Explainable AI Framework Enhances Materials Engineering Predictions

Researchers presented a high‑fidelity explainable AI (XAI) framework for non‑linear regression in engineering materials. The study compared four model architectures—Linear Regression, Support Vector Regression, Gradient Boosting, and Random Forest—using cross‑validation and SHapley Additive exPlanations (SHAP) to identify key factors influencing material performance. Results highlighted a sharp sensitivity of fracture energy at –40 °C, demonstrating that the XAI approach can provide accurate predictions while offering physical interpretability. The framework follows Green AI principles, reducing the need for extensive laboratory testing and non‑recyclable materials.

The work underscores how explainable machine‑learning models can support safety‑critical material design by balancing predictive accuracy with insight into underlying mechanisms.