Georgia Tech Savannah has been the home of the bi-annual international workshops on Reliable Engineering Computing, and is proud to once again serve as host for REC 2008. These workshops are unique in combining computer science, mathematics, and engineering analysis and design to discuss the reliability of engineering computations, as well as providing a common forum by which to continue cross-disciplinary advisements in the field. The focus of this cycle will be the NSF workshop on Imprecise Probability in Engineering Analysis and Design and its role in Reliable Engineering Computing.
Design of an engineered system requires the performance of the system to be guaranteed over its lifetime. One of the major difficulties a designer must face is that neither the external demands of the system nor its manufacturing variations are known exactly. In order to overcome this uncertainty, the designer currently provides excessive capabilities and over designs the system. As analysis tools continue to be developed, the predictive skills of designers have become finer. In addition, the demands of the market place require that more efficient and reliable designs be developed. In order to satisfy these current requirements in designs subject to uncertainties, the uncertainties in the performance of the system must be included in the analysis. Quantification of uncertainties requires knowledge about the probability of various system parameters; for continuous parameters, uncertainty can be defined by a probability density function (PDF).
In light of the above, the routine application of probability based design has been slower than one would expect. We hypothesize that there are two fundamental reasons for this slow adoption: the first is the difficulty of acquiring the needed information (PDFs) for risk based design and the second is the lack of viable engineering tools allowing for imprecise or incomplete information to be employed in a less than fully probabilistic design.Over the last three decades a number of methods have been developed to handle such situations in which information is incomplete, scarce, vague or conflicting. These methods include interval probability, probability bounds analysis, Dempster-Shafer theory, fuzzy-based, information-gap theory, clouds, and the most general approach that comes from the theory of imprecise probability.
We believe that the development of methods under general imprecise uncertainty representation would result in a major change in engineering design, promoting the universal adoption of probabilistic and extended probabilistic design and concomitant improvement in manufactured products.