Citation

Belanger C. Transp. Res. Rec. 1994; 1467: 23-29.

Copyright

(Copyright © 1994, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

In this study, empirical Bayesian methods were applied to the estimation of the safety of four-legged unsignalized intersections. This application can be described as a two-step process. First, multivariate models were developed to estimate the number of accidents from various flow functions at these intersections. The best model was obtained from the product of major and minor flows, raised to a power. Attempts were made to develop models for specific patterns of collisions and to incorporate variables other than traffic flow functions to these models. The modeling results were then combined with the accident count of a four-legged unsignalized intersection to estimate its safety.

RESULTS were used to identify blackspot locations and to evaluate the effects of interventions more accurately.

Language: en

Keywords

Accident prevention; Highway accidents; Intersections; Mathematical models; Regression analysis; Highway traffic control; Estimation; Statistical methods