Comparison between quantile value method and partial factor method for stability evaluation of soil slopes

Since the coefficients of variation of soil properties vary in a large range, the real reliability index for the soil slope designed using the partial factor method also fluctuates in a range. This drawback limits the application of PFM in stability evaluation for soil slopes. The quantile value method is a new method and adopts a single quantile value (η) as the design factor in slope reliability design. The quantile value method is introduced to analyze the reliability of soil slopes. In the analysis, both the quantile value (η) and the partial factors（c,f）are calibrated and validated. The results demonstrate that: (1) the quantile value (η) is more robust and stable comparing to the partial factors（c,f）when the fluctuations of the mean values and coefficients of variation of shear strength parameters of soil, the angle and height of slope are considered. (2) The quantile value method it is more conservative and safer than the partial factor method. (3) The quantile value method involves only one single value for all variables and is simpler than the partial factor method including many different partial factors for variables.