According to the current sluice design specifications, five indicators that characterize the safety status of sluice chambers were selected as the research object, and 10 corresponding influencing factors were screened out. Based on Delayed Rejection Adaptive Metropolis(DRAM)algorithm, BP neural network and Garson algorithm, the importance analysis method of the influencing factors for the single and overall safety status of a sluice chamber was constructed. The case study shows that the above method is reasonable and feasible. The research results reveal that for the single safety status index, the water depth in the upstream and downstream, and the wind speed have greater impact on anti-sliding stability and base stress. The seepage slope at the exit section is also influenced by the water depth in the upstream and downstream. The water depth and the wind speed have greater impact on the overall safety status index.