The Outage Prediction Modeling (OPM) system, developed by UConn EEC, offers pre-storm power outage forecasts in electric distribution systems. The machine learning based OPM is trained offline with historical weather data, then applied in forecast mode using weather forecast products for upcoming storms. During the past decade, EEC has been continually enhancing the OPM by expanding its storm database, refining weather data customization methods, and experimenting with diverse machine learning and AI algorithms. In this project, we seek the Senior Design Team's expertise to enhance outage prediction accuracy, particularly for thunderstorms and/or high-impact weather events. Thunderstorms are among the most common weather phenomena in the Northeast region especially during spring and summer seasons. However, the uncertainties inherent in predicting these convective storms make it challenging to accurately forecast the thunderstorm-related outages. This project aims to improve the predictive analytical methods for forecasting outages caused by thunderstorms. The outages caused by high-impact weather events such as hurricanes and nor'easters tend to be underestimated by OPM. We seek alternative methods to enhance the model's outage prediction capabilities for high-impact events, allowing utilities to better prepare in advance.