Advanced reliability analysis of road-slope stability in soft rock geological terrain

Abstract

Most of the national, regional, and local roads in Limpopo Province have been developed through a rugged topography and artificial slopes have been created with loose rocks scattered across the slopes as a results road slope instability is the common challenge. The objective of this research study is to conduct an advanced reliability analysis of road-slope stability in soft rock geological terrain using the national road (N1) and its tributary (R71) as case studies. Limit analysis, limit equilibrium, finite element methods, finite difference methods, machine learning and GIS-based tools have been used for this purpose. Meanwhile, the accuracy classification chart of limit equilibrium methods in homogenous slope and a new method for predicting the stability of slope in multiple faulted slopes were developed. The reproduction of failure evolution of slope instability was also performed, followed by reliability analysis of the slope based on probabilistic analysis. Lastly, an integrated approach to slope stability assessment based on machine learning, geographic information systembased tools and geotechnical methods was presented. To achieve the above, field observations and measurements, structural mapping, limit equilibrium, limit analysis, Monte Carlo simulation, fuzzy inference analysis, and GIS digitization and analysis were performed. Software packages such as SLIDE, FLACslope, Optimum 2G, DIPS, RocLab, and ArcGIS, were used. The accuracy classification chart for Limit Equilibrium Methods LEM), a new method for performing stability analysis in multiple faulted slopes, reproduction of failure evolution of slope was developed. Monte Carlo simulation was established as the most reliable and effective technique to analyze slope stability. The steepness of the slope, rock and soil properties, extreme rainfall and geological features were demonstrated to influence slope instability based on an integrated approach as stated above. From the above-mentioned major findings, it was concluded that the developed accuracy error classification chart of LEMs and the new method of slope stability in multi-faulted slopes are useful. Though the reproduction of failure evolution of slope was successfully achieved, for material to flow for a longer distance, high kinetic energy and more shearing of material are expected to take place during this process. It is recommended that other sophisticated methods be utilized to expand the results.