Multi-Physics Topology Optimization and Machine Learning Surrogates for Lightweight Cast Automotive Parts

This paper proposes an integrated optimization framework for lightweight cast vehicle structures by combining Multi-Physics Topology Optimization (MPTO) with machine learning (ML)-driven surrogate modeling. Conventional structural optimization approaches often rely on simplified load cases—such as 3 g inertial loading, frame torsion, or bending—rather than capturing the true complexity of performance-critical conditions like crashworthiness and fatigue durability. As a result, designs optimized for a single criterion frequently degrade in other performance areas, leading to costly redesigns or over-engineered solutions. The proposed methodology introduces MPTO through four structured steps that progressively build design fidelity. Step 1 establishes a baseline primitive topology optimization workflow using a standardized package checklist, ensuring that the design domain, boundary conditions, constraints, and key manufacturing rules are consistently defined as the foundation for subsequent studies. Step 2 applies MPTO under equivalent static loads (ESLs) derived from transient crash and durability simulations to identify fundamental structural load paths, guaranteeing that multiple disciplines are addressed simultaneously and realistic performance trade-offs are explicitly captured. Step 3 refines the candidate design by introducing optimized bead patterns along the identified load paths, where ML-based surrogate models trained on design of experiments (DOE) samples are used to efficiently predict stress responses and evaluate multiple reinforcement variants with drastically reduced computational cost. Step 4 conducts Pareto-based multi-objective analysis to select balanced solutions that simultaneously meet crashworthiness and durability requirements without compromising casting feasibility. The framework is validated on a front engine mounting bracket of a heavy-duty truck—a safety-critical cast component. The application demonstrates a successful material transition from iron to aluminum, achieving more than 30% weight reduction while maintaining compliance with durability and frontal crash performance targets. The results confirm that integrating MPTO with ML-driven surrogate modeling provides a systematic and scalable methodology for generating robust, lightweight, and casting-feasible designs for next-generation electric and commercial vehicle platforms.