Title

Identifying Crash Distributions and Prone Locations by Lane Groups at Freeway Diverging Areas

Document Type

Article

Publication Date

1-2011

Digital Object Identifier (DOI)

https://doi.org/10.3141/2237-10

Abstract

Crash distributions and the locations of all crashes and severe crashes at freeway diverging areas were analyzed and compared. Contributing factors to crashes were identified; these included exit ramp types, geometric design features, traffic conditions, and crash-related attributes. Exit ramp types are defined by the number of lanes exiting freeways and AASHTO's lane-balance theory. Four typical exit ramp types were considered: one-lane exit with lane-balanced design (Type 1), one-lane exit with lane-unbalanced design (Type 2), two-lane exit with lane-balanced design (Type 3), and two-lane exit with lane-unbalanced design (Type 4). Lanes are further classified as exit and drop lane group, impact lane group, and interior lane group. Proportionality tests were used to compare the crash distributions on different lanes. The results indicate that lane-balanced designs (Type 1 and Type 3) have a statistically significant higher percentage of severe crashes on the impact lane group than that for lane-unbalanced designs (Type 2 and Type 4). For the interior lane group, Type 4 ramps have a statistically significant higher percentage of severe crashes than the other design types. In addition, ordered probit models were developed for all crashes and severe crashes by one-lane exits and two-lane exits, respectively. Outcomes from the models suggested that more crashes occurred on the exit and drop lane group for the lane-unbalanced designs and more severe crashes occurred on the interior lane group for the lane-unbalanced designs. The study will help engineers have a better understanding of crash distributions and locations for different exit types and help them to develop effective countermeasures.

Was this content written or created while at USF?

Yes

Citation / Publisher Attribution

Transportation Research Record, v. 2237, issue 1, p. 88-97

Share

COinS