Comparisons of IRI-Based Pavement Deterioration Prediction Models Using New Mexico Pavement Data

The paper investigates the applicability of IRI-based pavement deterioration prediction models, which include four deterministic models for pavement performance prediction (i.e., the NCHRP model, Al-Omari-Darter model, Dubai model, and the NMDOT model used by the New Mexico Department of Transportation). Comparisons of these models are made using the data from both the NMDOT Pavement Management System (PMS) and the Long-Term Pavement Performance (LTPP) sites in New Mexico. It is found that the NCHRP and the Dubai models are relatively good tools for pavement performance prediction in New Mexico, while the Al-Omari-Darter model and the NMDOT model do not provide a sound statistical basis. In addition, one probabilistic model for pavement service life prediction, i.e., the survival curve, is also presented. The service life estimated by survival curve is compared with those determined from two approaches — pavement age and traffic loading. The result of traffic loading approach is found to be more reasonable.