Lecturer

Specializes in computational materials science, machine learning for materials design, and steel hardenability modeling, with expertise in Bayesian optimization and process improvement. Research focuses on applying advanced computational intelligence techniques to predict and optimize steel hardenability and manufacturing processes. Developed and evaluated Feed-forward Neural Networks (NN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS) to estimate steel hardenability, determining optimal model complexity for improved generalization and proposing strategies for application across various steel grades. Integrates statistical design of experiments with hybrid Al methods such as Response Surface Methodology (RSM), Artificial Neural Networks (ANN), and genetic algorithms (GA) to optimize CNC machining parameters for high-strength steels. Current work employs state-of-the-art regression models, including Tabular Prior-data Fitted Network (TabPFN) and CatBoost, enhanced with explainable AI (XAI) techniques like SHapley Additive explanations (SHAP) and Shapley Interaction Quantification (SHAP-IQ) to ensure accurate, interpretable predictions. Research findings, published in peer-reviewed journals, demonstrate the ability to bridge materials science and engineering, mechanical engineering, and AI to solve industrial challenges and improve process efficiency.