Tunneling-induced ground settlements present a major challenge in urban settings, particularly in densely built environments. While traditional approaches like empirical formulas and finite element modeling are widely used to predict tunneling-induced deformations, they often require significant computational effort and are less adaptable to varying project conditions. This study explores the potential of machine learning (ML) techniques to predict ground settlements caused by tunneling activities. A comprehensive dataset, comprising key influencing factors such as tunnel and soil parameters, is utilized to train ML models to identify patterns and relationships that influence settlement outcomes. Algorithms including decision trees (DT), random forests (RF), and gradient boosting (GB) are employed to establish predictive relationships between these parameters and maximum settlements while offering insights into the importance of each factor. Model performance is evaluated using standard metrics like coefficient of determination (R2), root-mean-squared error (RMSE), and mean absolute error (MAE). Feature importance analysis provides insights into the relative influence of each parameter, aiding in geotechnical decision-making. The study highlights the potential of ML as a computationally efficient, adaptable, and accurate tool for assessing tunneling-induced ground settlements, offering a transformative approach for future geotechnical designs. The results emphasize the feasibility of integrating ML into tunneling projects, contributing to safer and more sustainable urban infrastructure development.