
Conditional inference tree‐based analysis of hazardous traffic conditions for rear‐end and sideswipe collisions with implications for control strategies on freeways
Author(s) -
Li Zhibin,
Wang Wei,
Chen Ruoyun,
Liu Pan
Publication year - 2014
Publication title -
iet intelligent transport systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.579
H-Index - 45
eISSN - 1751-9578
pISSN - 1751-956X
DOI - 10.1049/iet-its.2012.0203
Subject(s) - collision , inference , hazardous waste , crash , upstream (networking) , transport engineering , computer science , collision avoidance , engineering , computer network , artificial intelligence , computer security , waste management , programming language
Identifying hazardous traffic conditions related to traffic collisions is important to the development of real‐time traffic control measures for preventing collision occurrences. The primary objective of this study is to explore the hazardous traffic situations on freeways for rear‐end and sideswipe collisions to assist the development of control strategies for mitigating collision risks using the conditional inference tree method. Based on the 3‐year crash data and traffic data from a freeway corridor on the Interstate 880 in California, the conditional inference tree was developed and validated for each collision type separately. Results showed that the hazardous traffic conditions were different between different types of collisions. The occurrence of rear‐end collision was mainly related to the magnitude of lengthwise traffic variation between upstream and downstream locations. The occurrence of sideswipe collision was related to the type of freeway segment as well as the crosswise traffic variation between adjacent lanes. The control strategies for the mitigation of collision potentials were discussed according to the appearances of the conditional inference trees developed in this study.