Invitation to Research SYSTEMS THINKING Roger Clarke, Xamax Consultancy, Canberra Visiting Professor, CSIS, Uni of Hong Kong Visiting Fellow, Australian National University http://www.anu.edu.au/people/Roger.Clarke/... ...Res /15-Sys.ppt ebs,

1
Invitation to ResearchSYSTEMS THINKINGRoger
Clarke, Xamax Consultancy, CanberraVisiting
Professor, CSIS, Uni of Hong Kong Visiting
Fellow, Australian National University
http//www.anu.edu.au/people/Roger.Clarke/......R
es /15-Sys.pptebs, 16-20 January 2003

2
Agenda

• Systems
• Models
• Cybernetics
• Complexity of
• System
• Model
• Behaviour

3
A System

• A set of interacting Real-World Entities
• Interacts with its Environment across its
  Boundary
• Entities within a System
• gather inputs via Sensors
• transmit outputs via Effectors
• interact with one another via Interfaces
• Systems may
• exist at the same level (e.g. an mammals
  cardio-vascular and digestive systems)
• be related hierarchically (e.g. corpuscles within
  blood within the cardio-vascular system)

4
System Characteristics

• Degree of Closedness/Opennessto external Stimuli
  or Triggers
• EntropyThe tendency to lose energy and order
• NegentropyThe importation of energy to sustain
  order
• Unguidedness cf. Purposiveness
• Adaptiveness/LearningThe ability to change as a
  result of stimuli

5
A Model

• An Abstract-World Representationof a Real-World
  System
• Iconic or Symbolic/Mathematical
• Capable of being
• viewed (e.g. a map
• manipulated (e.g. a model aircraft in a
  wind-tunnel)
• in ways that the System cannot be
• Necessarily (greatly) simplified
• The only reliable model of a real-world system
  is the system itself (Pratchett)

6
Categories of Models

• Deterministic Models Automata
• Computable by Analytical Methods
• Too Complex to Compute, hence Requiring Numerical
  Methods
• Probabilistic Models
• Non-Deterministic / Stochastic Models
• Entities exercisingSelf-Determination / Free
  Wille.g. Humans and Organisations

7
Simple Process Model
8
Process Model with Controller

9
Cybernetics

• Control Loop
• Inputs or Process Parameters changedas a result
  of measurement of Outputs
• FeedbackPositive / Reinforcing or Negative /
  Retardent
• Static Equilibrium / Steady-State cf.Dynamic
  Equilibrium / Homeostasis
• Delays, Control Limits, Tolerances
• gt Hunting / Hysteresis, Over-Correction
• Feedforward
• Tiered Control Loops exhibit Complex Behaviour

10
Levels of Discourse About Systems (Boulding)

• Framework Anatomy
• Balanced Dynamic Clockwork
• 1st-Order Control Thermostat
• Open, Self-Maintaining Cell (in Biology)
• Genetic-Societal Plant
• Self-Aware, Purposive Animal ( Orgn?)
• Self-Conscious Human
• ...