Study on Bayesian Networks based Aviation Safety Analysis

Security plays significant role in aviation field which directly influences the survival and development of the aviation enterprises.Appropriate scientific security analysis may help find existed security problems in time and prevent initiation factors happening which may induce serious aviation accidents.By utilizing the analysis method it can enhance the scientific and maneuverability nature of the aviation security management.A FT-based multi-modal BN security analysis system is presented in this paper based on aviation security characteristics.A comprehensive application of diagnostic reasoning and causal reasoning is used to achieve a combination of qualitative and quantitative analysis of aviation security.A real case of aviation accident has been investigated to the method validity.On this basis,an interval Bayesian networks analysis model of aviation security is used to overcome the disadvantages of the limitation of the single-node probability calculation.Compared with the traditional security assessment system,a more objective and more persuasive one is presented.