<![CDATA[The rapid advancement of Industry 4.0 has catalyzed the integration of artificial intelligence (AI) into smart manufacturing, with predictive maintenance emerging as a crucial application to reduce downtime and optimize operational efficiency. This study aims to develop and evaluate a deep learning-based predictive maintenance model by leveraging real-time sensor data from a smart factory environment. A convolutional neural network (CNN) architecture was implemented to detect anomalies and predict machinery failures in advance. The dataset, consisting of multivariate time-series signals from industrial sensors, was preprocessed and used to train, validate, and test the model’s predictive performance. Results indicate that the proposed deep learning model achieved a prediction accuracy of 94.6%, outperforming traditional statistical and machine learning methods in both precision and recall. The implementation of this AI-driven system enables proactive maintenance strategies, minimizing production losses and extending equipment lifespan. In conclusion, the research demonstrates the feasibility and effectiveness of deep learning in predictive maintenance applications for smart manufacturing systems and offers a scalable solution adaptable to diverse industrial settings. ]]>
Copyrights © 2025
