The Impact of Narrow Lane on Safety of the Arterial Roads: Incorporating Akaike Information Criterio

This study investigates the impact of lane width on safety for urban arterial roads. The Generalized Linear Model with Negative Binomial distribution was used to estimate the model, which is selected by Akaike information criterion (AIC). Using lane width with two level of categories (9~10ft, 11~12ft), the research found interaction effects of lane width with average annual daily traffic (AADT), presence of shoulder and on-street parking. This finding suggests that the impact of AADT on safety can be either positive or negative depending on lane width, and the impact of lane width on safety increases where shoulder or on-street parking exists. Therefore, the authors may consider narrowing lanes to provide additional spaces to reduce crash rates or address specific issues such as traffic congestions without significant increases on crashes.