This work introduces an innovative computational framework tailored for high-fidelity, part-scale thermo-mechanical simulations in Metal Additive Manufacturing (MAM), with a focus on the Direct Energy Deposition (DED) process. MAM offers exceptional design flexibility and material efficiency by fabricating complex metal parts layer by layer. However, the process is challenged by steep thermal gradients, evolving residual stresses, and intricate material behaviors that can affect the structural integrity and performance of the manufactured component. Accurate predictive modeling is therefore essential for optimizing process parameters and ensuring part reliability. Traditional full-order Finite Element (FE) simulations can capture the detailed physics involved, but their computational cost often renders them impractical for large-scale analyses or iterative design optimizations. To address these limitations, the proposed approach employs a local-global simulation strategy based on a Virtual Domain Approximation (VDA) method. By confining the numerical analysis to the critical Heat-Affected Zone (HAZ), the method significantly reduces computational demands while maintaining high accuracy. A residual-based VDA technique is used to approximate boundary conditions effectively, thereby ensuring a robust coupling between the local and global solution domains. Comparative evaluations against standard FE simulations indicate that the new framework achieves substantial computational speed-ups without compromising the precision of thermal and mechanical predictions [1]. This efficient approach facilitates a high-resolution analysis of MAM-induced phenomena, supporting both process optimization and part qualification. As a result, the framework has the potential to bolster the broader industrial adoption of MAM, especially in high-performance sectors such as aerospace, automotive, and energy where reliability and precision are critical. REFERENCES [1] Moreira, Carlos A., Chiumenti, Michele, Caicedo, Manuel A., Baiges, Joan, and Cervera, Miguel. “High-fidelity part-scale simulations in metal additive manufacturing using a computationally efficient and accurate approach.” Additive Manufacturing, April 2025, ISSN 2214-8604, 104748. Elsevier BV, doi:10.1016/j.addma.2025.104748.