Reliability-based robust geotechnical design of braced excavations considering multiple failure modes

Reliability design of braced excavation is still a challenge for geotechnical community. Optimization design is a normal method to control the safety and cost of braced excavations. This study presents an advanced reliability-based robust geotechnical design method, which can consider multiple failures and uncertainty of statistical information. A universal design sample was conducted to verify the necessity of considering the uncertainty of statistical information. Ultimate limit state and serviceability limit state of braced excavations were defined, and point estimating method was used to evaluate the standard deviation of failure probabilities. Two-objective and threeobjective optimization models were developed to illustrate the application of proposed methods in detail. In addition, the performance of optimization algorithms and further application of multiple-objective models were discussed. The results from this study indicate that the proposed method has a good performance in determining the optimal design with reasonable robustness and cost. New algorithms have higher efficiency in solving nonlinear and multiple-objective optimization problems than the 2nd Non-dominated sorting genetic algorithm. This study can guide the design of retaining systems of braced excavations in clay.