To address the fragmentation among “stability assessment – motion process – protective design” in current landslide research, this study breaks through the conventional single-stage paradigm by proposing an innovative full-life-cycle analysis framework. This framework enables comprehensive analysis covering potential landslide evaluation, post-failure movement simulation, and protective measure design. Based on engineering geological investigations of the Luoboli Village Landslide hazard in Puge County, we developed for the first time an integrated technical system that couples Abaqus finite element method and CoSim discrete element method. This “failure mechanism – movement simulation – risk mitigation” framework was used to conduct a full-life-cycle analysis of the landslide. The results show that: ① Under strong seismic loading （PGA＝1.0g）, the slope stability coefficient rapidly drops to 0.89, indicating an impending failure characterized by a combined rear-tension and front-shear failure mechanism. ② The dynamic landslide simulation reveals a pronounced velocity gradient, with the landslide front reaching 4−6 m/s, 3 times faster than the rear （1−3 m/s）. This gradient causes tensile deformation within the sliding mass and results in a deposit with a maximum thickness of 12 m and an area of 3.65×10⁴ m², posing direct threats to 16 buildings and major transportation routes. ③ Comparative analysis of multiple mitigation strategies indicates that individual approaches, such as slope cutting （60.7% reduction in displacement） or retaining walls （85.7% reduction in displacement）, face performance bottlenecks. In contrast, a combined slope cutting and retaining wall system significantly improves control, achieving a 94.3% reduction in sliding distance and a 99.99% reduction in run-out area. This confirms the scientific validity of an integrated "source-load reduction and path-blocking" strategy. The proposed “full-element simulation – multi-objective optimization” method integrates precise prediction of landslide disaster chains with intelligent decision-making for mitigation. This approach provides a systematic technical pathway for landslide prevention and control in seismically active regions.

The research findings indicate:

1） Under strong seismic loading, the potential risk of landslide instability increases significantly. Numerical simulations indicate that the safety factor of the Luoboli Village Landslide is 1.45 under natural conditions, suggesting a relatively stable state. When subjected to a seismic load of 0.15g, the safety factor decreases to 1.13, approaching the threshold of instability. Under a 1.0g seismic loading, the safety factor further declines to 0.89, indicating a high likelihood of failure during strong ground motion. The failure mode under seismic loading is characterized by a composite mechanism involving tensile cracking at the rear and shear extrusion at the front. These results highlight the urgent need for appropriate mitigation measures to reduce the potential disaster risk.

2） During the initiation stage of landslide movement, a clear velocity stratification phenomenon is observed. The leading edge of the sliding mass exhibits significantly higher velocities than the middle and rear sections, resulting in stretching or deformation of the sliding body. In the early phase of dynamic simulation, the front moves at 4−6 m/s, while the rear lags behind at 1−3 m/s. This phenomenon may be attributed to thrust forces acting on the front from the rear, while tensile forces act on the trailing edge, limiting its motion. Additionally, intense shear at the front facilitates particle detachment, reducing inter-particle contact and friction, and allowing higher acceleration. These findings underscore the necessity of reinforcing the leading edge in landslide mitigation design.

3） Simulation of the post-failure process suggests that, in the event of collapse, the Luoboli Village Landslide could deposit material over an area of approximately 30 065 m², burying 16 buildings and a local road. The maximum deposition thickness may be up to 12 m, implying severe disaster consequences. Among the three mitigation strategies assessed, slope cutting alone results in a 60.7% reduction in sliding distance and an 80.7% reduction in affected area, but additional retaining structures are required. A single retaining wall achieves better performance, reducing sliding distance by 85.7% and affected area by 98.6%, although some residual sliding risk remains. The combined strategy （slope cutting＋retaining wall）, despite higher costs, provides the most effective long-term solution, achieving a 94.3% reduction in sliding distance and a 99.99% reduction in run out area.

Overall, this study demonstrates the feasibility of a full-process approach−spanning failure mechanism analysis, dynamic simulation, and mitigation optimization−based on the Abaqus-CoSim coupling method. To further improve landslide prediction and control, future research should consider the combined effects of rainfall and seismic activity, as well as the influence of complex topography on landslide behavior.