ABSTRACT
In many engineering applications insufficient data are available to justify the choice of a particular PDE This lack-of-data problem has hampered the adoption of probabilistic techniques in several engineering fields. We propose the use of the probabilistic models that are characterized by the avoidance of the selection of a specific PDF for a variable in the absence of specific knowledge about the variable. A hierarchical model of a continuous family of PDF’s is used instead. Sometimes the only data available are in the form of interval estimates which represent, often conflicting, expert opinion. The classical Bayesian estimation methods are expanded to make use of imprecise interval data. Each of the expert opinions (interval data) are interpreted as random interval samples of a parent PDE Consequently, a partial conflict between experts is automatically accounted for through the likelihood function. An example illustrates how this epistemic uncertainty modeling can be integrated in existing reliability techniques to assess the confidence in the reliability computations.
