Michael North - Agent Models Argonne Labs

Michael J. North, MBA, Ph.D.
https://www.linkedin.com/in/drmichaelnorth

Group Leader
Integrated Analytics Group
Systems Science Center
Global Security Sciences Division
Argonne National Laboratory
9700 S. Cass Avenue
Argonne, IL 60439-4854
north@anl.gov
www.gss.anl.gov
630-252-6234

Senior Fellow
Computation Institute
The University of Chicago
Searle Laboratory
5735 South Ellis Avenue
Chicago, IL 60637-1403
north@uchicago.edu
www.ci.uchicago.edu
773-702-3946

Agent-Based Consumer Market Modeling

A number of my former colleagues at Procter & Gamble have co-authored a paper with some smart people I met at Argonne National Labs. Abstract below. Fascinating application of agents to markets. With very interesting details. To appear in Complexity Magazine. Congrats for its publication to everyone.

Multiscale agent-based consumer market modeling
Michael J. North 1 , Charles M. Macal 1, James St. Aubin 2, Prakash Thimmapuram 3, Mark Bragen 2, June Hahn 4, James Karr 4, Nancy Brigham 4, Mark E. Lacy 4, Delaine Hampton 4
1Center for Complex Adaptive Agent Systems Simulation, Argonne, Illinois 60439
2Modeling, Simulation, and Visualization Group, Argonne, Illinois 60439
3Center for Energy, Environmental, and Economic Systems Analysis, Argonne National Laboratory, Argonne, Illinois 60439
4The Procter & Gamble Company, Cincinnati, Ohio 45202

Abstract
Consumer markets have been studied in great depth, and many techniques have been used to represent them. These have included regression-based models, logit models, and theoretical market-level models, such as the NBD-Dirichlet approach. Although many important contributions and insights have resulted from studies that relied on these models, there is still a need for a model that could more holistically represent the interdependencies of the decisions made by consumers, retailers, and manufacturers. When the need is for a model that could be used repeatedly over time to support decisions in an industrial setting, it is particularly critical. Although some existing methods can, in principle, represent such complex interdependencies, their capabilities might be outstripped if they had to be used for industrial applications, because of the details this type of modeling requires. However, a complementary method - agent-based modeling - shows promise for addressing these issues. Agent-based models use business-driven rules for individuals (e.g., individual consumer rules for buying items, individual retailer rules for stocking items, or individual firm rules for advertizing items) to determine holistic, system-level outcomes (e.g., to determine if brand X's market share is increasing). We applied agent-based modeling to develop a multi-scale consumer market model. We then conducted calibration, verification, and validation tests of this model. The model was successfully applied by Procter & Gamble to several challenging business problems. In these situations, it directly influenced managerial decision making and produced substantial cost savings.

Managing Business Complexity

Re-Discovered: Managing Business Complexity: Discovering Strategic Solutions with Agent-Based Modeling and Simulation by Michael J. North and Charles M. Macal . Read parts of this back in 2007 when it was published as an introduction. Now I am re-examining the use of Agent-Based Modeling (ABM) and the book forms an excellent study of the methodology and a number of applied examples. Both authors work at Argonne National Labs and we very successfully used their consulting services. Instructional pieces in this area are fragmented, but this is the only the book I know that puts it all together. I suggest at the least the first few chapters and browsing the rest for examples.

Managing Business Complexity

Just received and have been exploring the book: Managing Business Complexity: Discovering Strategic Solutions with Agent-Based Modeling and Simulation by Michael J. North and Charles M Macal. The Amazon link is a 'search inside the book' so you can take a quick look. The second chapter 'The ABMS Paradigm' is an especially useful and relatively detailed introduction of the idea. The following chapter gives you an idea about what agents are all about, and why you should care. Later chapters are more academic, but also include examples of both approach and business problems being addressed by these models. Of particular interest, a store simulation model.

I was struck while scanning and reading parts of this book by how similar this approach is to the AI methods of the 90s. What it adds to the idea is the aspect of the agent, a simple entity that forms the basis of the modeling approach. And, finally, but also very important, these models are all simulations, often adaptive, that need to be understood statistically.

The authors are practitioners of this form of modeling at Argonne National Labs, and are consultants for ABM work. I have heard North talk on the topic and he knows the approach very well. I saw some early drafts of this book as well. Recommended.