Simulation of Landslides after Slope Failure Using the Material Point Method in Shaking Table Tests

The seismic safety of slopes is generally estimated through stability analysis using a conservative approach such as the Fellenius method. However, it is also important in terms of the seismic safety of slopes to estimate the displacement of landslide masses. Accordingly, a series of shaking table tests for slope models with a lower-strength cohesive sand layer were carried out in consideration of the effect of the slope gradient and the width of the sand layer on landslide conditions after slope failure. The results indicated that the behavior of landslide masses depended on the type of slope model used in the test, i.e., the models exhibited ductile failure (which can be simulated using a simple analysis method such as Newmark's sliding block approach) and brittle failure. Additionally, simulation using the material point method (a particle-in-cell technique that can deal with large deformation of soil using simple constitutive laws) for the test was also carried out. The outcome showed that the same trend of landslide mass behavior according to the type of slope model as that seen the shaking table test results can be obtained using the material point method with a simple elasto-plastic Drucker-Prager model for the constitutive law in consideration of the properties of the weak layer in the slope model.