Effective road infrastructure maintenance is essential to ensure the sustainability and safety of the transportation system, particularly in urban areas. This study focuses on the application of the Surface Distress Index (SDI) in road condition assessment using a regression model to explore various factors affecting road damage. The methodology used included data collection through field surveys on 42 urban road segments with 2,467 observational data points. The collected data comprised information on crack length and width, potholes, and rutting. Multiple linear regression analysis and the exploration of non-linear models were conducted to evaluate the relationship between these variables and road conditions. The results showed that the number of potholes had the strongest correlation with the SDI, followed by crack width and rutting. The logarithmic model proved to be the most efficient in predicting road conditions, with an R2 value of 0.75, an Akaike Information Criterion (AIC) of 22,856.35, and a Bayesian Information Criterion (BIC) of 22,879.59, indicating a balance between simplicity and the ability to explain variance in the data. This study contributes to the development of data-driven road maintenance methodologies, which can be applied in planning road maintenance that is more accurate, efficient, and sustainable.