Unveiling Energy Insights: An Explainable AI-Driven Framework for Precision Household Consumption Forecasting

Abstract-Effective energy management is essential for promoting sustainability, reducing carbon emissions, conserving resources, and cutting costs. However, traditional energy forecasting methods often fall short in terms of accuracy, indicating a need for more advanced solutions. Artificial intelligence (AI) has emerged as a valuable tool for energy forecasting, but its lack of transparency and interpretability makes it difficult to understand its predictions. To address this, Explainable AI (XAI) frameworks have been developed to enhance the transparency and interpretability of AI models, particularly those considered “black-box” models. This paper examines household energy consumption predictions by comparing various forecasting models using evaluation metrics such as the Coefficient of Determination (R2), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE). After testing with unseen data, the best-performing model is selected, and its predictions are explained through two XAI techniques: Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP). These methods help identify key factors influencing energy consumption forecasts, such as current consumption patterns and previous energy usage. The study also highlights the importance of XAI in developing predictive models that are both reliable and consistent.