Abstract: It is prone to encountering water inflow for the karst tunnel. An extension model for assessing the risk of water inflow in karst tunnels was established, in which a multi-index evaluation system of water inflow in karst tunnels was set up and the evaluation indices were nondimensionalized; a matter-element to be evaluated was built based on the measured data of the evaluation indices of water inflow in karst tunnels; comprehensive weighting method was used to assign weights for the evaluation indices, the objective weights are determined by measured data based correlation function, and the subjective weights are determined by fuzzy analytical hierarchy process based on the expert knowledge; the correlation function value of the matter-element to be evaluated with respect to the risk grades of water inflow was calculated, and the maximum correlation degree principle was adopted to determine the risk grade of water inflow of the tunnel to be evaluated; the variable characteristic value was calculated to determine the tendency of the risk grade being classified to the next risk grade. This extension model was used to assess the risk of water inflow of a section of certain karst tunnel and the assessment results agree well with the measured data, verifying the reliability of this extension assessment model.