<p>Accurate prediction of industrial energy consumption plays a crucial role in improving manufacturing sustainability and optimizing resource utilization. This study proposes a hybrid deep learning framework that combines a Convolutional Neural Network and a Multilayer Perceptron to predict energy consumption in manufacturing systems. The proposed model integrates Binary Particle Swarm Optimization for feature selection and a Genetic Algorithm for model optimization to enhance prediction accuracy and reduce redundant input variables. The performance of the proposed approach is assessed on a real-world smart manufacturing problem with 10,000 instances containing various operational features as well as material-related features. An optimization-based feature selection technique is used to select the most relevant features to be included in the model. This improves the overall robustness of the model. The proposed model is evaluated on the selected features. The data set is divided into a training set, a validation set, and a testing set using a 70%–15%–15% distribution. Experimental results show the proposed optimized model outperforms the existing deep learning models. The proposed model is compared with the Convolutional Neural Network model, Multilayer Perceptron model, Long Short-Term Memory model, Gated Recurrent Unit model, and Bidirectional Long Short-Term Memory model. The proposed model shows better performance compared to the existing models. The proposed model shows a Mean Squared Error of 0.0097, Mean Absolute Percentage Error of 0.0126, Mean Absolute Error of 0.0780, Median Absolute Error of 0.0647, and a coefficient of determination of 99.82%. The proposed framework offers a reliable and computationally efficient approach for intelligent manufacturing systems and supports data-driven decision-making toward sustainable industrial operations.</p>