Workshop on Uncertainty Quantification and Error Estimation

Johns Hopkins University, Baltimore MD

Investigators

Abstract

WORKSHOP ON UNCERTAINTY QUANTIFICATION AND ERROR ESTIMATION A workshop will be held at the Johns Hopkins University on November 13-14, 2003 to developa research plan for the role that Uncertainty Quantification (UQ) can and must play within Computational Science and Engineering. In particular, it is postulated that UQ contributes to broaden the definition of Error Estimation to include errors associated with models, data, statistical analysis, and inverse analysis. From this perspective, it is desirable to seek a framework whereby all of these errors can be abstracted and modeled simultaneously. This will provide a rigorous and rational bridge between model-based predictions and experimental evidence. The workshop will bring together experts in the fields of Error Estimation, Uncertainty Quantification, Experimental Mechanics, and a variety of Application areas. Their charter is to identify an ambitious albeit achievable mission for UQ. The first day of the workshop will involve seven overview presentations that will underscore the state-of-the-art in the key individual science areas. The second day will involve eight break-out sessions developed around an application area which is meant to highlight certain the significance of Uncertainty Quantification to the state-of-practice, and a path for achieving that. It is anticipated that a synthesis of the outcome from all break-out sessions will give a more defined shape to the significance of UQ and its role in science and technology.