Application of Random Set Finite Difference Method into a deep urban excavation

ABSTRACT: Engineers have been attempting to design geotechnical structures in such a way to achieve the best performance with the minimum budget. To satisfy this demand, reliability methods have been proposed to increase the performance and serviceability; but, decrease the probability of failure in geotechnical structures by considering parameter uncertainties in design. Among various reliability methods, in the present study, Random Set theory in combination with finite difference method (RS-FDM) was used to investigate its applicability and efficiency in geotechnical practice. The 27-meter-deep urban excavation located in Tehran, Iran was chosen as a case study. The excavation was modeled numerically using FLAC3D software. The crown horizontal displacement and factor of safety were considered as desired performance functions. To find the most influential parameters on the desired performance functions, a sensitivity analysis was conducted for 19 different input parameters. The RS-FDM results fall well within the range of site measurements. The probability of failure was estimated to be 0.00002, which indicates that the excavation is safe against failure.