Opportunities to Preventing Rear-End Vehicle Crashes: Findings from Analyzing Actual Crash Data

To date most traffic crash analyses were conducted using aggregated crash data. The main focus was given to determining relationship between crashes and corresponding variables such as traffic volume, speed, speed variance and geometry conditions. Relatively little has been studied what caused such crashes at individual vehicular level. This is mainly due to lack of accurate data for such individual crashes. Recently, the Minnesota Traffic Observatory at the University of Minnesota developed a set of data containing 5 actual rear-end crashes at individual vehicular level. This paper analyzed these data and attempted to establish a trigger factor preventing crashes. An inverse of time-to-collision value of 0.4 detected all 5 actual crashes before the collision. However, it also generated a large number of false alarms. Additional trigger factor (i.e., the deceleration rate difference between leading and following vehicles set at 15 ft/sec2) completely eliminated those false alarms. In addition, it was found that an advanced warning intended to alert a driver to prevent crashes has little help. This was because a driver needs a reaction time of about 0.57 seconds – too late to avoid crash for most cases (only 20% actual crashes were avoided). This indicated that an automated braking system should be applied to prevent crashes or mitigate the crash impacts.