Evaluating random set technique for reliability analysis of deep urban excavation using Monte Carlo simulation

Abstract Over the last few years, the importance of non-deterministic techniques in Geotechnical engineering is highlighted in literature. Random set (RS) theory is recently being utilized for reliability analysis of geotechnical problems, however, its application in deep urban excavation is not yet investigated for a real case study. Hence, in this paper, an attempt is made to investigate the feasibility of the RS method in determining the reliability of deep excavations. For this reason, a 27-meter-deep urban excavation in Tehran, Iran is considered as a case study. The excavation is numerically modelled. Subsequently the reliability analysis is performed. Findings indicate that the probability of collapse of excavation is as low as 10−7 which suggests that the project is safe against collapse. The aforementioned conclusion is confirmed when RS results were compared with site observation. Additionally, RS results were evaluated using Monte Carlo (MC) technique. Although further studies for more real case studies are recommended, good agreement between the results of MC and RS methods suggests the feasibility of RS technique for estimating the reliability of deep urban excavations.