A hybrid reliability methodology is proposed for the rock tunnel stability analysis considering the polymorphic nature of uncertainties in input properties. This method combines a fuzzy approach to model epistemic uncertainties in Geological Strength Index, considers transformational uncertainty of empirical models, and stochastic methods to model aleatory uncertain properties. Further, detailed guidelines are proposed for the characterization and fuzzification of transformational uncertainties by using the data collected through an extensive literature review. A modified Convergence-Confinement method is proposed and demonstrated by performing the hybrid analysis of a railway tunnel in Jammu and Kashmir, India. Hybrid Monte-Carlo simulations were performed to estimate p-boxes of response parameters of the tunnel, i.e., plastic zone radius and radial deformation. Further, the results of the hybrid analysis evaluated from the developed methodology were compared with the traditional methods-based results and it was concluded that the proposed methodology is appropriate for the tunnel stability analysis with input parameters having different uncertainty types.
10th European Conference on Numerical Methods in Geotechnical Engineering (NUMGE2023)
5. Probabilistic and inverse analysis