The use of machine learning (ML) and artificial intelligence (AI) in predictive maintenance for civil engineering structures is investigated in this study. This study uses machine learning (ML) and artificial intelligence (AI) to present a novel approach for predictive maintenance in civil engineering structures. Class imbalances in maintenance applications present intrinsic issues that are addressed by this study through the substantial use of a rigorous machine learning process workflow and techniques, especially in the prediction of unusual failures. Using publicly accessible datasets, the study demonstrates how to statistically analyze telemetry data to reveal descriptive statistics and sensor activity that are essential for making well-informed decisions. While taking particular concerns for component relevance into account, performance evaluations of the Random Forest and Artificial Neural Network models in the validation and test sets show generally excellent results. In addition to offering definitive results, the research highlights the significance of taking into account actual failure predictions and optimizing metrics, acting as a methodological manual for managing various data kinds in predictive maintenance applications. In the end, this study adds to the field of artificial intelligence (AI) and machine learning (ML) applications in civil engineering by providing a viable method for improving maintenance plans using cutting-edge data analytics