Reliability-Based Optimization Design for Drilled Shafts/Slope System-Article

Article

Pdf Summary

The paper presents a reliability-based method for designing rows of drilled shafts to stabilize unstable slopes at minimum cost while meeting a target reliability index. It extends a limit-equilibrium method of slices by incorporating soil arching: shafts transfer part of the soil’s driving force, represented by a load-transfer factor. Because soil properties and the semi-empirical load-transfer model are uncertain, the authors use Monte Carlo simulation with lognormally distributed soil cohesion, friction angle, and unit weight, and account for bias in the load-transfer factor. These methods are implemented in the P-UASLOPE program.<br /><br />The method is demonstrated using the ATH-124 slope in Ohio, with a target reliability index of 3. The design procedure tests shaft locations, diameters, and spacing, then considers shaft forces, structural performance, constructability, and cost. The selected economical option uses 0.6 m diameter shafts, spacing-to-diameter ratio of 2, and a location 30.5 m from the slope reference point; it meets the reliability and deflection requirements.<br /><br />The results show that factor of safety alone does not reliably represent slope safety: different safety factors can correspond to similar reliability, and identical factors can yield different reliability. The reliability-based approach explicitly accounts for uncertainty and supports more informed design optimization. Sensitivity analysis indicates that variability in friction angle has the greatest influence on failure probability, making its characterization especially important.

Keywords

reliability-based design

drilled shafts

slope stabilization

limit equilibrium

soil arching

Monte Carlo simulation

failure probability

factor of safety

friction angle variability

P-UASLOPE