The safety situation of railway stations is facing great challenges due to the characteristics of huge transport task load, fast operation speed and larger station scale. Based on this, this paper proposes a decision-making method for safety risk factors of high-speed railway stations, which combines fuzzy fault tree (FFTA) and Bayesian network (BN) algorithms. Firstly, the basic risk factors and events of each subsystem of high-speed railway station in the actual construction and operation process are statistically analyzed and modeled, and a total of 22 basic events affecting the safe operation of railway stations are determined. Secondly, the limitation of data collection is overcome by means of expert experience evaluation, scoring and fuzzy rough set theory. Then, the fusion structure of fault tree analysis and Bayesian network algorithm is used to overcome the shortcomings of single algorithm in data processing ability and accuracy, effectively and accurately reveal the logic and probability relationship between safety risk factors, and four basic events such as the lack of technical means to master the station resource status and operation progress are identified as the main influencing factors. Finally, through sensitivity analysis, the risk reduction value importance (RRW) indicators of 22 basic events that affect the safety of high-speed railway station construction and operation are obtained. The results show that the RRW of the 4 main influencing factors are all greater than 15%, which verifies the correctness of the results obtained by FFTA and BN. The proposed decision-making method for safety risk factors of high-speed railway stations can realize accurate positioning and decision-making of key indicators and corresponding probabilities affecting the safety of high-speed railway stations, thus providing effective reference for ensuring the safe operation of high-speed railway stations. [ABSTRACT FROM AUTHOR]