Generally, the design approach for tunnel by traditional deterministic approach assumes rock mass as uniform distributed material properties where it ignores natural heterogeneity present in the field. Whereas the rock mass in Himalayan region where often encountered with highly varying ground conditions due to tectonic activities in addition to weathering and seismic conditions. Hence, negligence of these heterogeneity becomes highly unreliable for the infrastructures such as slopes and tunnels. The present study investigates the impact of spatial variability in rock mass properties on tunnel deformation, with a focus on the geologically complex Himalayan region. The study adopts Spectral Representation Method (SRM) to simulate random fields representing spatially varying Geological Strength Index (GSI) and Young's Modulus (E) is presented. Statistical parameters such as mean, standard deviation, Coefficient of Variation (COV), and Scale of Fluctuation (SOF) are used to generate input data. The paper outlines the development of a 2D numerical tunnel model in PLAXIS 2D, incorporating spatial variability through Python-based scripting. The influence of increasing COV and decreasing SOF on tunnel deformation is discussed, with findings indicating higher deformations under greater heterogeneity. A case study on Tunnel 49A from the Udhampur-Srinagar-Baramulla Rail Link (USBRL) project is presented to validate the approach and demonstrate real-world applicability. The study showcases how incorporating spatial variability leads to more realistic deformation predictions and better-informed decisions regarding excavation and support strategies. Finally, the paper emphasizes the need for adopting probabilistic and adaptive modelling practices in tunnel engineering for safer, more efficient, and economically viable tunnel designs.
Spatial variability, Tunnel, Coefficient of Variation, Scale of Fluctuation, PLAXIS
