Method for Probabilistic Assessment of Tunneling-Induced Damage to Surface Structures Considering Soil-Structure Interaction Effects

Abstract

A new method is presented to quantify the uncertainty in early-stage assessment of surface structure damage caused by tunneling-induced ground movements. The method employs the Monte Carlo simulation technique and a numerical model that simulates soil-structure interaction with an elastoplastic interface. The assessment method results in a probabilistic prediction of the level of structural damage. The sources of uncertainty in tunneling-induced ground movements are quantified using field monitoring data, and the uncertainties in soil and structure properties are characterized from published experimental tests and empirical analyses. The proposed assessment method is then demonstrated by two case studies. Variance-based sensitivity analysis and factor mapping analysis are conducted to demonstrate how the method can identify dominant sources of damage assessment uncertainty. For the both case studies, volume loss was found to contribute most to the uncertainty in building damage prediction; the contribution of other parameters varies significantly for different tunneling scenarios. However, the factor mapping analysis shows that the uncertainty of building properties and ground movements are almost equally responsible for the uncertainty in structural damage. More generally, the results lay the foundation for a performance-based design method to evaluate potential structural damage and to plan mitigation procedures in future tunneling projects.