Nomogram for Predicting Asphalt Pavement Roughness After Preventive Maintenance Based on Long-Term Pavement Performance Data

Relative effectiveness has been used as the primary criterion when choosing pavement preventive maintenance (PM) treatments; however, the most effective treatment is not necessarily the most cost-effective treatment because of the higher cost. Cost-effectiveness should be determined by predictive models of PM-treated pavement performance, which has rarely been investigated. In this study, a predictive model for post-treatment pavement roughness—as defined by international roughness index (IRI)—was established utilizing the data from the Long-Term Pavement Performance Program (LTPP) Specific Pavement Study 3 (SPS-3). The generalized least squares (GLS) model was employed to improve the predictive performance by exploiting the characteristics of LTPP panel data. A nomogram was also provided to help highway agencies manually obtain the predicted post-treatment IRI values. Results show that post-treatment IRI was significantly higher in dry and non-freeze areas than in other climate areas. The effect of pavement structure on post-treatment IRI was time-dependent; it was insignificant at the beginning and gradually increased after 4 years. Although post-treatment IRI was affected by pavement structures and climate, the relative effectiveness of different PM treatments was only related to the pre-treatment IRI. Thin overlay significantly improved the pavement IRI, and when pre-treatment IRI was 2.0 m/km, the post-treatment IRI of the thin overlay would be reduced to 0.6 times that of the control. However, there was no significant difference in pavement IRI between different seal treatments.