Abstract: Water inrush hazard has been a major issue that seriously imperils safe construction of tunnels. In order to achieve real-time monitoring of evolution process of water inrush and disaster warning, the electrical resistivity tomography (ERT) method is introduced for monitoring water inrush in the tunnels. A quantitative evaluation method of ERT is proposed based on the gray correlation theory to quantitatively describe the monitoring information of water inrush. First, the two typical kinds of water inrush hazards caused by geological defects and non-geological defects are generalized, and geoelectric models of different stages during the evolution process of water inrush are acquired, which lays a foundation for numerical forward of real-time monitoring of water inrush of tunnels. Then, the finite element method is adopted to carry out the numerical forward, and the response characteristics of water inrush evolution process are revealed. ERT gray correlation quantitative analysis of water inrush process is done. Sample points whose correlation coefficient is less than 0.5 or 0.3 greatly increase, and the standard deviations of correlation distribution and forward image dramatically increase, which are an important precursory pattern when water inrush happens. Finally, model tests on ERT of water inrush of tunnel are performed, and precursor information of water inrush is captured well and disaster early-warning of water inrush is realized, which shows it is feasible that ERT is applied to real-time monitoring.