Interpreting complex, non-linear Interferometric Synthetic Aperture Radar (InSAR) displacement time-series data for infrastructure risk assessment remains a significant geotechnical challenge. This is particularly evident in regions with established road and railway infrastructures, where the primary objective is monitoring the entire network to ensure safety and operational continuity. Because landslide displacement is a highly complex process affected by a combination of internal geological conditions and external triggers, time-series data inherently encode non-linear trends and periodic fluctuations. To address this, a data-driven framework utilizing a sliding-window transformation to engineer temporal-kinematic features is proposed, providing a broader framework for the contextualization of slope stability assessment from a network perspective. This is paired with Principal Component Analysis (PCA) for dimensionality reduction and evaluated across four unsupervised architectures: K-means, Self-Organising Maps (SOMs), Hybrid SOM-K-means, and PCA-K-means. The comparative evaluation reveals that the PCA-K-means pipeline performed best, offering a highly efficient and scalable workflow. The analysis revealed that the optimized PCA-K-means architecture successfully captured 79.20% of the kinematic variance across the first two principal components. Furthermore, it achieved a robust Between-Cluster-to-Total-Sum-of-Squares (BCSS/TSS) ratio of 71.70%, an optimal Silhouette Score of 0.320, and a low Quantisation Error (QE) of 0.90, demonstrating superior spatial separation and geometric accuracy compared to traditional heuristic methods. When cross-validated against static topographic susceptibility models, the dynamic kinematic clusters exhibited a 23% spatial convergence at the polar bounds of risk, successfully grounding the algorithm’s predictions in physically verified geomorphological features. Relying on the statistical volatility of displacements, this optimal model successfully partitioned the data into five distinct geotechnical risk classes, ranging from stable (Class A) to extreme risk (Class E). The results demonstrate that the developed dynamic framework provides a highly reliable, actionable tool for proactive, large-scale slope stability and infrastructure risk assessment.
Slope instabilities pose a significant threat to transportation infrastructure, creating the need for reliable data-driven approaches for displacement forecasting. This study proposes a scalable machine learning framework that integrates Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) displacement time series with a structured sliding-window transformation to predict short-term vertical slope displacements under multi-cadence satellite acquisition schedules. The transformation reformats raw InSAR observations into fixed-length input windows, where the four most recent displacement measurements and their corresponding elapsed time intervals are used to predict the subsequent displacement based on a local Markov-type assumption. An iterative modelling strategy, progressing from baseline to multi-interval configurations on balanced and unbalanced datasets, was implemented using Artificial Neural Networks (ANN) and Multiple Linear Regression (MR). Under a uniform 12-day satellite revisit schedule, ANN and MR achieved comparable predictive performance (RMSE = 2.85, R2 = 0.93). When evaluated on the more realistic multi-cadence datasets (12–144 days), ANN showed a modest performance advantage over MR, with lower RMSE (2.86 vs. 2.98) and slightly higher R2 (0.92 vs. 0.91). These results indicate that the relative performance of the two approaches is influenced by the temporal sampling structure, with ANN providing a modest improvement under heterogeneous temporal cadences. Sensitivity analysis further identified the most recent displacement stages as the dominant predictors, confirming the importance of short-term deformation history. The proposed framework provides an efficient, scalable, and data-driven solution for regional InSAR-based slope displacement forecasting, supporting early detection of deformation trends and risk-informed management of transportation infrastructure, particularly in data-scarce environments.