Simplified reliability assessment approach for tunnel structures considering the effects of adjacent excavation and soil uncertainty

Reliability assessment of tunnel structures is a fundamental concern in civil engineering, as tunnels are inevitably affected by adjacent excavation projects, resulting to additional deformations. The complexity of this subject is further amplified when considering the soil uncertainty. In this study, a direct Monte-Carlo simulation (DMCS) approach is proposed to evaluate the reliability of tunnels structures after disturbed by adjacent excavation. Random variables are employed to capture the uncertainty in soil mechanics parameters, while the tunnel is modeled as a Timoshenko beam placed on a Pasternak foundation, allowing for the consideration of generalized deflections and shearing deformation, as well as the bending deformation. A comparison between DMCS, Response Surface method (RSM) and First-order Second-moment (FOSM) method is performed through a typical engineering example. The results demonstrate that DMCS accurately captures the reliability of tunnel structures and requires less computational effort, as it eliminates the need for re-establishing response surface functions through numerical simulations. Further parametric analysis investigates the influence of coefficient of variation (COV), cross-correlation, and probability distribution types of soil parameters on reliability of tunnel, which indicated that the COV and probability distribution types had a significant influence, while the impact of cross correlation was limited. The reliability of tunnel structures is more sensitive to the variability of elastic modulus. Additionally, the reliability results obtained when assuming soil parameters follow a normal distribution are lower than those obtained when considering them as log-normal distributions. The proposed method offers an accurate and practical tool for evaluating the reliability of operational tunnel structures in the presence of adjacent excavations.

Automated summary

This study proposes a direct Monte-Carlo simulation approach to assess the reliability of tunnel structures after being affected by adjacent excavations. Parametric analysis shows that the variability of soil parameters, the cross-correlation, and the probability distribution types have significant effects on tunnel reliability, with the variability of elastic modulus being the most influential factor.