Abstract
Although various approaches have been reported for forecasting aviation safety risks, they frequently fail to fully consider the stochastic nature and complex interrelations of numerous real-time factors that influence safety. This highlights the urgency for flexible and realistic analytical tools that can quantify the probability of aviation accidents and identify critical vulnerabilities to support proactive safety management. In this study, a probabilistic multiplicative stochastic risk model was developed based on Reason’s “Swiss cheese” concept for quantitative risk assessment and dynamic forecasting. Key risk parameters (e.g., the probability of breaking through a protective barrier) were modeled as random variables with beta distributions based on empirical observations and expert judgment to capture uncertainty and variability. Large-scale scenario modeling (10,000 simulations per scenario) was conducted using the Monte Carlo method to obtain statistically stable estimates, acknowledging the limitations posed by sparse accident data and reliance on expert judgment. This study contributes to aviation safety by adapting the Swiss cheese model to a stochastic framework parameterized with regional data. The novelty of this approach is the quantification of barrier-specific probabilities and the demonstration of how scenario-based improvements affect overall accident likelihood. The results highlight the potential and limitations of applying rare-event risk models in regions with sparse accident statistics.