In electric and hybrid electric vehicles, the eccentricity of the hub motor rotor has a significant impact on the vehicle suspension system, as the resulting vibrations are transmitted to the suspension, increasing component fatigue. This study investigates the nonlinear dynamics of a vehicle suspension system with interval parameters for rotor eccentricity under road and motor excitations, quantifying and predicting the effects of interval uncertainties on the system's dynamic responses. A nonlinear dynamic model of the vehicle suspension system is established, with the rotor eccentricity modeled as an interval parameter. An efficient surrogate modeling approach based on polynomial chaos expansion (PCE) is developed to significantly improve the efficiency of interval uncertainty analysis. The accuracy and efficiency of the surrogate model is validated through comparisons with the results of Monte Carlo simulation (MCS). It is found that at the same accuracy level, the number of samples and computation time required by the PCE method are reduced to 1.6% and 20% of those by the MCS method, respectively. Frequency-domain analysis incorporating rotor eccentricity uncertainties demonstrates that the system complies with the design safety threshold across 95% of the sampled parameter space realizations. This finding establishes a theoretical foundation for subsequent control optimization and robustness validation efforts. Finally, a global sensitivity analysis of the output responses reveals that the interaction between the eccentricities of the front and rear rotors has a significant impact on the system's dynamic behavior.