Answers are grounded in this paper and cite their sources.
Reliable prediction of soil and structural deformation is important not only for safety but also for resource-efficient underground construction. Excessively conservative constitutive modeling may lead to unnecessary use of retaining, bracing, and cementitious ground-improvement materials, whereas underestimated deformation can result in infrastructure damage and resource-intensive remediation. In this study, parameters for the Modified Cam-Clay (MCC), Hardening Soil (HS), and Hardening Soil with small-strain stiffness (HSS) models were determined for silty clay sampled at the investigated site in Nanning, China, using laboratory tests on undisturbed samples, in situ shear-wave velocity measurements, and empirical correlations. The three parameter sets were then implemented in PLAXIS 2D to simulate a representative excavation section above existing twin metro tunnels, and the calculated responses were compared with field monitoring data. The MCC and HS models reproduced the overall deformation patterns of the retaining system and tunnels but systematically overpredicted tunnel heave. The HSS model provided closer predictions of tunnel vertical displacement and central-column rebound than the MCC and HS models at the final excavation stage, although the agreement varied with monitoring location and excavation stage. For tunnel sidewall displacement, however, the HSS model tended to underestimate deformation near the central column, whereas the HS model showed better local agreement. These findings suggest that accounting for small-strain stiffness and adopting site-specific constitutive parameters may help reduce overprediction of excavation-induced rebound in the investigated case. The field-evaluated parameter sets provide a modeling basis for safer excavation design and may support more rational use of retaining, bracing, and ground-improvement measures, although the associated material and environmental benefits were not directly quantified in this study.
No takes yet. Share an insight, caveat, or question.
Lv et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: