Integrating AI-Driven Deep Learning for Energy-Efficient Smart Buildings in Internet of Thing-based Industry 4.0

Abstract

The integration of Industry 4.0 technologies has paved the way for rapid advancements in smart, energy-efficient buildings. This research focuses on optimizing energy consumption in IoT-enabled infrastructures through the application of data-driven modeling techniques. A comparative analysis is conducted using several machine learning and deep learning models, including Random Forest (RF), Gradient Boosting (GB), Deep Neural Networks (DNN), and Artificial Neural Networks (ANN). These models are trained and validated using real-world datasets, with appropriate preprocessing methods applied to enhance data quality. Evaluation metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and Scatter Index (SI) are used to measure performance. The findings suggest that RF and GB models strike a practical balance between accuracy and computational efficiency, while DNN delivers the highest predictive accuracy but demands significantly more processing power.