The spatial variability in rock mass mechanical parameters critically affects slope stability assessments. This study investigated the southern slope of the Bayan Obo open-pit mine. A representative elementary volume (REV) with a side length of 14 m was determined through discrete fracture network (DFN) simulations. Based on the rock quality designation (RQD) data from 40 boreholes, a three-dimensional spatial distribution model of the RQD was constructed using Ordinary Kriging interpolation. The RQD values were converted into geological strength index (GSI) values through an empirical correlation, and the generalized Hoek–Brown criterion was applied to develop a spatially heterogeneous equivalent mechanical parameter field. Numerical simulations were performed using FLAC3D, with the slope stability evaluated using the point safety factor (PSF) method. For comparison, three homogeneous benchmark models based on the 5th, 25th, and 50th percentiles produced profile-scale safety factors of 0.96–1.92 and failed to replicate the observed failure geometry. By contrast, the heterogeneous model yielded safety factors of approximately 1.03–1.08 and accurately reproduced the mapped sliding surface. These findings demonstrate that incorporating spatial heterogeneity significantly improves the accuracy of slope stability assessments, providing a robust theoretical basis for targeted monitoring and reinforcement design.

To address the limitations of traditional early warning methods in open-pit slope displacement monitoring—particularly their neglect of spatiotemporal correlations and their difficulty in analyzing multi-scale non-stationary sequences—this study proposes an early warning framework that integrates spatiotemporal clustering with multi-scale decomposition. Taking the southern slope of the Bayan Obo Main Pit as a case study, high-risk deformation zones were identified using DBSCAN-based spatiotemporal clustering applied to slope radar monitoring data. The displacement time series were decomposed using Variational Mode Decomposition (VMD) into trend and periodic components, for which Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM) models were respectively developed. The results indicate that (1) DBSCAN effectively detects clusters characterized by high average cumulative displacement and broad spatial distribution, while filtering out isolated outliers. (2) The trend component prediction achieved a coefficient of determination (R²) of 0.99755, while the periodic component prediction yielded a root mean square error (RMSE) of just 0.0978 mm. The reconstructed total displacement achieved an R² of 0.9973, verifying the proposed multi-scale decomposition and hybrid modeling framework’s high accuracy and robustness in slope deformation modeling and early warning.

Conventional unconditional random-field (URF) models were shown to neglect in‑situ monitoring data and thus to misrepresent real slope stability. To address this, a constrained random-field (CRF) generator was proposed, in which Fast Fourier Transform (FFT) filtering was coupled with co‑Kriging to assimilate site observations. A representative three‑bench slope was adopted, and the failure‑mode distribution and the statistics of the factor of safety (FoS) produced by the URF, the independent random field (IRF), and the CRF were examined across strata dip angles of 15°–75° and two cross‑correlation states ( = −0.2, 0). It was found that eliminating cross‑correlation decreased the mean FoS by 0.006, increased its standard deviation by 10.26 %, and raised the frequency of low‑FoS events from 7.49 % to 12.30 %. When field constraints were imposed through the CRF, the probability of through‑going failure was reduced by 12 %, the mean FoS was increased by 0.01, the standard deviation was reduced by 15.38 %, and low‑FoS events were suppressed to 2.30 %. The CRF framework was thus demonstrated to integrate stochastic analysis with field measurements, enabling more realistic reliability assessment and proactive risk management of slopes.

Conventional unconditional random field (URF) models were shown to neglect in-situ monitoring data and thus misrepresent real slope stability. To address this, a conditional random field (CRF) generator was proposed, in which Fast Fourier Transform (FFT) filtering was coupled with co-Kriging to assimilate site observations. A representative three-bench slope was adopted, and the failure-mode distribution and the statistics of the factor of safety (FoS) produced by the URF, the independent random field (IRF), and the CRF were examined across bedding-dip angles of 15–75° and two cross-correlation states ($\rho$ = −0.2, 0). It was found that eliminating cross-correlation decreased the mean FoS by 0.006, increased its standard deviation by 10.26%, and raised the frequency of low-FoS events from 7.49% to 12.30%. When field constraints were imposed through the CRF, the probability of through-going failure was reduced by 12%, the mean FoS was increased by 0.01, the standard deviation was reduced by 15.38%, and low-FoS events were suppressed to 2.30%. The CRF framework was thus demonstrated to integrate stochastic analysis with field measurements, enabling more realistic reliability assessment and proactive risk management of slopes.