Calibration of Information-Sensitive Partial Factors for Assessing Earth Slopes

Use of the limit state design with the calibrated load and resistance factors (or partial factors of safety) has a long tradition, especially for structural design codes. The load and resistance factors are calibrated using statistics, reliability, probabilistic analyses and selected target safety levels. To take advantage of the reliability-based design approach, to achieve a greater consistency in the safety level for designed or assessed earth slopes, and to cope with the degree of uncertainty in soil properties, in this study, calibration of the information-sensitive partial factors is carried out. The calibration is based on the first-order reliability method, and considers that the critical slip surface for a given set of soil properties and geometric variables of slope can be estimated based on the generalized method of slices. The calibrated factors depend on the degree of uncertainty in the soil properties (i.e., coefficients of variation of cohesion and friction angle), and on the selected target reliability levels. Results of calibration are used to develop empirical equations for estimating the partial factors that are to be used for slope stability analysis and to assess the adequacy of slope for a selected target safety level. It is hoped that the developed relations could be used to aid the development of reliability-consistent design and checking of earth slopes, and to promote the practical application of the limit state design in geotechnical engineering.