RESUMEN
This article investigates the effects of highway-related correlates such as geometric design features and traffic factors on fatal accident occurrences in small-sized Indian cities, Patiala and Rajpura, by estimating Negative Binomial regression models. Between 2013 and 2015, 61 and 188 fatal accidents were recorded in Rajpura and Patiala, respectively. The study reveals that the number of traffic lanes positively influences accident occurrences in Rajpura. On the other hand, in Patiala, the width of paved road surface, segment length and average daily traffic (ADT) are positively correlated with accident rates. Intercity highways passing through rural areas near the city boundary are more prone to fatal accidents than the highway segments within urban limits. Elasticity analysis suggests that an increase in paved road width and ADT by one percent would increase accident rates by 0.7% and 0.436% in Patiala and Rajpura, respectively. The findings of this research concur with the results of previous studies that infrastructural improvements such as increasing the number of lanes and paved road width are positively correlated with accident rates. It is expected that understanding the association between accident rates and non-behavioral factors would help transportation engineers identify the deficiencies in roadway designs in the study cities.