Date of Award


Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy in Computational Science and Engineering (PhD)

Administrative Home Department

Department of Social Sciences

Advisor 1

Mark Rouleau

Committee Member 1

Audrey Mayer

Committee Member 2

Donald Lafreniere

Committee Member 3

Thomas Oommen

Committee Member 4

Daisuke Minakata


This dissertation examines the application of agent-based modeling (ABM) to complex systems with the intent of developing a means of overcoming limitations present in existing tools. This is done though the development of two ABMs intended to address complex systems present in the fields of sustainability studies and chemistry.

After introductory information, Chapters 2 - 4 of this dissertation address the limitations of tools intended to project the environmental, economic, and social impacts of woody biomass based biofuels. Chapter 2 begins by discussing the limitations in tools to study timber harvest decision making and its impact upon the landscape, and develops an ABM platform to address this gap. Next, Chapter 3 presents a life cycle assessment (LCA) of a proposed biorefinery in Ontonagon, Michigan is conducted. This study acts as a benchmark benchmark for the case study presented in Chapter 4, where an argument for the integration of ABM and life cycle sustainability assessment (agent-based LCSA) is presented. The argument is followed by a case study demonstrating the applicability of the technique. The case study finds that while Ontonagon is a promising site for a biorefinery, there are concerns regarding the quantity of woody biomass that may be delivered as a feedstock and potential impacts upon regional wetlands.

Chapter 5 of this dissertation addresses the limitations of models of advanced oxidation processes (AOPs) using ordinary differential equations (ODEs). We argue that these limitations can be addressed by modeling the AOP as a complex system, including the complete elementary reaction pathway using ABM. To demonstrate the applicability of this novel approach, an ABM is developed and two in silico studies of acetone degradation induced by hydroxyl radicals are performed. We found that when using a comprehensive list of elementary reaction pathways, the ABM was able to replicate concentration curves for major chemical species in our laboratory study. As a novel application of ABM to AOPs we conclude that the technique shows considerable promise.