Reliability Based Design for Slope Stabilization Using Drilled Shafts and Anchors

by Lin Li

Institution: University of Akron
Department: Civil Engineering
Degree: PhD
Year: 2014
Keywords: Civil Engineering; Engineering; Geotechnology; slope stability; Monte Carlo; drilled shaft; anchor; reliability; optimization design; deep foundation; landslide; multiple rows
Record ID: 2038402
Full text PDF: http://rave.ohiolink.edu/etdc/view?acc_num=akron1415117410


Landslides and slope failures occur frequently every year to have major impact on the operational safety of roadways and to add financial burden to the highway agencies for slope repairs and maintenance. In this dissertation, a reliability-based computational algorithm is developed for design of a row of equally spaced drilled shafts and/or anchors to stabilize an unstable slope while achieving the required target reliability index with minimum volume of drilled shafts. The Monte Carlo simulation (MCS) technique is used in the previously developed deterministic computational program, in which the limiting equilibrium method of slices is modified to incorporate the arching effects of the drilled shafts in a slope. Uncertainties of soil parameters in the slope are considered by statistical descriptors, including mean, c.o.v., and distribution function. Model errors of the semi-empirical predictive equation for the load transfer factor for characterizing the soil arching effects are considered by statistics of bias. A PC-based program has been developed based on the above methodology. In order to dealing with small probability events in the drilled shaft/slope system and reduce the large number of MCS calculations, a more advanced methodology, importance sampling technique (IST), is proposed to determine the probability of failure and the reliability index of a drilled shaft/slope system. The performance function and the design point are determined by the ordinary method of slices (OMS) with the accompanying load transfer factor. To permit system reliability analysis for an anchor/slope system considering the effects of stochastic corrosion, the Monte Carlo simulation technique is used in conjunction with the modified limiting equilibrium method of slices. Meanwhile, the time-dependent deterioration of bond capacity of corroding soil anchors is developed in this methodology due to the attack of chlorides. The importance of using a system reliability–based computational method is illustrated by showing that the anchor/slope system may fail unpredictably during design life – even though the initial design results have a high degree of safety – if uncertainties of soil parameters, anchor force, and the time-dependent stochastic corrosion process are not accounted for systematically. In order to preventing a large earth thrust applied to the drilled shafts due to the dimensions of the failed slope, the use of multiple rows of drilled shafts could be a feasible solution to both ensure that the global factor of safety of the stabilized slope meets the target factor of safety and the amount of reinforcement used in the drilled shafts is constructible and economical. A limiting equilibrium based methodology, incorporating the method of slices and arching effects of the drilled shafts, is developed for optimizing the use of multiple rows of drilled shafts. It is shown that as compared to one row of drilled shafts, multiple rows of drilled shafts can effectively increase the global factor of safety and at the same time reduce the…