Artificial Neural Network Metamodels of Stochastic Computer Simulations

Abstract

A computer simulation model can be thought of as a relation that connects input parameters to output measures. Since these models can become computationally expensive in terms of processing time and/or memory requirements, there are many reasons why it would be beneficial to be able to approximate these models in a computationally expedient manner. This research examines the use of artificial neural networks (ANN), to develop a metamodel of computer simulations. The development and use of the Baseline ANN Metamodel Approach is provided and is shown to outperform traditional regression approaches. The results provide a solid foundation and methodological direction for developing ANN metamodels to perform complex tasks such as simulation optimization, sensitivity analysis, and simulation aggregation/reduction. Artificial Neural Networks, Computer Simulation Metamodel, Regression, Response Surface Methods, Simulation Optimization.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 10, 1994
Accession Number
ADA285951

Entities

People

  • Robert A. Kilmer

Organizations

  • University of Pittsburgh

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence
  • Computational Science
  • Computer Programming
  • Computer Programs
  • Computer Simulations
  • Computers
  • Data Science
  • Databases
  • Experimental Design
  • Industrial Engineering
  • Information Science
  • Mathematical Models
  • Neural Networks
  • Simulations
  • Systems Engineering
  • Two Dimensional

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence
  • Computational Modeling and Simulation

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks