Distress-Based Performance Prediction of Pavements and the Master Curve Approach

With the continuous deterioration of pavements, highway agencies are examining alternative methods for predicting performance with field distress data collected regularly and stored in the long-term pavement performance (LTPP database). While mechanistic–empirical (M–E) models are typically developed for pavement design, performance prediction models are used in pavement management systems (PMS) to predict future conditions and optimize maintenance and rehabilitation strategies (M&R). This study aimed to develop prediction functions with a new approach based on the concept of master curves, where performance curves for pavements of similar characteristics and environmental conditions are represented by a universal function. The suggested modeling approach used multiple linear regression (MLR) with LTPP data. Asphalt pavements of similar design and material properties and four key distresses, including fatigue cracking (FC), wheel-path longitudinal cracking (WP), rutting (RD), and transverse cracking (TC), were analyzed. The MCs are constructed using a fixed-slope logarithmic form with vertical shift functions linked to mixture and pavement properties, enabling transferability across climates and structural designs. The study findings indicated that the approach provides a very good fit with distress data from two selected states, with R2 values generally above 0.9 and p-values less than 0.05. While other modeling approaches have been proposed for pavement distress predictions, this MC based modeling uses field- distress progression with material property-based vertical shift functions, linking laboratory characterization to field performance. Because pavement distresses are material and section-specific, this modeling enables highway agencies to develop state-specific prediction models while preserving a generalized methodology for similar material, climatic and structural conditions. Benefits of the suggested modeling approach include reduced distress surveys and modeling frequency, and provides improved performance predictions for enhanced planning of M&R strategies and more reliable scheduling. Potentially unique and/or extreme climatic events (such as extreme weather patterns, high-altitude locations), or new nonstandard mixes may require additional calibration.