Analyses of factors of crash avoidance maneuvers using the general estimates system

Taking an effective corrective action to a critical traffic situation provides drivers an opportunity to avoid crash occurrence and minimize crash severity. The objective of this study is to investigate the relationship between the probability of taking corrective actions and the characteristics of drivers, vehicles, and driving environments.

Using the 2004 GES crash database, this study classified drivers who encountered critical traffic events (identified as P_CRASH3 in the GES database) into two pre-crash groups: corrective avoidance actions group and no corrective avoidance actions group. Single and multiple logistic regression analyses were performed to identify potential traffic factors associated with the probability of drivers taking corrective actions.

The regression results showed that the driver/vehicle factors associated with the probability of taking corrective actions include: driver age, gender, alcohol use, drug use, physical impairments, distraction, sight obstruction, and vehicle type. In particular, older drivers, female drivers, drug/alcohol use, physical impairment, distraction, or poor visibility may increase the probability of failing to attempt to avoid crashes. Moreover, drivers of larger size vehicles are 42.5% more likely to take corrective avoidance actions than passenger car drivers. On the other hand, the significant environmental factors correlated with the drivers' crash avoidance maneuver include: highway type, number of lanes, divided/undivided highway, speed limit, highway alignment, highway profile, weather condition, and surface condition. Some adverse highway environmental factors, such as horizontal curves, vertical curves, worse weather conditions, and slippery road surface conditions are correlated with a higher probability of crash avoidance maneuvers. These results may seem counterintuitive but they can be explained by the fact that motorists may be more likely to drive cautiously in those adverse driving environments.

The analyses revealed that drivers' distraction could be the highest risk factor leading to the failure of attempting to avoid crashes. Further analyses entailing distraction causes (e.g., cellular phone use) and their possible countermeasures need to be conducted. The age and gender factors are overrepresented in the "no avoidance maneuver." A possible solution could involve the integration of a new function in the current ITS technologies. A personalized system, which could be related to the expected type of maneuver for a driver with certain characteristics, would assist different drivers with different characteristics to avoid crashes. Further crash database studies are recommended to investigate the association of drivers' emergency maneuvers such as braking, steering, or their combination with crash severity.