Citation

Jagannathan R, Gimbel M, Bared JG, Hughes WE, Persaud BN, Lyon C. Transp. Res. Rec. 2006; 1953: 187-200.

Copyright

(Copyright © 2006, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

New Jersey jug handle intersections (NJJI) have been around for the past few decades. The basic design philosophy behind implementing jug handle intersections at suitable locations is to improve traffic operations by the elimination of the left-turn phase on a major road and to improve traffic safety by a reduction of the total number of potential conflict points and specific conflicting maneuvers at the intersection. This study, based on statistical analyses of intersection crash data, investigates the differences between and similarities in safety performance of NJJIs and conventional intersections for a limited sample set of 44 NJJIs and 50 conventional intersections. Results from raw data indicated that conventional intersections tended to have more head-on, left-turn, fatal-plus-injury, and property-damage-only accidents and relatively fewer rear-end accidents than NJJIs. These observations were confirmed by negative binomial crash prediction models that were developed to account for the influence of other causal factors. Models were estimated for total, fatal-plus-injury accidents, rear-end, and sideswipe accidents for both sets of intersections.