the TRACE project quantified the expected safety benefits for existing and future safety applications. As for existing safety functions or safety packages, the main striking results show that any increment of a passive or active safety function selected in this project produces additional safety benefits. In general, the safety gains are even higher for higher injury severity levels. For example, if all cars were Euro NCAP five stars and fitted with EBA and ESC, compared to four stars without ESC and EBA, injury accidents would be reduced by 47%, all injuries would be mitigated by 68% and severe + fatal injuries by 70%. As for future advanced safety functions, TRACE investigated 19 safety systems. The results show that the greatest additional safety gains potential are expected from intelligent speed adaptation systems, automatic crash notification systems, and collision warning and collision avoidance systems. Their expected benefits (expected reduction in the total number of injured persons if the fleet is 100% equipped) are between 6% and 11%. Safety benefits of other systems are more often below 5%. Some systems have a very low expected safety benefit (around or less than 1%).