Syllabus
Modeling and Simulation of Social and Organizational
Behavior
Course Number: S604
Time: Tuesday 9:30-12:15
Place: LI001
Instructor: Hamid R. Ekbia
Office Hours: Tuesday 1:00-2:30
Friday 2:30 – 4:00
Or by appointment
Introduction
Scientists, professionals, and policy-makers increasingly deal with complex systems
and phenomena – that is, with large aggregates of heterogeneous, structured, and
interactive (human or nonhuman) components. Modeling and simulation present effective
techniques for understanding and analyzing such systems and phenomena. Traditionally,
methods such as business process, mathematical, cognitive, distributed AI, or
computational organization modeling are used for this purpose. However, they all have
serious limitations that make them less useful for real situations and practices. In
recent years, various modeling techniques and platforms have been developed for this
purpose. NASA, for instance, has developed Brahms -- a general-purpose language for
modeling and simulating how people work in realistic situations such as those involved in
NASA missions and projects. Systems such as Brahms have both theoretical and practical
advantages. Theoretically, they help us understand how people work together, how they
participate in a collaborative activity, and how communication happens, and also
illustrate the role of tools, artifacts, and space as well as individual motives, history
and culture. Practically, these modeling techniques and platforms use simulation as a
design and analysis tool, and conceptualize humans as social and collaborative agents
that are situated, deliberative, and cognitive. Information scientists, students, and
practitioners can benefit from these techniques in their quest to understand interactions
among humans, information sources, technologies, organizations, etc. These techniques
will give students a human-centered perspective on work and organizational processes.
Understanding how people actually work will enable students to better define requirements
for information technology in organizations. It may also serve as a source of ideas,
tools, and methodology to help students understand existing organizations, design their
own processes, and generating their own projects. For instance, they can learn how to
develop tools that can
reduce work and information overload.
There are many other tools and environments available for modeling, especially for
agent-based modeling and simulation (ABMS), which would be the focus of this course. ABMS
is a framework that allows experimentation with simulated complex systems. It conceives
of a complex system as a set of agents that interacts and adapts to changing
environments. Starting with a description or theory of individual agents, it seeks to
formulate rules of behavior such as rational behavior (behavioral economics),
stimulus-driven motion (cell biology), flocking (animal behavior), utility maximization
(organization behavior) and so on. Based on these rules, ABMS can then be used to study
the behavior of a system as a whole – that is, of how the system could evolve over
time. By generating unexpected patterns of behavior that might emerge from interactions
among simple agents, ABMS provides insights into both individual agent and overall system
behaviors. It can help the experimenter anticipate system interactions, structures, and
possible evolutionary paths, allowing him or her to answer questions such as the
following:
Description
This course is an introduction to the modeling and simulation of human social and organizational behavior. Most methods for modeling such behavior focus on the process or functional levels of the phenomenon under study. However, in the past decade social scientists and computer scientists involved in social informatics and human-centered design have argued that if we want to develop better knowledge processes and usable information systems, we need to understand the 'living work practice' of the people in an organization. In this course students will learn what 'practice' is and how it can be observed, modeled and simulated. We will learn how to observe and model human activity and practice for the analysis of processes and the design of information systems.
The course will interleave theoretical discussions and technical methods in order to give students both a theoretical grounding and technical skills required to model complex systems. The class readings and lectures will review the literature, but a significant part of the class and its labs will be devoted to learning computational tools and techniques for modeling and simulating human social and organizational behavior, as well as for developing intelligent agent systems.
The course will include a final student project, which can be done individually or in
teams of two or more students. Possible projects include:
Text
There is no required textbook for this course. However, every week there are required
readings that all participants are expected to have read before class meeting. In
addition, there are also additional optional readings available for those who want to go
deeper into a topic. Both groups of readings will be made available electronically or in
hardcopy.
Schedule
Week 1: Introduction to Modeling and Simulation
In this introductory session, we discuss models, simulations, their similarities and differences, as well as their role as analytic tools for understanding complex systems and behaviors
Reading
Required:
Axelrod, R. (2005). Agent-based Modeling as a Bridge between Disciplines. In K. L. Judd and L. Tesfatsion (Eds.), Handbook of Computational Economics, Vol. 2: Agent-Based Computational Economics, Handbooks in Economics Series, Amsterdam: North-Holland.
http://www-personal.umich.edu/~axe/
Glennan, S. (2000). A Model of Models. Philosophy of Science Archives. Retrieved on
August 31, 2008 from: http://philsci-archive.pitt.edu/archive/00001134/00/Models1.4.pdf
Harré, R. (1988). The Philosophies of Science: An Introductory Survey. Oxford:
Oxford University Press. pp. 168-183 (Chapter 6).
Optional:
Aris, R. (1979). Mathematical Modeling Techniques. San Francisco: Pitman Advanced
Publishing Program. pp. 1-38 (Chapters 1 & 2)
Harré, R. (1988). Cognitive Science: A Philosophical Introduction. London:
Sage, pp. 42-56.
Week 2: Introduction to ABMS and NetLogo
We introduce the conceptual basics of ABMS, using NetLogo as the modeling environment.
Reading
Resnick, M. (1999). Turtles, Termites, and Traffic Jams: Explorations in Massively
Parallel Microworlds. MIT Press. (pp. 49–68)
Axelrod, R. and Tesfatsion, L. (2005). A Guide for Newcomers to Agent-based Modeling in the Social Sciences. In K. L. Judd and L. Tesfatsion (Eds.), Handbook of Computational Economics, Vol. 2: Agent-Based Computational Economics, Handbooks in Economics Series, Amsterdam: North-Holland.
http://www-personal.umich.edu/~axe/
Optional:
Gilbert, N. and Terna, P. (1999). How to build and use agent-based
models in social science
Macal, C. M. and North, M.J. (2006). Tutorial On Agent-Based Modeling And
Simulation Part 2: How To Model With Agents. Proceedings of the 2006 Winter
Simulation Conference L. F. Perrone, F. P. Wieland, J. Liu, B. G. Lawson, D. M. Nicol,
and R. M. Fujimoto, eds.
Brown, D.G. 2006. Agent-based models. In H. Geist, Ed. The Earth's Changing Land: An
Encyclopedia of Land-Use and Land-Cover Change. Westport CT: Greenwood Publishing Group, pp.
7-13.
Macy, M.W. and R. Willer (2001). From Factors to Actors:
Computational Sociology and Agent-Based Modeling. Annual Review of
Sociology, Vol. 28, pp. 143-166. (Available from JSTOR)
Lab Session
Getting to know NetLogo
Read NetLogo Tutorial. http://ccl.northwestern.edu/netlogo/
Assignment
Week 3: Segregation
Many social behaviors can be understood as emerging from interactions among
individuals engaged in simple behaviors: residential segregation, cooperation, wars,
social influence, urban growth, etc. In recent years, agent-based models of many of these
phenomena have been developed by social scientists. These three weeks we study some of
these models.
Reading
Schelling, T. C. (1978), Micromotives and Macrobehavior, Norton, New York, NY, pp.
137-157.1
Schelling, T. C. (1969). Models of Segregation. The American Economic Review. 59 (2):
488-493.
Pancs, R. and Vriend, N. J. (2007). Schelling's Spatial Proximity Model
of Segregation Revisited. Journal of Public Economics. 91: 1-24 http://webspace.qmul.ac.uk/nvriend/pub/pubec.pdf
Sander, R., Schreiber, D. and Doherty, J. (2000). A Computational Model of
Housing Segregation. Accessed from: http://www.law.ucla.edu/sander/H_Seg/WPSA_Sander.pdf
Lab Session
We will examine Chris Cook's segregation model:
http://www.econ.iastate.edu/tesfatsi/demos/schelling/schellhp.htm
Or:
Denis Phan's Segregation Model: http://digemer.enst-bretagne.fr/~phan/complexe/schelling.html
Week 4: The Evolution of Cooperation
Reading
Axelrod, Robert (1984), The Evolution
of Cooperation (Chapters 1,2,9). Basic Books Inc., New York,
NY.
Axelrod, Robert (1986), "An Evolutionary Approach to Norms",
American Political Science Review, Vol. 80, pp. 1095-1111 (available from
JSTOR)
Lab Session
Axelrod's tournament
OR Toronto's interactive
tutorial
Week 5: Standing Ovation
Granovetter, Mark (1978), "Threshold Models of Collective Behavior",
American Sociological Review, Vol. 83, pp. 1420-1442 (available from JSTOR)
Miller, John, and Scott E. Page (2004), The Standing
Ovation Problem, Complexity, Vol. 9,
No. 5, May/June, pp. 8-16
Lab Session
A demo from Craig Reynolds boids
website
Assignment
Exercise 2 due
Week 6: The Theory of Simulation and The Object-Oriented Approach
In this session, we draw on Zeigler's theory to review the various formalisms,
specification levels, and observation frames for modeling and simulation
Reading
Zeigler, B. P., H. Praehofer, and T.G. Kim (2000). Introduction to System Modeling Concepts (Chapters 1, 2, 3), Theory of Modeling and Simulation. San Diego, CA, Academic Press.
Weisfeld, M. (2000). The Object-Oriented Thought Process, Indianapolis:Indiana, SAMS
Publishing (Division of Macmillan).
Lab Session
A tutorial on Java programming — e.g., Murphy's
tutorial
Week 7: Introduction to Repast
We introduce Repast, the ABM toolkit developed by Argonne National Lab. In Repast, Grid Agent is the default agent component in simulations with a grid topology such as those involving neighborhood relationships. In this session, we will learn how to create a grid model
Reading
Repast How To's. http://cscs.umich.edu/old/lab/documentation/RePastStuff/repast/repastj/docs/how_to/how_to.html
Case Study: The CarryDrop Model
Lab Session
John Murphy's Repast Tutorial: http://www.u.arizona.edu/~jtmurphy/H2R/HowTo01.htm
Week 8: Modeling in Repast
In this session we will continue our experiments with Repast, in particular some of
its Java capabiities
Reading
Tesfatsion, L. Self-Study Guide for Java-based Repast. http://www.econ.iastate.edu/tesfatsi/repastsg.htm
Week 9: Geospatial Modeling
The GIS model is the default model for creating simulations that take place on a GIS
topology and whose agents can be created from a shape file. A model of urban development
will be examined as an example of how ABM and GIS can be combined to provide useful
outcomes.
Reading
Parker, D.C., etal (2003). Multi-Agent Systems for the Simulation of Land-Use and Land
cover Change: A Review. â¨. Annals of the American Association of
Geographers 93(2).
â¨D.A.Bennett and W. Tang (2006). Modelling adaptive,
spatially aware, and mobile agents: Elk Migration in Yellowstone.
â¨. IJGIS 20(9): 1039-1066.
Lab Session
Urban Growth Model
Week 10: Descriptive Modeling in Brahms
Brahms is a multi-agent systems modeling language and environment, developed and used by NASA in simulating its missions (most recently the Mars Rover). We will have a brief introduction to Brahms and its application to the study of Apollo 12 mission.
Reading
Sierhuis, M. et al. (2003). Modeling and Simulation for Mission Operations Work System Design
Journal of Management Information Systems / Spring 2003, Vol. 19, No. 4, pp.
85–129.
Sierhuis, M. et al. (in print). A multiagent modeling environment for simulating work practice in organizations. Paper submission to the special issue on "Simulating Organisational Processes",
Journal for Simulation Modelling Practice and Theory.
http://www2.sims.berkeley.edu/academics/courses/is2903/s05/papers/Brahms_SMPT_final.pdf
Sierhuis. APOLLO 12 ALSEP Deployment. http://www2.sims.berkeley.edu/academics/courses/is290-3/s05/papers/Apollo_12_ALSEP.pdf
Sierhuis, M. Modeling and Simulating Work Practice: A Method for Work Systems Design
IEEE Intelligent Systems, September/October 2002.
Seah, C. et al. Multi-agent Modeling and Simulation Approach for Design and Analysis of MER
Mission Operations.
http://www2.sims.berkeley.edu/academics/courses/is290-3/s05/papers/SIMCHI05_MER_Brahms_v4.pdf
Lab Session
Apollo 12 case study
Assignment
Exercise 3 due
Week 11: Situated Activity
Human activity takes place in particular social, organizational, and spatio-temporal contexts not in abstract space, and is only understandable within those contexts. In this session, we review some of the dominant theories of activity.
Reading
Engestroèm, Y. (2000). Activity theory as a framework for analyzing and redesigning
Work. Ergonomics, 43(7), 960-974
Hutchins, E. (1995). How a Cockpit Remembers Its Speeds. Cognitive Science
19:265-288 (1995)
Clancey, W.J., (2002). Simulating activities: Relating motives, deliberation, and
attentive coordination. Cognitive Systems Research, 3:471–499
Week 12: Field Study Methods
Models need data collected from the "field" in order to provide realistic insights into systems. In this session we study a number of field study and data collection methods such as contextual inquiry.
Reading
Clancey, W. (2001). Field Science Ethnography: Methods for Systematic Observation on
an Arctic Expedition. Field Methods, Vol. 13, No. 3, 223–243
Clancey, W.J. Observation of Work Practices in Natural Settings
http://www2.sims.berkeley.edu/academics/courses/is290-3/s05/papers/Clancey_Expertise_Handbook_4.pdf
Schreiber, G., et al (1999). Knowledge Engineering Basics (Chapter 2) in, Knowledge
Engineering and Management: The CommonKADS Methodology, The MIT Press.
Hutchins, E. (1995). Welcome Aboard (Chapter 1). Cognition in the Wild. Cambridge, MA,
MIT Press.
Lab Session
Guest Speaker: Dr. Josh Goleberg
A Computational Model of Infant Habituation Using Matlab
Week 13: Overview of ABMS Toolkits
There are numerous toolkits and environments available for ABMS modeling. We review these and examine their capabilities, limitations, and utilities for various applications.
Reading
Tools for Agent-Based Modeling.
http://www.swarm.org/wiki/Tools_for_Agent-Based_Modelling
Week 14: Presentations
Week 15: Presentations
Grading
Individuals or teams of two students will develop an ABM model in a domain of their
choice. Each team will present its model to the seminar at the end of the
semester.
Grades will be based on:
1. Leading class discussion 20%
2. Short Project #1 20% Due week 4
3. Short Project #2 20% Due week 8
4. Term project 40% Due week 14
Project Descriptions
TBD
Resources
Wonderful web resources are available these days on ABM — papers, tutorials,
tools, languages, models (with code and documentation). The following are links to a
subset of these resources.