Data-Driven Prediction and Optimization of Mechanical Properties and Vibration Damping in Cast Iron–Granite-Epoxy Hybrid Composites

This study presents a framework involving statistical modeling and machine learning to accurately predict and optimize the mechanical and damping properties of hybrid granite–epoxy (G–E) composites reinforced with cast iron (CI) filler particles. Hybrid G–E composite with added cast iron (CI) filler particles enhances stiffness, strength, and vibration damping, offering enhanced performance for vibration-sensitive engineering applications. Unlike conventional approaches, this work simultaneously employs Artificial Neural Networks (ANN) for high-accuracy property prediction and Response Surface Methodology (RSM) for in-depth analysis of factor interactions and optimization. A total of 24 experimental test data sets of varying input factors (granite weight%, epoxy weight%, and CI filler weight%) were utilized to train and test the prediction models using an ANN approach and further analyze the interaction effects using RSM. Mechanical properties, including tensile, compressive, and flexural strength, elastic modulus, density and damping properties measured under various testing conditions, were set as output parameters for prediction. This study analyzed and optimized the performance of the ANN model using Bayesian Regularization and Levenberg-Marquardt algorithms to identify the best performing number of neurons in the hidden layer for achieving the highest prediction accuracy. The proposed ANN framework achieved an exceptional average determination coefficient (R²) exceeding 99%, with Bayesian Regularization demonstrating remarkable stability in the 22-neuron range and minimal variation across all properties. RSM and ANN form a powerful framework for predicting and optimizing hybrid G–E composite properties, enabling efficient design for vibration-critical applications with reduced experimental effort and performance optimization.