Abstract
Based on the identification and analysis of safety risks in TBM tunnel construction, we construct a comprehensive safety risk dynamic evaluation index system tailored to this context. To account for the dynamic nature of safety risks during TBM tunnel construction, variable weight theory is incorporated, allowing for a more accurate reflection of the time-varying importance of risk indicators. Building on the theoretical framework of attribute interval recognition theory, we establish a TBM tunnel construction safety risk evaluation model grounded in attribute interval theory. This model effectively addresses the various uncertainties inherent in the risk evaluation process for TBM tunnel construction. It determines the evaluation level and attribute intervals for each individual indicator, and subsequently calculates the interval values for both the individual indicators and the composite attribute. Finally, a case study demonstrates the model's effectiveness and practical applicability.