Abstract: Reliability analysis of slope stability is essential for improving the rationality of geotechnical design. Using a deep excavation as the study case, a three-dimensional numerical model was established based on its engineering geology, hydrogeological conditions, and excavation-support scheme. The finite-difference strength-reduction method was employed to calculate the slope stability factor. A total of 1 000 sets of soil parameter combinations were generated through Monte Carlo simulation. Based on the computed stability factors, system reliability theory was applied to evaluate slope stability reliability, and grey relational analysis was used to assess the sensitivity of soil parameters. The results indicate that the mean stability factor of the foundation-pit slope is F_s=1.249, with a failure probability of P_f=1.2%. Natural unit weight and cohesion are the relatively dominant factors influencing the stability factor, whereas internal friction angle and cohesion are the primary factors affecting slope displacement. Comparisons with field-measured data were conducted to examine differences in displacement and their underlying causes. This study provides a useful reference for reliability-based slope stability evaluation and support-system design for deep excavations.