Analysis of structural performance in the framework of imprecise probabilities

The analysis of structures with uncertain properties modeled as random variables with imprecise Probability Density Function (PDF) characterized by interval basic parameters (mean-value, variance, etc.) is addressed. A novel procedure able to provide approximate explicit expressions of the bounds of the interval mean-value and variance of the random stresses is proposed. The procedure stems from the joint application of the Improved Interval Analysis via Extra Unitary Interval and the Rational Series Expansion, introduced in the literature by the last two authors. The influence of imprecision of the PDF of the input parameters on structural performance is also investigated. For validation purposes, a 3D truss structure with uncertain Young's moduli is analyzed.