Comparative Assessment of Pavement Performance Using Analytical Models and Machine Learning under Traffic Loading
Comparative Assessment of Pavement Performance Using Analytical Models and Machine Learning under Traffic Loading
Kamal Omid1
1) M.Sc. in Urban Planning, Municipality of Tabriz Metropolis, Tabriz, Iran,
IOI Code
:
AACH-AAHEDD
Publication :
7th.International Conference & 8th.National Conference on Civil Engineering, Architecture, Art and Urban Design(8iccacs.ir)
Abstract :
Pavement performance under traffic loading is one of the most critical factors affecting the durability, safety, and maintenance costs of transportation networks. Considering the complexity of material behavior and their interactions with traffic loads and environmental conditions, accurate assessment and prediction of pavement performance require the use of advanced analytical models and artificial intelligence techniques. In this study, a comparative assessment approach is proposed to evaluate the performance of asphalt and concrete pavements. Initially, classical analytical models based on material mechanics and traffic load distribution were developed, and subsequently, long-term pavement behavior under real traffic loads was simulated using machine learning algorithms, including neural networks, decision trees, and support vector machines. The results indicate that machine learning models are capable of predicting pavement performance variations with higher accuracy than traditional analytical models and can better capture the effects of complex parameters such as variable traffic and climatic conditions. This study provides a practical framework for improving maintenance planning, reducing life-cycle costs, and enhancing the safety of road
Keywords : pavement performance analytical models machine learning traffic loading service life prediction