Modeling Probability-based Pavement Deterioration Using PASER Data

Abstract: Pavement deterioration models (PDMs) are essential elements of pavement management systems (PMS), facilitating informed decision-making on maintenance and rehabilitation strategies for road networks. This study aimed to develop a probabilistic PDM that accounts for uncertainties related to various factors influencing pavement deterioration. By utilizing the distribution of pavement condition data from the Pavement Surface Evaluation and Rating (PASER) system, the model effectively captured the inherent variability in pavement performance. The resulting model demonstrated a robust coefficient of determination (R² = 0.90), indicating its high accuracy in predicting pavement deterioration trends over time. This innovative approach to constructing probabilistic PDMs suggests significant potential for broader application across different contexts. By providing a more realistic representation of how pavements deteriorate, these models can enhance the effectiveness of PMS, allowing for better resource allocation and more strategic planning in maintenance and rehabilitation efforts. The ability to incorporate uncertainties into PDMs not only improves the accuracy of predictions but also aids in risk assessment, enabling managers to make more reliable long-term decisions regarding road maintenance. As pavement networks face increasing pressures from traffic loads and environmental factors, adopting probabilistic models becomes increasingly critical for sustaining infrastructure quality and safety. This study contributes to the existing body of knowledge by offering a validated method for developing probabilistic PDMs, which can be applied to various pavement types and conditions. Overall, the advancement of such models represents a significant step toward improving the management of road networks, ultimately benefiting both infrastructure sustainability and user safety.