A SAFETY FACTOR QUANTIFICATION METHOD BASED ON BAYESIAN CREDIBLE RELIABILITY

To address the issues of overly conservative and insufficiently quantified safety factor selection grounded in traditional experience in aerospace structural design, this paper proposes a safety factor quantification method based on Bayesian credible reliability. Building on the analytical framework that relates safety factors to reliability in probabilistic models, the mapping between safety factors and reliability is derived for non-probabilistic models, thereby establishing a safety factor quantification method on the plane of safety factor versus stress distribution parameters. According to Bayesian theory, credible modeling is performed for structural stress and material strength. By specifying prior distributions and incorporating Bayesian posterior updating, uncertainty intervals at different credibility levels are obtained and employed to compute non-probabilistic credible reliability. Furthermore, the uncertainty model parameters are dynamically updated utilizing actual sample data to subsequently calibrate safety factors, thereby establishing a dynamic quantification method for safety factors. Numerical case studies of a typical rod and a wing structure systematically validate the engineering applicability of the proposed method, and discuss the influence of credible reliability indicators and sample size on safety factor selection in structural design, revealing the effectiveness of the credible reliability method in structural weight reduction. The numerical results demonstrate that the proposed method can effectively mitigate the conservatism of traditional experience-based safety factors, offering new foundations and guidance for the lightweight design of advanced structures in the aerospace field.