Stochastic Stability and Failure Mechanism of Embankments Supported by Deep Mixing Columns Considering Anisotropic Spatial Variability in Clay Shear Strength

Deep mixing (DM) columns are widely used to enhance soft soil beneath embankments. These columns are subject to vertical compression and horizontal pressure, developing multiple failure mechanisms, including compression, tension, shear, bending, and rotation. Inaccurate failure assessment can compromise embankment stability. This study investigates how anisotropic soil variability affects failure mechanisms and the stability of embankments supported by DM columns. A random finite difference method with anisotropic random field modeling and Monte Carlo simulation is employed. The stochastic stability is analyzed for various undrained shear strengths, area replacement ratios, unconfined compressive strengths of columns, and horizontal and vertical correlation distances of clay. Innovative digital image processing techniques enable precise quantification of failure mechanisms from anisotropic random soil models. The interaction between the failure slip and column failure mechanisms reveals diverse failure patterns and sliding surface dimensions. This study reveals how these patterns are influenced by variations in the column strength parameters and the anisotropic spatial variability of soil. Insights from this analysis elucidate the crucial role of these alterations in failure mechanisms and slip surface sizes, providing new perspectives on their intricate interaction and revealing implications for embankment stability.