Reliability and parameter sensitivity analysis on geosynthetic-reinforced slope with considering spatially variability of soil properties

Abstract

The Bishop method is often adopted to evaluate the stability of slope, but this method can’t effectively consider the inherent variation characteristics of soil. This study focuses on the reliability of geosynthetic-reinforced slope considering the spatial variability of soil properties. The improved Bishop method is adopted to evaluated stability of slope. The Karhunen-Loève expansion is adopted to generated 2-D random fields of spatial variability of soil properties. The Monte Carlo simulation combined with Latin hypercube sampling is adopted to analyze the reliability of geosynthetic-reinforced slope considering the spatial variability of soil properties. As a result, the spatial variability of soil shear strength has an important impact on the reliability of reinforced slope. The reliability of reinforced slope is more sensitive for variation of cohesion (COVc) compared with coefficient of variation of internal friction angle (COVφ). Considering slope’s spatial variability of cohesion (c) and internal friction angle (φ), the reinforced slope’s reliability is greater than that considering the spatial variability of φ or c. The reliability is sensitive for the autocorrelation distance especially the vertical direction. The correlation coefficient (r (c, φ)) has also a great influence on the reinforced slope’s reliability. The study may help provide more scientific help for the design of reinforced slope and make the construction of the project running safely and smoothly.