A SAFETY PLANNING TOOL FOR EVALUATING SAFETY IMPACTS OF ROAD INFRASTRUCTURE PROJECTS AT THE LINK AND INTERSECTION LEVEL

摘要

The current traffic safety literature on methods and factors affecting vehicle-vehicle and vehicle5 pedestrian collision frequency is very extensive; however, the integration of collision prediction models (CPMs) at the planning stage to assess the regional safety impacts of roadway infrastructural changes is an area that requires further study. Keeping this in mind, the objectives of this research were two-fold: i) develop a crash-risk prediction tool that combines a macroscopic regional-level traffic assignment model and safety performance functions to evaluate the impact of new road infrastructure and ii) to illustrate its applicability through the evaluation of a major highway project at the intersection and link level. The developed tool automates the collision estimation and mapping for different road network scenarios and road users in the Montreal region. In terms of model specification, different negative binomial (NB) model settings were attempted and selected for the two types of network elements (intersections and links) and the two main types of crashes (vehicle-vehicle and vehicle-pedestrian collisions). After applying the developed tool to a case study, it was found that the model results predict a significant re-distribution of crashes on the network. The cause of the re-distribution is mainly due to the change in the traffic patterns resulting from the opening of the new toll bridge. In addition, the overall impact of this new infrastructure is found to be marginally positive given that the total number of collisions in the overall network slightly decreased. The application also shows the network elements in which safety deteriorates.