PARAMETER VALIDATION USING CONSTRAINT OPTIMIZATION FOR MODELING AND SIMULATION

Published

May 11, 2009

Author(s)

Guodong Shao, Charles R. McLean, Alexander Brodsky, Paul Ammann

Abstract

Modeling and simulation (M&S) techniques are increasingly being used to solve problems and aid decision making in many different fields. Results of simulations are expected to provide reliable information for decision makers. But potential errors may be introduced during the M&S development lifecycle. It is critical to ensure to build the right model and the model is built right. M&S community has had intensive Verification and Validation (V&V) research. But V&V activities are often not formally performed in most of the cases. For those who perform V&V activities, they normally wait until development of the simulation modeling is finished. Practical and solid validation techniques are needed. In this paper, the authors propose a novel validation methodology that allows parallel simulation development and model parameter validation, i.e. first the simulation model can be built with unknown parameters included; and then, those parameters can be estimated using a built-in constraint optimizer. Finally the initially unknown parameters are replaced with the found optimal values. The model is ready for future output prediction. A validation method for the technique is discussed. As an example application, a simple supply chain cost simulation model was discussed using the proposed methodology.

Proceedings Title

Proceedings of the Industrial Simulation Conference 2009

Conference Dates

June 1-3, 2009

Conference Location

Loughborough

Pub Type

Conferences

Download Paper

Local Download

Keywords

Modeling and Simulation (M&S), Verification and Validation (V&V), Model Parameter, Optimization

Citation

Shao, G. , McLean, C. , Brodsky, A. and Ammann, P. (2009), PARAMETER VALIDATION USING CONSTRAINT OPTIMIZATION FOR MODELING AND SIMULATION, Proceedings of the Industrial Simulation Conference 2009, Loughborough, -1, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=902524 (Accessed September 13, 2024)