Considering the innovations in the automotive industry with the implementation of EV technology, building high fidelity models is the key component in the success of the vehicle for NVH performance. Building an accurate CAE model with complex subsystems that represent the physical model is the crucial step in the evolution of the vehicle simulations. One of the reasons for the discrepancies between the finite element NVH model and the physical model is due to the linearization of the systems. This is due to difficulties in determining accurately material properties like damping in rubber mounts. Due to this the CAE models will not represent its counterpart accurately. One approach to solve this issue is to build a physics-based model, having parameters which will control the overall behavior of the system. The engineers must fine-tune these few critical modal parameters of the component of the complex subsystems with respect to certain load conditions. The responses for the full vehicle model are compared against the targets that define the expected behavior of the model and parameters are adjusted to achieve correlation. This entire problem is manual and time-consuming process and mostly depends on the expertise of the engineer. In this paper, we attempt to formulate the more systematic approach to the whole model correlation process. The methodology is demonstrated with a case study employed on a reduced FRF assembly based full system vehicle model built and managed through BETA CAE NVH Console. Use simulated annealing optimization algorithm and modal perturbation as a novel technique is showcased in this process. This technique can be used to make a full vehicle CAE model a digital twin to the physical model.