A Comprehensive Python Package for Forecasting Power Outages

Abstract

Michigan has the second-highest rate of weather-related power outages, posing significant public health risks and resulting in millions of dollars in economic losses. While several studies have employed forecasting methods, including machine learning and artificial neural networks no comprehensive libraries exist to streamline data engineering, correlation testing, and forecasting. This study aims to create a fully integrated library in Python facilitating its integration into diverse applications.