As is known to all, the modeling of vehicles and durability simulation is becoming more accurate and more compatible with physical testing, resulting in shortening of the analysis process, and a lower cost. It would be more advantageous in the future to simulate the full vehicle system before the physical testing. Thus, in the analysis of vehicle durability performance, the need for more precise rigid and flexible vehicle modeling and more precise external loadings acquisition method is increasing.In view of the typical difficulties faced in the vehicle multi-body dynamics (MBD) simulation and in the associated loading extraction, this paper proposes a method to achieve accurate external vehicle loadings by virtual simulation. This method is performed based on the physical testing and compensates for the imperfections in the MBD modeling, thus being able to improve the quality of fatigue life prediction (FLP). Key technologies to implement this approach in this study include finite element analysis (FEA), MBD simulation and virtual iteration technique (VIT).A rigid and flexible multi-body dynamics model with less modes is constructed in order to reduce the computation time and to make a better convergence in the computing of the MBD analysis simultaneously. Integrated with the statistics of physical testing, the multi-channel vehicle loadings can be achieved through virtual iteration (VI). Meanwhile, this iteration process identifies corrections to model parameters, thus helping to get a more precise MBD model. Accounting for the nonlinear effects of some of the vehicle components, by using the corresponding precise vehicle loads to excite the dynamic model, body loading spectrums of vehicle are obtained for subsequent FLP. All these measures have been demonstrated for a commercial vehicle analysis. The result showed that a better acquisition of the loading-time history has been obtained, thus reducing the time and effort used in FLP process.