Fuzzy Bayesian based bowtie risk assessment of an unstabilized approach: A method for airline flight operations

Risk is inevitable part of life. As in all areas of life, risk management is of great importance in aviation. Matrices are one of the most frequently used methods in safety risk management in aviation. In this method, risks are classified according to probability and severity. In this sense, matrix risk assessment process has a static and discrete structure. However, operations in aviation consist of dynamic components. Outputs of dynamic components change over time. Dynamic and continuous system risk assessment in aviation can contribute to the risk assessment process and improve overall performance of safety risk management. In this study, a dynamic risk assessment system design based on fuzzy logic and Bayesian network is combined with the bowtie method to analyze unstabilized approaches in flight operations. This approach forms a network structure with conditional probabilities and can be used with both quantitative and/or qualitative data. Each event in the system is connected to each other with conditional probabilities. Any change in probability of each basic event will update overall probability of the analysis. Hence, autonomous, and dynamic risk assessment process that produces fast results can be achieved.