Agent-based modeling and simulation (ABMS) is a new approach to modeling systems composed of autonomous, interacting agents. Computational advances, new modeling know-how, and specialized agent-based modeling toolkits have enabled the development of agent-based models spanning the full range of application domains. Applications range from modeling trader behavior in the stock market to modeling consumer purchasing decisions, from predicting the possible spread of an H1N1 epidemic to modeling the immune system at the cellular level, from modeling the behavior and genetic evolution of microbes to identifying plausible causes of the fall of ancient civilizations, from modeling the predator-prey relationship between killer whales and other marine mammals to assessing the viability of new markets for space tourism, from building whole economies of agents “from the bottom-up” to modeling military units operating in urban environments, and many more.

Such progress suggests that ABMS could have the potential to have far-reaching impact on the use of models, whether the impact is on business and government use of computers to support decision-making and policy analysis or whether it is on scientists’ use of agent-based models as electronic laboratories for extended experimentation beyond what is possible in the traditional laboratory setting. Some even contend that ABMS is a “third way of doing science” whereby knowledge discovery proceeds through computational experimentation to augment traditional deduction and induction.

But is agent-based modeling really a revolutionary new modeling technique or just discrete event simulation in another guise? Will the full potential of agent-based modeling ever be realized or remain a pipe dream due to key hurdles that ultimately cannot be overcome? ABMS is inherently interdisciplinary, with deep connections to many fields, including complex adaptive systems (CAS), artificial life (ALife), behavioral, cognitive, and social sciences, as well as traditional fields such as operations research, discrete-event simulation, and systems engineering. As such, there is a rich set of diverse views concerning the purpose, utility, and directions for the field of agent-based modeling.

In this talk, we introduce agent-based modeling, describe the foundations of the field, discuss illustrative applications, show how ABMS is related to other modeling and simulation techniques, address agent-based model development methods, and offer some thoughts about where the field is going.

Charles Macal is Senior Systems Engineer and Director of the Complex Adaptive Systems Group at Argonne National Laboratory. He applies computational modeling and simulation tools to complex systems in a variety of fields, including energy, national security and defense. He is a principal investigator for the development of the widely used Repast agent-based modeling tookit. Dr. Macal serves on the editorial boards of the ACM Transactions on Modeling and Simulation and the journal Simulation, and he has co-authored a book, Managing Business Complexity: Discovering Strategic Solutions with Agent-Based Modeling and Simulation, published in 2007 by Oxford Press. He is Adjunct Professor at The University of Chicago, where he teaches a course on Complex Adaptive Systems for Threat Management and Emergency Preparedness. Dr. Macal received a Ph.D. in Industrial Engineering & Management Sciences from Northwestern University and holds an M.S. in Industrial Engineering and a B.S. in Engineering Sciences from Purdue University. He is a registered professional engineer in the State of Illinois and is a member of the Association for Computing Machinery (ACM), Institute of Electrical and Electronics Engineers (IEEE), Society for Computer Simulation International (SCSI), the Institute for Operations Research and the Management Sciences (INFORMS), and the Systems Dynamics Society (SDS).