Safety Assessment of Road Network Using Traffic Engineering Software (TES). Application of Generalized Estimating Equations and Empirical Bayes

This paper describes how identifying sites with potential for safety improvements, network screening is the initial step that is usually taken by many transportation agencies in their safety management programs. However, identifying and conducting detailed engineering studies of candidate improvement sites is very time consuming and very expensive. Since the funds for safety improvements are limited, it is important to spend the resources as effectively as possible. The purpose of network screening is to review the entire roadway network under the jurisdiction of a particular agency and identify and prioritize those sites that have promise as sites for potential safety improvements. This paper details the development of an automated ranking tool using Traffic Engineering Software (TES) for identifying and prioritizing problem intersections and road segments for the Region Municipality of Halton, Canada. The proposed approach uses the Empirical Bayes method and collision prediction models for estimating the potential safety improvement that can be achieved for each intersection and roadway segment. The generalized estimating equations technique with the assumption of negative binomial error distribution was used for development of the collision prediction models.