GEORGE WASHINGTON UNIVERSITY
 SCHOOL OF ENGINEERING AND APPLIED SCIENCE

 EMSE135/235 - Systems Thinking and Policy Modeling

General Information
Course Description and Objectives
Text and Materials
Requirements and Grading
Schedule
Projects
Other Resources 

GENERAL INFORMATION

Instructor:            John H. Saunders, Ph.D.,  Work #: (202) 685-2078
E-mail:                 jsaunders@erols.com or saunders@ndu.edu 

COURSE DESCRIPTION and OBJECTIVES

Stock-flow analysis of feedback systems presented for policy analysis and management. System dynamics; principles of systems employed to structure the problem solving process. Problems and case studies solved using microcomputers.
Upon successful completion of this course, the student should be able to:
1. Model, both mathematically and visually, concrete and abstract dynamic structures of a system environment.

2. Provide a foundation for the consequences of feedback, time delay, and non-linear activity in problem analysis and synthesis.

3. Develop and test the influence of endogenous and exogenous policy change variables upon environmental performance.

4. Provide cases where stock and flow analysis has provided unique insight into major issues.
 

REQUIRED TEXTS AND MATERIALS
1) Clark, Rolf. System Dynamics and Modeling. Operations Research Society of America. 1990. Modified for online usage 1998. (SD&M)
 Chapter 1  Chapter 2   [Incomplete Chapter 3  Chapter 4   Chapter 5  Chapter 6]
2) An Introduction to Systems Thinking. 2001. High Performing Systems.(IST)

3) ithink Strategy software v 8.0 from High Performing Systems. Manuals packaged with the software include: Getting Started with the ithink software and Technical Documentation, both 2001. (GS) Purchase of the software is not required. It is available in the lab, but must be used at that location.

4) Link to Models

Other current articles may be assigned throughout the semester.
 

COURSE REQUIREMENTS AND GRADING

1. Examination (25%):  There will be an in-class, closed book mid-term examination to test the studentís understanding of the basic concepts discussed early in the course. Format includes multiple choice questinos, causal loop diagramming, analysis and creation of stock and flow diagrams.

2. Individual Project (30%): Each student will create a system dynamics model in an area of their choosing. The model should be the studentís own work. Students may wish to refer to past International System Dynamics Conference Proceedings on the System Dynamics Society Homepage or to Prior Student Projects to get a feel for model domains.

3. Group Project (25%): Students will be assigned to a small group to create and exercise a joint model. See the following detail project description. A joint grade will be assigned to all members of the group. Additionally a peer evaluation will be utilized to measure and evaluate group cooperation. Individual grades may be adjusted accordingly.

4. Homework (10%): Two homework assignments will be completed and submitted. The assignments will be graded to insure adequate progress in absorption of the material. Students should work alone on these assignments. Homework Assignment 1  Homework Assignment 2

5. Class Participation (10%):  A great deal of the graduate learning experience is gained through a discussion of shared experiences and methods for improving quality and processes. In-class exercises and discussion will therefore be evaluated.  The criteria (from most important to least important) for judging a studentís contribution to the discussions are quality (well thought out, relevant to the topic), clarity, and frequency.

The course grade of "B" represents the benchmark.  It indicates that the student has fulfilled all course requirements and demonstrated competency in the subject matter of the course.  Only those students who fully meet this standard and who demonstrate exceptional comprehension of the course subject matter, merit an "A".  Students who do not meet the benchmark standard of competency will earn the course grade of "C".  In those cases where there is substantial failure, the student will earn an "F". Plus and Minus grades are awarded.
 
 
 
 
 
 
 


SCHEDULE
 
Session/Date General Topic / Links
Readings / Assignments Due 
1
Introduction, Course Review, Overview of ST, SD & the MFS
-
2
 
Systems and Causal Loops IST:Preface&Chpt1; SD&M: Chpt 1
Kirkwood: Chapter 1
3 Systems and Causal Loops Causal Loops Problem Set
4 Stocks and Flows; Lab 1  IST:Chpt 2; SD&M: Chpt 2
 5 Building Structures; Lab 2 IST:Chpts 3,4;  SD&M: Chpt 3; 
GS: Hand-in Clients Model p 17-33
In-class exercise
6 Functions; Lab 3 Technical Documentation: Chpt 7
7 Group Project Presentations Group Project Papers and Models
8
Mid Term Examination  1 page individual project description
9
Model Validation; Lab Session  Forrester & Senge; Sterman;
10
Optimization / Forecasting 
-
11
Guest Speaker
 -
12
Survey of System Dynamics & Mgmt Flight Simulator Applications  Saunders
13 
Connections with Chaos
 -
14 Individual Presentations

Individual Project Papers and Models 

     



PROJECTS


Individual Modeling Project

The individual project consists of a unique systems study, further manifested through a dynamic model created in ithink. Students should create an accompanying 3-5 page paper which provides a description of the variables, explains the context for the model and talks to discoveries made while exercising the model. At the eighth class period each student should submit a brief one page project description of the model they plan to build. An example follows below:

Model: USAID Technology Effectiveness

Units of Analysis: Dollars, Numeric Effectiveness Factor (Interval Scale)

Sectors to be included: U.S. Congress, USAID Senior Management, USAID Technology Management

Time Period: 20 years in 1 year intervals

Example Output Variables: Ratio of Effectiveness to Prior Year Funding; Empowerment Utility

To Study the Effects of: Delays and Effectivness in increasing/decreasing budget dollars allocated to information technology.

A simple causal loop diagram or stock and flow sketch should be included.

Final Deliverables

The final project will be presented to the class in Session 14. The final ithink model and written 3-5 page paper are due at that time. The student will demonstrate and discuss the model in class and provide each instructor and each classmate with a 1 page description (diagram printout also very helpful) of the model.

Students should refer to prior student models (available from the instructor) to get a feel for the depth and complexity that is expected. A typical student model would have about 4 - 12 stocks and 15 - 30 converters. Real world models typically have hundreds to thousands of stocks and converters. These prior student projects may provide some aid in understanding appropriate modeling topics.



Group Causal Loop Diagram (CLD) Modeling Project
Small groups will be formed to produce an causal loop diagram which reflects the dynamics surrounding a current event or a public process. The group should decide collectively on the subject area and then find data or expert opinion to substantiate their model. An accompanying  paper should describe the model as well as provide a critique of the CLD method as it relates to the area of study and to the diagram (e.g. what is missing or perhaps misleading?) The group will present the results of their work to the class.

Examples:

The group should download the Vensim software from Ventana Systems to draw the causal loop diagrams. Using your internet browser go to Ventana Systems and download the Vensim PLE demonstration program (1.177 MB) and users guide (2.383 MB in Adobe Acrobat Reader format - 298 pages). The URL is http://www.vensim.com/download.html. The appropriate section to follow in the Vensim Users guide is chapter 4, pages 41-62.

Deliverables (due on evening of presentation - hardcopy and softcopy):

     a) Model - completed model with minimum 10 activities/variables and 4 loops.
     b) Paper - minimum 5 page paper providing model foundations and conclusions.
     c) Presentation - 20-30 minute in-class.
     d) Handout - single page for all class members on the substance of the model.



OTHER RESOURCES

Internet Links:

MIT -  http://sysdyn.mit.edu/sd-group/home.html
Tom Fiddamans Model Library - http://home.earthlink.net/~tomfid/models/models.html
London Business School -  http://pluto.lbs.lon.ac.uk/sysdyn
System Dynamics Society -  http://www.albany.edu/cpr/sds/
LISTSERV - e-mail to majordomo@world.std.com containing the following text subscribe system-dynamics end
Gene Bellingerís home page - http://www.radix.net/~crbnblu/systems/sim/simulation.html
Arizona State University Professor Craig Kirkwood's pages - http://www.public.asu.edu/~kirkwood/sysdyn/SDRes.htm
Books in Systems Thinking and System Dynamics:
Forrester, Jay. Industrial Dynamics. Productivity Press. Portland OR. 1961.

Hannon, B. and Ruth, M.

Modeling Dynamic Biological Systems. Springer Verlag. 1997.
Modeling Dynamic Economic Systems. Springer Verlag 1997.
Dynamic Modeling. Springer Verlag.  1994.
Hargrove, James L. Dynamic Modeling in the Health Sciences, Springer Verlag. 1998.

Morecroft & Sterman, eds. Modeling for Learning Organizations. Productivity Press. Portland OR. 1994.

Richardson, G., ed. Modelling for Management. Dartmouth Publishing Company. Brookfield, VT. 1997.

Roberts, E.B.  Managerial Applications of System Dynamics. Productivity Press. 1981.

Senge, P. The Fifth Discipline Fieldbook. Doubleday. New York NY. 1994.

Periodicals
System Dynamics Review, Wiley & Sons.
The Systems Thinker. Pegasus Communications.
Readings

Forrester, Jay and Senge, Peter. Tests for Building Confidence in System Dynamics Models. TIMS Studies in Management Sciences 14 (1980) 209-228.

Kirkwood, Craig. ďSystem Behavior and Causal Loop DiagramsĒ, Chapter 1 in Integrated Business Process Analysis. Unpublished monograph. 1996. Available at  http://www.public.asu.edu/~kirkwood/sysdyn/SDIntro/SDIntro.htm

Sterman, J. D. 1984. Appropriate Summary Statistics for Evaluating the Historical Fit of System Dynamics Models. Dynamica, 10 (Winter), 51-66.