To address the insufficient consideration of spatial variability of rockfill rheological model parameters in stochastic finite element analysis, taking an actual concrete-faced rockfill dam (CFRD) project as an example, a deterministic finite element model considering the rheological effects of rockfill materials was established. The Cholesky covariance decomposition method was employed to generate random fields of rheological model parameters. Through the integrated interaction of ABAQUS, MATLAB, and Python, a non-intrusive stochastic finite element calculation accounting for the spatial variability of material parameters was achieved. The Monte Carlo method was then used to perform statistical analyses of the stress and deformation responses of the CFRD. The results show that, compared with the finite element analysis based on a single set of deterministic material parameters, the stochastic finite element analysis considering rheological properties yields more conservative stress and deformation estimates. The differences in the extreme values of various indicators range from 2% to 6%, and the exceedance probabilities for all indicators are all greater than 50%. The proposed stochastic finite element method incorporating the spatial variability of rockfill rheological model parameters is more consistent with actual engineering conditions and can improve the reliability of deformation predictions for CFRDs.