Introduction <p>Efficient finite element simulations are crucial for tackling the computational demands of large-scale models and high-frequency ranges in noise, vibration, and harshness (NVH) modeling.</p> Objective <p>This paper introduces novel methods for compressing reduced impedance matrices (RIM) to optimize computational resources while preserving simulation accuracy.</p> Methods <p>The proposed methodology employs diverse reduction techniques, including Singular Value Decomposition (SVD), pellicular modes, and thin shell modes, to identify relevant vibration modes and optimize modal projection of impedance matrices. By eliminating less relevant modes or generating pertinent modal information of the coupling surfaces, the method aims to significantly compress RIMs without sacrificing accuracy.</p> Results <p>A comparative analysis on a car model demonstrates the effectiveness of the compression technique. The numerical results with the developed approach show significant reduction in storage and computational time, highlighting the method's ability to reduce complex NVH models with precision.</p> Conclusions <p>This proposed approach presents a promising solution to the computational challenges inherent in NVH modeling, offering opportunities for enhanced efficiency and reliability in NVH simulations. With its capacity to reduce computational time and storage requirements, the method is well-suited for large-scale NVH models across distinct engineering industries.</p>