Machine Learning Driven Prediction and GUI Based Optimization of Quasi-Static Mechanical Properties in SLM Fabricated Ti6Al4V Alloy

This work explores the application of machine learning techniques to predict the mechanical properties of Ti6Al4V alloy produced through Selective Laser Melting (SLM). Dataset comprised of 201 results was extracted from published literature, encompassing six key SLM process parameters and three tensile properties: yield strength, ultimate tensile strength and elongation. Several machine learning models, such as Support Vector Regression, Random Forest, K-Nearest Neighbors, Gradient Boosting, Gaussian Process Regression, and Decision Tree were individually applied to predict each mechanical property, however, the predictive accuracy of these models was moderate. In contrast, as Artificial Neural Networks (ANN) was applied, it captured the complex relationships more effectively, achieving R2 scores of up to 0.84 across all properties. To improve model interpretability, SHAP (SHapley Additive exPlanations) analysis was implemented on ANN, offering insights into the relative importance and physical influence of input features, and helping to bridge the gap between data driven prediction and underlying process physics. Subsequently, a graphical user interface (GUI) was established by reverse training the ANN models, allowing researchers and engineers to obtain process parameters based on mechanical properties required. This GUI can be a practical tool for pre-production evaluation, offering substantial benefits for aerospace and biomedical applications where material performance and precision are critical.