Professor Alba Sofi, Department of Civil, Energy, Environmental & Materials Engineering, University Mediterranea of Reggio Calabria, Italy

Analysis of structures with UNCERTAIN parameters MODELED AS interval VARIABLES
When Feb 23, 2015
from 02:00 PM to 03:00 PM
Where LR8, IEB Building, Engineering Science
Contact Name
Contact Phone 01865-283302
Add event to calendar vCal


Uncertainties affecting structural parameters, such as material or geometric properties, have traditionally been modeled in the context of the classical probability theory as random variables or random fields. Well-established probabilistic methods have been developed to analyze the effects of uncertainties on structural performance. However, available data are often insufficient to build credible probabilistic distributions of the uncertain parameters, especially in early design stages.

Over the last decade, several non-probabilistic approaches have gained much popularity as alternative tools for quantifying and processing uncertainties described by fragmentary or incomplete data. In this context, the interval model, originally developed on the basis of the interval analysis, is widely used when only bounds are known for the parameters involved in the engineering problem. This model does not provide any information on the frequency of occurrence of values between the lower and upper bounds. The analysis of structures with interval basic input parameters is thus oriented to estimate the range of variation of the response quantities. Solutions obtained by applying the classical interval analysis are often useless from an engineering point of view due to excessive conservatism. Therefore, much research effort has been devoted in the literature to the development of alternative procedures able to limit the overestimation affecting the classical interval analysis.

In this presentation, I will illustrate the main features of the so-called improved interval analysis recently developed by my research group for analyzing the behavior of linear structures with uncertain geometric and/or material properties modeled as interval variables. The improved interval analysis has proved to be a very efficient approach for obtaining sharp bounds on the interval response quantities useful for decision-making in practical engineering.