Structural reliability analysis using evidence theory and fuzzy probability distributions

A structural element failure probability (as a structural reliability measure) is an indicator of the structural safety. Approaches for structural reliability analysis with incomplete statistical data are a special scientific problem. In the development of this scientific direction, the article proposed a method for structural reliability analysis based on a combination of evidence theory and fuzzy probability distributions when the problem of reliability analysis involves quantitative and qualitative uncertainty at the same time. The article presents an experimental study of reliability analysis for a steel truss by the truss members strength criterion based on various approaches to reliability analysis. The reliability interval [0.99272; 1] of the proposed method covers the FOSM (First Order Second Moment) reliability value of 0.99354. From the experiment results, it follows that the proposed approach can be used in practice for a more cautious assessment of the structural reliability with incomplete statistical information. The proposed approach also allows reducing the number of tests and getting an operational (preliminary) assessment of the structural element reliability. The value of the acceptable reliability level in discrete or interval form should be set individually for each design situation taking into account the risk of economic and non-economic losses.