There has been an increased use of Machine Learning in predicting the pavement performance of roads. It can help reduce costs in maintaining them. Machine learning has a wide scope in maintaining road quality without requiring a significant investment by predicting the health of roads in advance. Countries like India do not have long term pavement performance data for Tier-2 cities. Moreover, collecting this historic data poses its own challenges. The paper aims to overcome this hurdle so that scientific policy planning can be done by governments. Polynomial regression, SVR- Support Vector Regression, Decision Tree, and Random Forest regression models were chosen based on their established performance in predicting similar road-related variables. The RMSE (root mean square error) values calculated for each model indicated promising predictive accuracy, with the support vector regression model performing the best. The study suggests Support Vector Regression is the best way to measure the international roughness index, giving 0.91 as the explained variance score also referred as R- square value.