SYSEN 507: Systems Thinking

Overview

This course in Systems Thinking provides you with the tools to understand and describe complex systems and to identify emergent properties, feedback mechanisms, and their effects. Students will understand the difference between systematic and systemic approaches, the pitfalls of reductionism, and the necessity for holistic system understanding and description. They will apply these ideas through a variety of approaches and methodologies including Causal Loop Diagrams, Stock and Flow Models and Rich Pictures.

Syllabus

Course Objectives

The fundamental objective of this course is to help students understand Systems Thinking. Students who complete this course successfully will be able to:

Discriminate between systemic and systematic approaches,
Articulate the role one’s mental model plays in perception,
Describe system behavior over time as a graph,
Construct causal loop diagrams of complex systems,
Recognize system archetypes in social systems,
Translate causal loop diagrams into stock and flow models.

Course Materials

There is no required textbook for this course, all materials you will need are provided either as Course Reserves, links within lesson modules, or as included readings. However, there are a number of excellent books on systems thinking and systems dynamics that you may find useful. I have highlighted some below:

Dennis Sherwood, “Seeing the Forest for the Trees: A Manager’s Guide to Applying Systems Thinking,” Nicholas Brealey Publishing, 2002. ISBN-10: 185788311X
- Focused on using causal loop diagrams in a managerial context.
Jamshid Gharajedaghi, “Systems thinking. Managing Chaos and Complexity,” Butterworth-Heinemann. ISBN-10: 0123859158. Available through University Libraries as an electronic book.
- An excellent treatment of systems thinking centered around the work of Russell Ackoff and his contributions to the field.
John Sterman, “Business Dynamics: Systems Thinking and Modeling for a Complex World,” McGraw-Hill, 2000. ISBN-10: 0071179895
- This is the best book on systems dynamics (Stock and Flow modeling), but can cost over $150.

Grading and Examinations

A grade is given solely on the basis of the instructor’s judgment as to the student’s scholarly attainment (see the Penn State Graduate Degree Programs Bulletin, p. 41). The following grading system applies to graduate students:

“A” (Excellent) indicates exceptional achievement.
“B” (Good) indicates substantial achievement.
“C” (Satisfactory) indicates acceptable but substandard achievement.
“D” (Poor) indicates inadequate achievement and is a failing grade for a graduate student.

Assignment	Quantity	Percentage of Final Grade
Weekly Homework Assignments	7	100%

Grades will be based on the following scale:

A = 95 – 100, A- = 90 – 94, B+ = 87 – 89, B = 84- 86, B- = 80 – 83, C+ = 77 – 79, C = 70 – 76, D = 60 – 69, and F = below 60.

Assignments

There will be seven (7) assignments – one due each week of the course.

Discussions

There will be several activities and discussion forums associated with each lesson. Some of these are ungraded, but you are expected to complete them, as they will help you comprehend the concepts presented in the course.

University Policies

Be sure to review the University Policies and Resources Links to an external site. which include important information regarding academic integrity, student disability resources, educational equity, counseling services, and technical requirements.

*Syllabus subject to change

Sample Lesson

Stocks & Flows
Stock & Flow Model
Causal Loop Diagram

Stocks & Flows

In order to simulate a model, we need to define the nature of the variables, their units, scales, and how they affect each other. The first step is to determine which variables are stocks and which are flows.

Stocks accumulate (or deplete) over time and can be measured at an instant in time.
Flows increase (flow into) or decrease stocks (flow out of) stocks and can be measured only over a period of time. Hence, flows are rates.

Select the headers below to test your understanding of stocks and flows.

Please login.

Stock & Flow Model

Let’s take a look at the following example.

From the description provided below, we can formulate a stock and flow model of the fish population. The model was created using Vensim.

Details for this Example

Lake Sunbeemtee is a small lake nestled in the Pocono Mountains in Pennsylvania. Today, approximately 120,000 fish, mainly wall-eyed pike and bass, swim in the deep and shallow waters of Lake Sunbeemtee. The lake is a mid-sized lake, with a surface area of 20,000 square meters and an average depth of 10 meters. Although female fish lay several thousand eggs a year, on average only three of each female’s eggs will hatch successfully each year. When the density of fish is at an optimum level, as it is now, the fish live eight months on average. The beautiful bass that give proud fishermen a reason to boast are typically three to four years old, but those fish are few and far between. The average lifetime of the fish is short because most fish die young; big fish devour the little ones. The average lifetime of the fish, however, varies with the crowding in the lake because of the competition for food and other nutrients. For example, if the lake were twice as dense in fish, the average lifetime of the fish would drop down to approximately four and a half months. If the lake were half as dense in fish, the average lifetime of the fish would rise to nine months.

Building a Causal Loop Diagram

To draw a causal loop diagram, you should begin by drawing behavior over time (BOT) graphs. Here’s how to create BOTs:

Formulate the core problem
Tell the story of the problem behavior
Choose the key variables you want to work with
Name the variables precisely and don’t forget to:
- Use nouns or noun phrases
- Be sure our variable name fits into phrases such as “level of” or “size of”
- Use a neutral or positive term whenever possible
- Include intangible variables, such as morale, where appropriate, as well as tangible variables
Graph the variables’ behavior over time
Hypothesize about how the variables might be interrelated

There are several methods for moving beyond these steps to draw a CLD. Let’s start off with a new case and then follow the steps up to drawing a BOT graph. Then, we’ll describe two ways to use the graph to create a causal loop diagram.

Case study: The Case of the Collapsing Banks

Throughout its history, the United States has suffered periodic rashes of bank failures. During these episodes, depositors seemed to lose confidence in a bank and began withdrawing their funds. If word of this worry got around, more and more depositors lost confidence, and more and more funds were withdrawn from banks. Eventually, the volume of these withdrawals threatened the solvency of the bank, and when bank funds fell too low, the bank failed. Worse yet, the failure of one bank could trigger a rash of other bank failures. Over the course of several months, depositors at other banks got nervous when they heard about the failure of the first bank, whether they had any reason to worry about their own banks or not. So, they withdrew their funds from their banks, and, if funds got low enough, these banks, too, lost solvency and failed.

Here’s how you might produce a BOT graph of this story’s variables:

Formulate the Problem
- To formulate the problem, ask yourself: “What was going on?”. As we see it, the problem is that many banks were failing over the course of several months.
Tell the Story
- The story, in brief, is that as depositors lost confidence in their banks, they withdrew their funds, and the banks began failing in a kind of domino effect. As more and more banks failed, depositors lost even more confidence and withdrew yet more funds. Then, even more banks failed as shown in Figure 2.10.
Choose Your Key Variables and Name Them Precisely
- The significant variables that we detect in the story are
  - Bank failures
  - Bank solvency
  - Funds withdrawals
  - Depositors’ confidence
Graph the Key Variables’ Behavior Over Time
- “The Bank Story BOT Graph,” shows how we see they key variables behaving over time