A Hybrid Gaussian Process-Extreme Gradient Boosting Machine Learning Framework for Robust Knockdown Factor Prediction and Reliability-Based Design of Thin Spherical Shell Caps
The study developed a hybrid Gaussian Process–Extreme Gradient Boosting Machine Learning model to predict knockdown factors (KDF) for thin spherical shell caps, addressing conservatism in current design methods. The hybrid model outperformed other machine learning techniques by achieving the highest accuracy and providing reliable uncertainty estimates, leading to more accurate and less conservative reliability-based designs.