A comparative analysis of regression and neural networks in simulation metamodeling

Abstract

In today's business, a company must be able to make efficient use of its available resources. Decision making techniques play a significant role in enabling companies to optimize their limited resources. Discrete-Event Computer Simulation is one of the more widely used decision making tools. However, simulation in itself is a descriptive tool, and provides solutions to different alternatives that an analyst wishes to explore. Because running a simulation is computationally expensive, randomly trying different alternatives results in inefficient practice at a high cost. Simulation Metamodeling is an inexpensive method of exploring various alternatives that can be used with simulation models as decision-making tools. Simulation metamodels reduce the time needed to explore and optimize a simulated scenario. However the appropriate simulation metamodel to be used for a specific goal has not yet be identified. This thesis will investigate the construction of two types of metamodels for a given simulated scenario (regression metamodels and neural network metamodels), and compare their ability to accurately predict the simulation model. Conclusions will be drawn on the ability of each to accurately predict the simulation model, and suggestions will be made on the type of characteristics to which each method is suited.

Notes

This item is only available in print in the UCF Libraries. If this is your thesis or dissertation, you can help us make it available online for use by researchers around the world by STARS for more information.

Thesis Completion

2000

Semester

Spring

Advisor

Mollaghasemi, Mansooreh

Degree

Bachelor of Science (B.S.)

College

College of Engineering

Degree Program

Industrial Engineering

Subjects

Dissertations, Academic -- Engineering;Engineering -- Dissertations, Academic;Computer simulation;Decision making -- Computer simulation;Neural networks (Computer science)

Format

Print

Identifier

DP0021624

Language

English

Access Status

Open Access

Length of Campus-only Access

None

Document Type

Honors in the Major Thesis

This document is currently not available here.

Share

COinS