EOSimulator Documentation

1.1.1

• What is EOSimulator?
• EOSimulator Features
• How to read this
• Acknowledgments
• Support

EOSimulator support discrete event simulation models using event scheduling world view with event method or three phase approach.

EOSimulator is a discrete event simulation library written in C++. It is intended to be used by programmers and students in Discrete Event Simulation courses.

EOSimulator was designed and implemented in the Departamento de Investigación Operativa - Facultad de Ingeniería - Universidad de la República by Sebastián Alaggia, under supervision of Maria Urquhart and Antonio Mauttone.

What is EOSimulator?

As we said above, EOSimulator is a discrete event simulation library written in C++. It's a object oriented framework for developing discrete event simulations. EOSimulator supports discrete event simulation models using event scheduling world view with event method or three phase approach. In this world view the model is structured by specifying those time points that are relevant to the system we are modeling.

EOSimulator is based in two very different discrete event simulation libraries: PascalSim (1989) and DesmoJ (1999-2005). It has some commons features of both libraries and some of its own. EOSimulator could be roughly described like PascalSim written through DesmoJ perspective.

In this page you will find full API documentation and some guidence to use the package correctly. We remind you that this is an experimental library, so we are very interested in the users and every suggestion is welcomed.

EOSimulator Features

EOSimulator has some interesting features:

• It separates the model form the structures needed to run a simulation. Those concepts are encapsulated in two classes: Model, which defines the system we are modeling; and Experiment, which provides the structures and operations needed to run a simulation.
• Capabilities to add several pseudo random number generators and distributions.
• Built-in containers for events, entities and distributions.
• Data output accumulators.

How to read this

This documentation is intended to be both an API documentation and User manual. The information is distributed all over the documentation and divided in modules. These modules are sets of clases which are not equal to the subsystem of EOSimulator. The modules are an intruduction to a subject, but the detailed information is distributed among the clases contained.

The modules are:

• Modeling Clases: explains how to create models with EOSimulator.
• Running Simulations: explains how to run a simulation.
• Distributions: this module is about sampling from distributions.
• Data Collectors: in this module the use of histograms in EOSimulator is explained.
• Collections: we explain how to use the collections.
• Graphic Simulations: finally we explain how to use write graphical simulations.

First read Modeling Clases (and maybe Collections), then Running Simulations. Finally read Distributions and Data Collector when you need them, Collections for models with priority queues and Graphical Simulations.

Acknowledgments

Many people have provided help with the development of this library:

• Bruno Martínez and Fernando Pardo for helping with the implementation of the library.
• The students of from the "Simulación a Eventos Discretos 2005" who "dare" to use the first version of this package and whose suggestions were used to improve the usability.

Support

EOSimulator is an experimental simulation library. For problems, bugs or comments, send an me e-mail with the subject "EOSimulator:". Every comment about EOSimulator is welcomed.

---