This dissertation endeavors to introduce a novel supervised Structural Health Monitoring (SHM) methodology for the detection of damage and the prediction of fatigue life in asphalt concrete materials. Grounded in the principles of S-N (strain-number of cycles until failure) curves, this research addresses the intricate task of proficiently monitoring asphalt pavements through the utilization of sensor data, thus optimizing maintenance schedules. The successful implementation of this methodology holds the promise of significant cost savings, primarily by facilitating timely inspections and judicious resource allocation. The comprehensive approach encompasses an extensive literature review, encompassing topics such as asphalt properties, fatigue life, and damage prediction models. It also involves the strategic deployment of piezoelectric sensors, with a specific emphasis on Lead Zirconate Titanate (PZT), as well as four-point bending (4PB) testing. This methodology is further enriched by the incorporation of supervised machine learning techniques for the precise prediction of strain levels, subsequently utilized in fatigue life prediction and damage modeling.