Dimensions of credibility in models and simulations

  • Authors:
  • Martin J. Steele

  • Affiliations:
  • National Aeronautics and Space Administration (NASA), Kennedy Space Center, FL

  • Venue:
  • Proceedings of the 2008 Summer Computer Simulation Conference
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Based on the National Aeronautics and Space Administration's (NASA's) work in developing a standard for models and simulations (M&S), the subject of credibility in M&S became a distinct focus. This is an indirect result from the Space Shuttle Columbia Accident Investigation Board (CAIB), which eventually resulted in an action, among others, to improve the rigor in NASA's M&S practices. The focus of this action came to mean a standardized method for assessing and reporting results from any type of M&S. As is typical in the standards development process, this necessarily developed into defining a common terminology base and common documentation requirements (especially for M&S used in critical decision making), along with a method for assessing the credibility of M&S results. What surfaced in the development of the NASA Standard was the various dimensions of credibility to consider when accepting the results from any model or simulation analysis. The eight generally applicable factors of credibility chosen in the NASA Standard proved only one aspect in the dimensionality of M&S credibility. At the next level of detail, the full comprehension of some of the factors requires an understanding along a couple of dimensions as well. Included in this discussion are the prerequisites for the appropriate use of a given M&S, the choice of factors in credibility assessment with their inherent dimensionality, and minimum requirements for fully reporting M&S results.