Systems Thinking and Causal Loop Diagrams

Uncovering System Structures: Building Causal Loops

• Causal loop diagrams (CLDs) are used to explore dynamic interrelationships among variables and identify systemic structures.
• CLDs allow us to hypothesize about solutions to problems and test them in a risk-free environment.

Systems Thinking

• Systems thinking consists of identifying feedback processes and dynamics that determine system behavior.
• Systems can be modeled using reinforcing (amplifying) processes, balancing processes, and delays.
• Reinforcing processes amplify small actions, creating accelerating growth or decline.
• Balancing processes stabilize the system, operating whenever there is goal-oriented behavior.
• Delays are periods of time between causes and effects.

Systems Thinking Levels of Understanding

• Event level: focus on immediate responses to problems.
• Pattern level: identify trends and changes in events over time.
• Structure level: examine the underlying causes of patterns and events.

Seeing the Structure

• The structure of a system is often hidden, like the submerged portion of an iceberg.
• We can see events and patterns, but not the underlying structure that causes them.
• To see the structure, we must ask questions like "What would explain these patterns?" and look for causal connections.

Causal Loop Construction

• Good loops are the result of understanding the story and the insightful conversation between the mind and the diagram.
• Start by writing down variables, using language from the story.
• Draw arrows and label them as S (same direction) or O (opposite direction).
• Connect links into a loop and label it as R (reinforcing) or B (balancing).
• Expect to go through many iterations to refine the loop.

Variables

• A variable is an element whose value changes during the time period of interest.
• Variables have behavior that can be measured over time.
• Choose brief, one-to-two-word descriptions for each variable.
• Avoid using adverbs or lengthy adjectives.

Systems Theory Principles

• System structure causes its behavior.
• System interrelationships cause their own crises.
• There are no villains.
• Understanding structural interrelationships is necessary to understanding system behavior.
• Changing system output requires changing the system structure.

Applications of Systems Thinking

• Developing a systems perspective: there are multiple levels of explanation for any complex situation.
• Creating systems models: simplified diagrams that capture the essential dynamic complexity of the system.
• Systems models are constructed using combinations of reinforcing loops, balancing loops, and delays.

Techniques for Building a System Model

• Use generic system models as a template.
• Identify problem symptoms and look for patterns of behavior.
• Use the "multiple why" process to identify the causes underlying these recurring problem symptoms.
• Look for structural relationships creating these problem symptoms.
• Use the systemic structure level of explanation to identify the underlying causes of patterns and events.### Causal Loop Diagrams (CLDs)
• A CLD is a tool used to identify and analyze complex systems and behaviors
• It consists of two or more variables connected by links, which form a feedback loop
• The links have signs (s or o) that convey information about the relationship between variables

Reinforcing Loops

• A reinforcing loop is a feedback loop that amplifies or accelerates a change in a system
• It can be seen as the engine of growth and collapse
• Characteristics:
  • Compounds change in one direction with even more change in that direction
  • Accelerates movement in a particular direction
  • Can create virtuous or vicious cycles
  • Designated with an "R" in a CLD
• Example: Employee-supervisor reinforcing loop, where support leads to improved performance, which in turn leads to more support

Balancing Loops

• A balancing loop is a feedback loop that tries to bring a system to a desired state and keep it there
• It resists change in one direction by producing change in the opposite direction
• Characteristics:
  • Goal-seeking behavior
  • Tries to bring conditions into equilibrium
  • Designated with a "B" in a CLD
• Example: Inventory control balancing loop, where the system maintains a desired inventory level by adjusting the actual level whenever there are too many or too few parts

Delays

• Delays are a critical component of systems and can lead to misperception
• Types of delays:
  • Perceiving feedback
  • Measuring results
  • Deciding how to respond to results
  • Implementing solutions
• Delays can cause problems, such as overcontrolling or excessive actions that create system instability

Building a Causal Loop Diagram

• Steps to build a CLD:
  1. Formulate the core problem
  2. Tell the story of the problem behavior
  3. Choose the key variables
  4. Name the variables precisely
  5. Graph the variables' behavior over time
  6. Hypothesize about how the variables might be interrelated
• Three methods to build a CLD:
  1. Begin at the beginning of the story
  2. Work backward from the problem symptom
  3. Go back and forth through the sequence of the story

Case Studies

• Case of the collapsing banks: depositors' loss of confidence leads to withdrawals, which leads to decreasing solvency and eventually bank failures

• Case of the sugar-free cereal: an increase in demand leads to a delay in production, which leads to a price increase, and eventually an oversupply of cereals### Building Causal Loop Diagrams

• Identifying variables, graphing their behavior over time, and describing their causal relationships can bring out diverse knowledge and perspectives within a group.

• The "back and forth" method is time-consuming but valuable when you don't have full knowledge of the system.

• The key is to stick with tracing a loop completely through before diverging onto other branches of the diagram.

Guidelines for Building Loop Diagrams

• Use links and arrows to show the direction of variables' cause-and-effect relationships.
• Mark links with "s" for same or "o" for opposite to show the nature of the link.
• Label the center of every loop with either "R" for "reinforcing" or "B" for "balancing".

Grouping and Unpacking Variables

• Lump multiple consequences of a variable into one term while finishing the rest of the loop.
• Unpack the consequences later when exploring specific strategies.

Showing Short-term and Long-term Consequences

• Add new loops or links to show increasingly longer-term consequences or side-effects.
• Distinguish between short-term and long-term consequences, e.g., the effect of stress on alcohol consumption.

Clarifying Complex Connections

• Redefine variables or insert intermediate variables to clarify unclear connections between variables.
• For example, show how increased demand raises production pressure, leading to reduced quality.

Checking the Reasoning Behind a Causal Loop Diagram

• Go around the loop and tell the story depicted by the links and the "s"s and "o"s.
• Ensure the story makes sense and the connections are logical.

Interpreting Multiloop Diagrams

• Start by tracing through one loop, usually the one containing the variables from the first part of the story.
• Read subsequent loops in the diagram, choosing the variable in the central loop that first branches away from the central loop.