Developing correlations for advance prediction of slope factor of safety using linear regression analysis – karachi landslide as a case study

Abstract

This study examines the slope stability analysis of a soil slope to correlate the Factor of Safety (FS) with the slope angle (b). Using linear regression analysis, a study is performed to predict FS for slopes in different conditions, such as saturated, unsaturated, seismic and non-seismic states. A total of 400 analysis are performed. The mechanical properties of slope is taken from the experimental analysis such as Atterberg limits, sieve analysis, compaction and other required tests. The slope angle is changed linearly and FS is calculated to develop correlations. In seismic analysis, the horizontal seismic coefficient is taken as 0.3, which is the maximum value for horizontal load in case of seismic analysis. The coefficient of determination (R2) for all the developed equations is higher than 99%. All the developed correlations are provided in the results section. These correlations are applicable in slope stability projects to find the slope safety in advance. A landslide in Karachi is considered as a case study and these correlations are applied to know about the future stability of the slope, which gives perfect result.