Discrete Event Simulation Model Decomposition

Abstract

Simulation models are currently being used for a multitude of purposes. Some simulation models are concise and well organized thereby facilitating their usage. However, some of these models may be quite lengthy and complex which causes development costs to rise beyond an acceptable level. The creation of an effective development environment has helped to eliminate this problem. Tools which are used to support modeling and analysis have been incorporated into this environment. Such an environment uses a model specification language which has condition specifications as its basic constructs. This research focused on interpreting a discrete event simulation model's condition specification primitives and their associated actions. A network representation was created using these condition action pairs (CAPs) as network nodes. The arcs, or edges, of the network represent information being transferred such as specific attributes of the CAPs. This network representation was decomposed into smaller networks, or sub-networks, by taking advantage of the structure of the network. The structure of the network was translated via a software interface into an edge-incidence matrix (E-matrix). The E-matrix was then transformed into a pseudo-covariance matrix (C-matrix). The C-matrix was used in the creation of a SAS data set which served as the input necessary to do principal components analysis. Two examples were used to demonstrate this procedure. Theses.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1988
Accession Number
ADA206048

Entities

People

  • Scott R. Matthes

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Abstracts
  • Air Force
  • Computer Languages
  • Computer Programming
  • Computer Programs
  • Computers
  • Covariance
  • Data Science
  • Data Sets
  • Factor Analysis
  • Information Science
  • Language
  • Mainframe Computers
  • Network Science
  • Simulations
  • Specifications
  • Statistical Analysis

Readers

  • Computational Modeling and Simulation
  • Database Systems and Applications
  • Neural Network Machine Learning.