This study explores the application of Artificial Neural Networks (ANNs) to predict critical operational parameters of internal combustion engines, which are essential indicators of engine performance. Predicting these parameters is challenging due to the complexity of high-dimensional and noisy data. To overcome this, we developed a novel approach that integrates clustering techniques into ANN training. By preprocessing the data with clustering algorithms, we uncovered hidden patterns that informed the training process. These clustering-based insights were incorporated into the ANN’s cost function, enhancing its accuracy. Using real-world engine data from Wärtsilä, the proposed method significantly improved prediction for critical parameters. Our findings highlight the potential of combining clustering and ANN methodologies to address challenges in predictive modeling for industrial applications.