ABSTRACT: Probabilistic slope analysis of embankment on soft clay.

CGRG Bibliography of Canadian Geomorphology
Search Results

Author :

Date :

Title :

Publication :

Issue :

Page(s) :

Abstract

Despite decades of accumulated experience, the ability of the geotechnical profession to make reliable forecasts of slope failures remains poor. This is largely due to the numerous sources of uncertainty that dominate performance projections in geotechnical engineering. Inherent natural variability of soil properties, lack of representative data, and limitations of models and correlations used in the analyses are some of the uncertainties that geotechnical engineers are faced with daily. Probabilistic techniques are better tools to quantify and incorporate uncertainty into slope analysis and design than the conventional approach based on the deterministic factor of safety.A probabilistic slope analysis methodology based on Monte Carlo simulation using the Excel and @Risk software is applied to evaluate the stability of the Muar trial embankment constructed in Malaysia in 1988. The embankment was built to failure on top of a stiff clay crust underlain by a thick deposit of soft marine clay. The strengths of both clay deposits exhibit linear trends with depth. At failure, the thickness of the placed fill was 5.4 m and the average settlement of the embankment was about 0.70 m. Construction time up to failure was 100 days and failure occurred rapidly, one day after the development of a longitudinal crack along the centreline of the embankment.Starting with field and laboratory data, the paper demonstrates the techniques used to quantify uncertainties in shear strengths of embankment fill material, clay crust, and underlying soft clay. The analysis accounts for the uncertainty due to the spatial variability of soil deposits, the statistical uncertainty in the linear trends of undrained shear strengths (because of the limited data available), and the bias in Bjerrum’s vane correction factor. Probabilistic and deterministic slope assessments are conducted to estimate the probability of unsatisfactory performance (or failure probability) for the maximum embankment height, and the variation in factor of safety and probability of unsatisfactory performance at different stages during construction. The probability of unsatisfactory performance and factor of safety associated with the maximum embankment height at failure are estimated to be 0.24 and 1.11, indicating that failure was expected. For an embankment height of 3.3 m, which would have been deemed stable based on conventional slope practice, the probability of unsatisfactory performance and the factor of safety are 0.014 and 1.42, respectively. The results of the two analyses combined with those of three case histories previously published by the authors shed some light on the meaning of computed probability figures in relation to observed field performances. This process of calibrating slope performances in terms of probabilities is of paramount importance in establishing a consistent probabilistic design criterion for slopes.A sensitivity analysis based on Spearman rank correlation coefficients indicates that the impact of uncertainty in Bjerrum’s vane correction factor on the reliability of the calculated factor of safety is significant, and might be even larger than that of the spatial variability of the undrained shear strength. It also shows that the contribution of fill strength to stability is a major source of uncertainty in the design of embankments on soft soils.

Bibliography of Canadian Geomorphology