Abstract: To effectively assess the resilience of highway tunnel construction safety systems during the tunnel entry stage in landslideprone areas, a Bayesian network resilience evaluation model is developed using Dempster-Shafer(D-S) evidence theory. This model establishes 16 resilience evaluation indices for the construction safety system of highway tunnel openings from the perspectives of pressure, state, and response according to the pressure-state-response model. Subsequently, a Bayesian network is constructed based on the binary relationships among these indices. Finally, D-S evidence theory is applied to compute conditional probabilities of BN node, and the resilience level is analyzed using Netica software. The model was implemented for the portal construction of the Baiyashan tunnel on the Zhuoni-Hezuo expressway in Gansu province, China. The calculation results indicate that the resilience level of the construction safety system at the tunnel entrance is robust under the influence of landslide disasters. These findings agree well with the monitoring results, thus validating the feasibility and applicability of the proposed evaluation model.

Key words: highway tunnel, resilience evaluation, Bayesian network, landslide disaster, DempsterShafer evidence theory, pressurestateresponse model