Asphalt pavement in urban streets plays a key role in improving safety, comfort, and the quality of transportation services. However, its frequent deterioration in densely populated cities—such as Karaj—results in considerable economic and social costs. This study aims to identify the factors influencing asphalt pavement deterioration and to propose solutions for enhancing its durability and performance. Relevant data on pavement age, traffic volume, the proportion of heavy vehicles, climatic conditions, material quality, and maintenance status were collected, and the level of distress was classified into three categories: low, moderate, and high.To analyze the data and predict the distress level, the Random Forest machine learning algorithm was employed. The results indicated that the model achieved an accuracy of approximately 96.7%, and the most influential factors affecting pavement deterioration were, respectively, material quality, the percentage of heavy vehicles, pavement age, and daily traffic volume, while climatic variables such as temperature and precipitation had less impact.

The confusion matrix analysis also showed the model’s strong ability to identify moderate and severe distress levels, with minor misclassification occurring mainly in predicting low distress. Based on the findings, rapid asphalt deterioration is attributed to a combination of traffic-related, construction-related, and maintenance-related factors. Recommended strategies include using higher-quality materials, increasing asphalt layer thickness, managing heavy-vehicle traffic, implementing preventive maintenance, and utilizing modern additives. These measures can extend pavement service life and reduce maintenance costs.