|
|
 |
|
Francois Hemez with Los Alamos National Laboratory |
|
Date: October 6, 2004
Time: 10:30am
Location: NIA, Rm 404
Speaker: Francois Hemez with Los Alamos National Laboratory
Subject: "The Myth of Science-based Predictive Modeling (U)"
Additional Information: Webcast
In computational physics and engineering, numerical models are developed to predict the behavior of a system whose response cannot be measured experimentally. A key aspect of science-based predictive modeling is the assessment of prediction credibility. Credibility, which is usually demonstrated through the activities of model Verification and Validation (V&V), quantifies the extent to which simulation results can be analyzed with confidence to represent the phenomenon of interest with accuracy consistent with the intended use of the model.
The presentation develops the idea that assessing the credibility of a mathematical or numerical model must combine three components: 1) Improving the fidelity to test data; 2) Studying the robustness of prediction-based decisions to variability, uncertainty, and lack-of-knowledge; and 3) Establishing the expected prediction accuracy of the models in situations where test measurements are not available. A Theorem is established that demonstrates the irrevocable trade-off between fidelity to data, robustness to uncertainty, and confidence in prediction. Clearly, fidelity to data matters because no analyst will trust a simulation that does not reproduce the measurements of past experiments. Robustness to uncertainty is equally critical to minimize the vulnerability of decisions to uncertainty and lack-of-knowledge. It may be argued, however, that the most important aspect of credibility is the assessment of confidence in prediction, which is generally not addressed in the literature. The antagonism between three objectives (fidelity to data, robustness to uncertainty and confidence in prediction) suggests a decision-making strategy in situations where knowledge is severely lacking. These concepts are illustrated with an engineering application for which simplistic numerical simulations are implemented, yet, severe sources of lack-of-knowledge are considered.
|
|
|
Home