Week 5 - Module 4 (Lecture 17)
System Thinking
Understanding the system
Integrating knowledge from different disciplines
Understanding the interactions among the structure and how it influences behaviour
Understand the relationship between system structure and behaviour then find points you can intervene whilst minimising unintended consequences
Examples
Transmission dynamics for Human Leptospirosis
click to edit
Levels of System Thinking
Events - Symptoms
Patterns - Behaviour over time
Systemic Structures -what the systems look like
Mental Models - Understanding of how system works
System Structure - Feedback
Dynamics controlled by interactions between components
Dynamics arise from the interaction of just 2 feedback loops - positive (reinforcing) and negative (or balancing)
Dynamics arise from the interaction of multiple loops
Casual Loop Diagrams
Cause increases whilst effect increases = Plus
Cause increases and effect decreases = Minus
Cause decreases and effect decreases = Plus
Cause decreases and effect increases = Minus
Loops
Reinforcing
Have zero or an even number of minus polarities
Balancing
Have an odd number of minus polarities
Common modes of behaviour in dynamic systems
Exponential Growth
Goal seeking
Oscillation
Arises from reinforcing loops
reinforcing loops can also generate exponential decline
Caused by balancing loops
Balancing loops counteract change
Balancing loops, reduces discrepancies and desires the goal
Also caused by balancing loops, but delay means the state of the system overshoots and undershoots the goal
Time delay cause the corrective actions to continue after the goal is reached, adjusts too much
S - shaped growth
There is a reinforcing and balancing loop that controls system behaviour
Balancing loops slows growth
Can be with delays in the balancing loop = overshoot and undershoots the goal
Overshoot and collapse - erosion of carrying capacity creates a second balancing loop that limits growth
Inferring Structure from behaviour
Link between structure and behaviour is important because it allows you to infer the structure of a system from its behaviour
EXAMPLE = system exhibits exponential growth then there must be a dominant reinforcing loop
Key Lessons
When managing any system, understand the relationship between the structure of your system and its behaviour
Be multidisciplinary
Use your understanding to carefully target your interventions
Multiple interventions may be needed