A Generic Object-Oriented Random Variate Model for Discrete Event Simulations Separating the Data Model from the Process Model

Date of Award

Fall 1999

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computational Modeling & Simulation Engineering

Program/Concentration

Modeling and Simulation

Committee Director

James F. Leathrum

Committee Member

John W. Stoughton

Committee Member

Roland R. Mielke

Call Number for Print

Special Collections LD4331.E55 F73

Abstract

Data is the lifeline of any simulation model. The validity of the data is a primary determining factor on the accuracy of a simulation. Hence the data model is an important part of any simulation model. But this is, in most occasions, the most neglected part of any model. Most people view the model correctness in te1ms of the process model functionality rather than the data that is being used in the simulation. As simulation gets more accepted as a reliable problem-solving tool, it is increasingly important to pay careful attention to the accuracy of the simulation and that can only be achieved if the data model is accurate and functionally capable.

Classically, all complexity associated with the functionality of the model is integrated in the process model. This is true even when the complexity is inherently within the data. This causes excess process model complexity and an undesirable rigidity in the model capabilities. The properties of the data itself might be highly varied and could require some complex modeling rather than simple calls to random number generators. Thus, to serve these purposes and address various basic software enginee1ing requirements, it is essential to model data in a more generic and detailed fashion so as to separate the complexity of the data from the process. To support this new simulation paradigm, new constructs are required at the data model level.

This study deals with the development of a generic advanced random variate model that supports several of the data modeling issues. The primary aim of the study is to create a separate data model that handles all the complexity that is associated with the data in the simulation. It is designed to provide added functionality that has been classically placed in the process model but which could easily be moved to the data model. This simplifies the process model considerably so as to enable easy debugging and enhancement work on the model. The design also aims at providing maximum flexibility in te1ms of reusability in different scenarios. The model is object-oriented and is kept as generic as possible to enable reuse of the model in other object-oriented discrete event simulations.

The following study involves the in-depth analysis of the requirements of the random variate object model and the steps carried out towards the designing of the model. It also includes tests that were carried out for testing its functionality.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/8bke-gq11

This document is currently not available here.

Share

COinS