Shield construction of deep tunnels for urban drainage involves many risk factors, and potential safety hazards are difficult to monitor and identify directly. In order to improve the risk management level of shield construction in urban drainage deep tunnel, this study proposes a method for identifying risk factors by combining text mining technology and the entropy weight method. By using this method, 34 safety risk factors were successfully extracted from the safety accident reports of urban drainage deep tunnel shield construction and the related text data. The results of this study show that the text mining method could play an important role in the risk management of urban drainage deep tunnel shield construction; the introduction of the entropy weight method further improved the accuracy of risk factor identification. The results of this study not only enrich the research content of risk management in urban drainage deep tunnel shield construction but also provide theoretical guidance for managers to formulate risk management measures and optimize risk management procedures.

Model tests are effective for studying the entire deformation and evolution process of reservoir landslides. The sensitivity of similar materials to seepage effects is crucial to the accuracy of landslide model testing. Based on a fuzzy evaluation of in situ sliding zone soil, this study compared three similar materials, using shear tests and microscopic SEM to assess the similarity. The optimal similar material (sliding zone soil: bentonite: standard sand = 50%: 20%: 30%) with a water content of 13.5% and a permeability coefficient of 3.8 × 10⁻⁶ cm/s was identified, simultaneously matching physical–mechanical properties and seepage effects. When the proportion of in situ sliding zone soil exceeds that of bentonite, the in situ sliding zone soil dominates the strength. Cohesion depends on interparticle cementation force and water film viscosity. Bentonite modifies these forces in stages, leading to a trend where cohesion (c′) first increases and then decreases with rising water content, while the internal friction angle (φ’) decreases continuously. Model test results indicate the failure mode of reservoir landslides is a three-stage traction-braking failure, evolving from initial shallow deformation to deep progressive failure and finally to overall large-scale instability. The proposed similar material exhibits reliable physical–mechanical and seepage similarity and can be directly applied in physical model tests of reservoir-induced landslides to reproduce the hydro-mechanical coupling behavior of sliding zones.