Abstract

This study investigates the integration of the universal discontinuity index (UDi) into digital twin (DT) models of rock masses, presenting a novel quantitative statistical semantic damage model (QSSDM) that aligns with the digitalization advancements of Industry 4.0. Unlike traditional empirical classification systems such as RMR, Q system, GSI, and RMi—which often lack a robust theoretical basis and fall short of encompassing all engineering designs—UDi offers an objective, comprehensive assessment grounded in probabilistic fracture mechanics and damage theory. The UDi framework classifies rock deformation processes into semantic categories based on a stress–strain ontology, including elastic, elastic–plastic, peak strength and post peak parts. This study validates the effectiveness of UDi through its application in various case studies, including experimental laboratory settings and numerical field-scale analyses for predicting rock strength under different loading conditions. The results show strong agreement between UDi-based rock strength predictions and those derived from both experimental and numerical analyses. These experiments demonstrate that theUniversal Discontinuity Index (UDi) effectively captures the anisotropic behavior of rock resulting from discontinuities and the stress field, as well as the influence of scale variations, confinement stress, and infill material within the discontinuities. Moreover, the findings underscore the UDi's versatility and adaptability across a wide range of applications. Additionally, UDi and other classification approaches were applied to predict the severity of overbreak along a tunnel at El Teniente mine. There is a notable correlation between the observed overbreak patterns and UDi variations, though this validation is more qualitative in nature. Overall, UDi represents a significant advancement in semantic modeling for geotechnical engineering and mining, facilitating a more coherent and practical application of DT concepts in rock mechanics and supporting the transition towards Mining 4.0. © The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature 2025.