<![CDATA[ The increasing complexity of modern engineering systems and the demand for efficient, cost-effective, and high-performance designs have driven the adoption of intelligent computational strategies in mechanical engineering and physical sciences. Traditional simulation and optimization techniques often struggle with nonlinear, multi-objective problems that span both material and structural design spaces. This study aims to develop a unified smart algorithmic framework capable of optimizing both mechanical systems and material properties concurrently. The research focuses on integrating data-driven models with physics-informed techniques to improve predictive accuracy, computational efficiency, and practical applicability. The proposed framework combines artificial neural networks (ANNs), physics-informed neural networks (PINNs), genetic algorithms (GAs), and Bayesian optimization to form a hybrid multi-objective optimization system. A case study on an electric vehicle (EV) suspension system is used to validate the approach. Surrogate models were trained on finite element analysis (FEA) data and applied within a Pareto optimization loop to explore trade-offs among mass, fatigue life, and material cost. The framework achieved a 27% reduction in structural mass, a 35% increase in fatigue life, and a 13% decrease in material cost. Surrogate models attained R² values exceeding 0.90, with validation showing less than 5% deviation from FEA results. Sensitivity analysis confirmed design robustness under input variation. The findings demonstrate the effectiveness of smart algorithms in co-optimizing systems and materials. The proposed framework enhances the speed, accuracy, and physical validity of intelligent engineering design.]]>
