Underlying issues associated with validation and verification of dynamic data driven simulation

Authors:
John G. Michopoulos;Samuel G. Lambrakos
Affiliations:
Center of Computational Material Science, Naval Research Laboratory, Washington, DC;Center of Computational Material Science, Naval Research Laboratory, Washington, DC
Venue:
Proceedings of the 38th conference on Winter simulation
Year:
2006

Citing 6
Cited 1

Theory of Modeling and Simulation

Theory of Modeling and Simulation
Inverse Problem Theory and Methods for Model Parameter Estimation

Inverse Problem Theory and Methods for Model Parameter Estimation
On a data-driven environment for multiphysics applications

Future Generation Computer Systems - Special section: Complex problem-solving environments for grid computing
Proceedings of the 5th international conference on Computational Science - Volume Part III

ICCS'05 Proceedings of the 5th international conference on Computational Science - Volume Part III
Dynamic data driven methodologies for multiphysics system modeling and simulation

ICCS'05 Proceedings of the 5th international conference on Computational Science - Volume Part II
On the fundamental tautology of validating data-driven models and simulations

ICCS'05 Proceedings of the 5th international conference on Computational Science - Volume Part II

A dynamic data-driven approach for rail transport system simulation

Winter Simulation Conference

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a brief exposition of three underempha-sized issues concerning modeling and simulation as they relate to Dynamic Data Driven Application Systems. One issue concerns the fact that conventional procedures for validation of data-driven models and simulations are unnecessary as they intrinsically contain validation in that they have been constructed according to acquired data. Another issue concerns the inherent coupling between the experimental frame used to measure system response and the system itself. This coupling can lead to an unrealistic simulation of the system in that the data contains the interaction of the system with the experimental frame. The final issue concerns the inherent pluralism of parametric representation and of potential mappings from data space into model space. This inherent pluralism imposes the need for optimal model and data space navigation procedures that are structured for appropriate sampling of data space and specification of faithful model parameterizations.