Enhancing the sustainability of slope stability in embankment construction by leveraging smart sensors and monitoring systems for data-driven insights

Slopes are highly susceptible to instability during earthquakes, floods, and other natural disasters, often leading to landslides that pose serious risks to human life and property. This study investigates embankment slope stability through smart monitoring and reinforcement using geosynthetics. A novel polymer composite fertilizer (PCF) was characterized for its surface curing performance, resistance to temperature extremes, freeze-thaw aging, wind and water erosion, and its ability to neutralize soil acidity and alkalinity. The results demonstrate that PCF improves loess slope fixation and overall stability through both physical and chemical mechanisms. An early warning system is integrated into the framework to prevent significant property loss and fatalities. Principal Component Analysis (PCA) was employed to pre-process soil settlement data and identify outliers. A novel attention-constrained neural network optimized using the Marine Predator Algorithm (MPA) was proposed to extract key features from complex datasets for seismic slope stability prediction. Compared with ANFIS and EHO-NF models, the proposed model achieved a higher sensitivity of 0.788. Furthermore, the incorporation of biopolymers significantly enhanced resistance to shallow slope failures, offering a sustainable and adaptable solution for improving soil strength and long-term slope stability.