Integrating uncertainty into geotechnical design of underground openings in tectonically undisturbed but lithologically varied sedimentary environments. (July 2021)

• Record Type: Journal Article
• Authors: Paraskevopoulou, Chrysothemis; Skolidis, Antonios; Parsons, Samuel; Marinos, Vassilis
• Is Part Of: Tunnelling and underground space technology. Volume 113(2021)
• Publication Date: 2021-07
• Subjects: Uncertainty in tunneling -- Probabilistic analysis -- Molasse -- Heterogeneous formations -- Lithological unit types -- Rock mass types -- Rock mass behaviour types; Tunneling -- Periodicals; Underground construction -- Periodicals; Tunnels -- Periodicals; Underground areas -- Periodicals; 624.193
• DOI: 10.1016/j.tust.2021.103979
• Languages: English
• ISSNs: 0886-7798

Abstract

Highlights: Provides example for probabilistic analysis in geotechnical engineering. Provides detailed analysis (step-by-step) to reduce uncertainty from geological to geotechnical models. Numerical investigation of Tunnel behaviours based on predefines rock mass behaviour and anticipated failure modes. Abstract: A systematic but straightforward probabilistic approach was adopted for the preliminary design using a tunnel case in Egnatia Odos Highway. This tunnel was excavated in a heterogenous molassic formation in Northern Greece, and more specifically in tectonically undisturbed but lithologically varied sedimentary series. This research aimed to identify sources of uncertainty and evaluate its governance during the selection of the geotechnical design parameters. Rigorous analyses of the critical parameters and factors were undertaken following a standardised method for designing underground structures based on the probability theory. Initially, the statistical analysis of the intact rock strength parameters and project-specific geological and geotechnical factors allowed the development of understanding in determining these sources of uncertainty and assisted in quantifying them. The lithological units encountered in the tunnel alignment were classified into first, Geological Units and, secondly, into Rock Mass Types (Geotechnical Units). Finally, the latter were grouped into anticipated Rock Mass Behaviour Types, which were numerically investigated using RS2 for the Finite Element Method analysis and Unwedge to assess these Rock Mass Types' failure modes. For this purpose, the Point Estimate Method was used in the FEM analysis to evaluate the effect of uncertainty in the strength parameters. Ultimately, the numerical analysis results subsequently led to the preliminary design of the respective support sections. Although this investigation was limited by the geotechnical information provided, it was concluded that the approach followed proved to be adequate in identifying the primary sources of uncertainty and its impact on geotechnical design. It was also noted that with more detailed geological information, this probabilistic approach could be used to produce a robust final support tunnel design.