Date of Award

Summer 8-2011

Degree Type


Degree Name

Doctor of Philosophy (PhD)




Computing Sciences and Computer Engineering

Committee Chair

Dia Ali

Committee Chair Department


Committee Member 2

Ray Seyfarth

Committee Member 2 Department


Committee Member 3

Beddhu Murali

Committee Member 3 Department


Committee Member 4

Christopher Winstead

Committee Member 4 Department

Physics and Astronomy

Committee Member 5

Grayson Rayborn

Committee Member 5 Department

Physics and Astronomy


A new paradigm for utilization of mobile agents in a modular architecture for scientific simulation is demonstrated through a case study involving Monte Carlo simulation of low energy electron interactions with molecular nitrogen gas. Design and development of Monte Carlo simulations for physical systems of moderate complexity can present a seemingly overwhelming endeavor. The researcher must possess or otherwise develop a thorough understanding the physical system, create mathematical and computational models of the physical system’s components, and forge a simulation utilizing those models. While there is no single route between a collection of physical concepts and a Monte Carlo simulation based on those concepts, this work develops a new paradigm based on agent-oriented architecture and modular design principles through a case study in which interactions between electrons and molecules are simulated. A methodology that incorporates both distributed and modular computing concepts is shown to facilitate the researcher’s selection of component granularity as well as the connectivity and interaction of the simulation components. The case study is specific, however, the techniques employed in addressing the encountered problems are general and applicable to a much broader range of scientific simulation. A paradigm is developed through which the burden of information management in distributed Monte Carlo simulations is lessened through realization of a modular system of agents that may be augmented as a virtual collaborative community.

An understanding of the physics to be simulated is a prerequisite of model development. Research has been conducted to provide the required understanding of the associated knowledge domain. The separable nature of the processes involved in air fluorescence provide suitable processes for a modular distributed simulation. Physical processes that can be decoupled and implemented as modular physics agents have been identified.

Models suitable for decoupling are implemented as OSGi bundles used by JADE agents. The OSGi architecture is used to define the types of data consumed and produced by models of a given physical process with no concern of specific implementation of the model. This allows third party developers freedom to implement models to their required level of detail with only the restriction of the model’s input and output upholding the contract defined by the framework. Agents for simulation of physical processes are based on published physics models and experimental data identified by a literature search. Agents simulating random processes employ a statistical technique known as the Monte Carlo Method.

The significance of this work extends beyond demonstrating the new paradigm for agency-oreinted Monte Carlo simulation that is both modular and extensible. The production of air fluorescence via interactions of ionizing radiation with atmospheric gases is a subject ongoing research. Development of Monte Carlo simulation of electron impact induced air fluorescence is of considerable value to related research efforts. Therefore, an opportunity exists not only to demonstrate the use of a modular agent-based paradigm for Monte Carlo simulation, but also to provide new capabilities for the investigation of physical phenomena.