Statewide Empirical Analysis of Asphalt Pavement Performance and Service Life at Intersections

Abstract

Abstract Intersections are consistently found to be among the most critical locations for pavement deterioration, yet large-scale empirical evidence quantifying their relationship with deterioration remains limited. This research conducts a comprehensive statewide empirical investigation into the influence of intersections on load-related asphalt pavement distress and service life, using data from Georgia’s state-maintained asphalt roads collected between 2017 and 2021. A two-pronged analytical approach was adopted, integrating fractional logit modeling with correlated random effects and Cox proportional hazards survival analysis, to concurrently examine pavement performance outcomes and deterioration rates in the vicinity of intersections. A fractional logit model with a correlated random effects approach was used to assess how intersection presence, signalization, and posted speed limits affect pavement conditions. Following, a Cox proportional hazards model was used to estimate the relative time until segments reached critical pavement condition thresholds. The analysis revealed that intersection segments exhibit 36% higher risk of reaching critical rutting thresholds. Furthermore, signalized intersections were significantly more likely to be in a deteriorated state, exhibiting a 7.6% point reduction in the probability of being in good condition, compared to a 1.8% point reduction for unsignalized intersections. The analysis supported the hypothesis that intersections with a more balanced distribution of traffic across all approaches tended to exhibit faster pavement deterioration, likely due to increased stopping and deceleration events. This study provides a statistically robust, statewide quantification of intersection–deterioration relationships, enabling intersection-specific design and maintenance strategies across diverse operational conditions.