抄録
In the context of the Global Technical Regulations (GTR) aimed at harmonizing safety standards for composite pressure vessels, discussions on safety margins are actively conducted. However, in order to control safety margins rationally, it is inevitable to incorporate probabilistic considerations. For the future establishment of safety factor settings based on rational justifications, it is considered essential to first establish a solid probabilistic foundation.
Particularly, in cases such as GTR standards where safety assessments for burst strength are conducted based on small samples, such as three or five samples, it is unavoidable to address the uncertainties associated with small-sample treatment. Specifically, when discussing safety factors based on sample data, consideration of both reliability and confidence levels is necessary1)2) . This perspective is crucial in the introduction of reliability design, and the author has already compiled introductory content and challenges in an explanatory article3).
In this article, emphasis is placed on the treatment of probabilistic safety factors, and explanatory additions are made to contribute to the discussion, referencing the approach of G.W. Mair4).
