Title: Web-based random variate generation for stochastic simulations

Authors: Mohammad Sabah, Osman Balci

Addresses: Department of Computer Science, 660 McBryde Hall, MC 0106, Virginia Tech, Blacksburg, VA 24061, USA. ' Department of Computer Science, 660 McBryde Hall, MC 0106, Virginia Tech, Blacksburg, VA 24061, USA

Abstract: This paper presents a random variate generation (RVG) world wide web (web) service and an RVG web-based software application, both of which are developed in compliance with the Java 2 Enterprise Edition (J2EE) architecture and run under the IBM WebSphere Application Server. The RVG web service can be invoked programmatically over the web by using simple object access protocol (SOAP) through the hyper text transfer protocol (HTTP) using the extensible markup language (XML). The RVG web service and software both generate random variates from 27 probability distributions and provide general statistics, scatter plot, and histogram of the requested random variates. The plots and histograms are created in the scalable vector graphics (SVG) format. The RVG web service can be used for stochastic input data modelling of a web-based simulation. The RVG software can be used as a web-based application for a variety of purposes including teaching random variate generation, probabilistic modelling and random experimentation.

Keywords: random variate generation; simulation experiment design; input data modelling; web service; web-based simulation; stochastic simulations; internet; XML; simple object access protocol; probabilistic modelling; random experimentation; software reuse; software interoperability.

DOI: 10.1504/IJSPM.2005.007110

International Journal of Simulation and Process Modelling, 2005 Vol.1 No.1/2, pp.16 - 25

Published online: 27 May 2005 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article