Exploring Driver Error at Intersections

Abstract

A large portion of crashes occur at intersections, and most such crashes are associated with driver mistakes. Severe mistakes may lead to serious injuries; therefore, it is necessary to investigate the factors that contribute to driver error and how those factors influence driver behavior. More information on these contributing factors can help researchers develop cost-effective countermeasures that might help mitigate driver error. The primary objective of this study was to examine key contributors to driver error that took place at uncontrolled, sign-controlled, and signalized intersections. An ordered-probit statistical model and a data-mining technique called “association rules” were implemented to explore these relationships. The results of both approaches were consistent. Association rules were found to be capable of discovering patterns in the data that could not be found in the ordered-probit statistical model. The secondary objective of this study was to provide new insights on how to improve intersection safety by adding to the knowledge regarding the contributing factors of those driver errors. Most, if not all, errors are related to human factors; thus, they can effectively be corrected through a holistic approach that involves engineering, enforcement, and education.