Validating Behavioral Models For Reuse

Author(s)
Paredis, Christiaan J. J.
Malak, Richard J., Jr.
Advisor(s)
Editor(s)
Associated Organization(s)
Series
Supplementary to:
Abstract
When using a model to predict the behavior of a physical system of interest, engineers must be confident that, under the conditions of interest, the model is an adequate representation of the system. The process of building this confidence is called model validation. It requires that engineers have knowledge about the system and conditions of interest, properties of the model and their own tolerance for uncertainty in the predictions. To reduce time and costs, engineers often reuse preexisting models that other engineers have developed. However, if the user lacks critical parts of this knowledge, model validation can be as time consuming and costly as developing a similar model from scratch. In this article, we describe a general process for performing model validation for reused behavioral models that overcomes this problem by relying on the formalization and exchange of knowledge. We identify the critical elements of this knowledge, discuss how to represent it and demonstrate the overall process on a simple engineering example.
Sponsor
G.W. Woodruff School of Mechanical Engineering Georgia Institute of Technology
NASA Ames Research Center (under cooperative agreement NNA04CK40A)
Date
2007
Extent
Resource Type
Text
Resource Subtype
Paper
Pre-print
Rights Statement
Rights URI