Proxel-Based Simulation

Lazarova-Molnar, Sanja and Horton, G. (2003) Proxel-Based Simulation. In: Proceedings of the Fourth Conference on Informatics and Information Technology. Institute of Informatics, Faculty of Natural Sciences and Mathematics, Ss. Cyril and Methodius University in Skopje, Macedonia, Skopje, Macedonia, pp. 41-48. ISBN 9989-668-45-0

[img]
Preview
Text
9989-668-45-0_pp41-48.pdf

Download (282kB) | Preview
Official URL: http://ciit.finki.ukim.mk

Abstract

Proxels ("probability elements") were recently introduced as a new algorithmic approach to analysing discrete-state stochastic models such as are represented by stochastic Petri nets or queueing systems. Usual approaches for this type of analysis are discrete-event simulation and solving partial differential equations derived from continuous-time supplementary variables. Both of these techniques have advantages and disadvantages. The proxel-based method provides a new way to simulate stochastic models and is designed to avoid the biggest disadvantages of each of the two existing approaches. The proxel-based method is deterministic, as opposed to discrete-event simulation, which uses a random number generator to imitate the random variables in the model. It thus avoids the problems associated with Monte-Carlo simulation such as the need for many replications in the presence of rare events. This paper focuses on the description of the analysis of stochastic models using the proxel-based method. Simulation experiments are used to demonstrate the behaviour of the method, and its advantages and disadvantages are discussed.

Item Type: Book Section
Uncontrolled Keywords: proxel-based simulation, discrete stochastic models, supplementary variables, stochastic Petri nets.
Subjects: International Conference on Informatics and Information Technologies > Computer Science
International Conference on Informatics and Information Technologies > Information Technology
Depositing User: Vangel Ajanovski
Date Deposited: 28 Oct 2016 00:15
Last Modified: 28 Oct 2016 00:15
URI: http://eprints.finki.ukim.mk/id/eprint/11268

Actions (login required)

View Item View Item