System Dynamics Modeling Overview

From SystemsWiki

Jump to: navigation, search

System Dynamics modeling was developed at MIT by Jay Wright Forrester in the 1950s. He had a background in computer hardware and feedback control systems based on servo-mechanisms for gun and missile guidance. The language of system dynamics is stocks, flows and feedback loops. This approach adopts the systemic perspective and builds formal computer models to help us make sense of the world. In 2007, Forrester restated the goals of System Dynamics as “Through an appropriate simulation model, one should know the structure causing the problem, should know how the problem is created, should have discovered a high-leverage policy that will alter behaviour, should understand the reasons why the low-leverage policies will fail, should be able to explain how strongly defended policies within the system are actually the cause of troubles, and should be able to argue for better alternative policies.” The argument is often stated in terms of causal loop diagrams, a type of influence diagram. System Dynamics is now considered as a branch of systems science which emphasizes designing better policies and management control structures based on an understanding of dynamic complexity. This explicit handling of time includes attention to sequences of actions, the pattern of forces that influence actions and time lags and delays in information processing feedbacks that affect stability.

Donna Meadows has expressed the key features of the System Dynamics view and approach in her 2008 (posthumous) book "Thinking in Systems: A Primer" Definition of a system A set of elements or parts that is coherently organized and interconnected in a pattern or structure that produces a characteristic set of behaviours, often classified as its function or purpose. Summary of Systems Principles A system is more than the sum of its parts. Many of the interconnections in systems operate through the flow of information. The least obvious part of the system, its function or purpose is often the most crucial determinant of the system’s behaviour. System structure is the source of system behaviour. System behaviour reveals itself as a series of events over time. This pattern of events over time is system behaviour.

A good system dynamics model focuses on explaining a specific problem or situation. The situation is best described by a puzzling trend over time that requires explanation and corrective action. It describes interactions among parts, especially information feedback interactions. All relevant assumptions are made explicit in the structure of the model. It takes an endogenous view of the problem, usually by zooming out on the problem so we have a causally closed boundary involving feedbacks (rather than forcings by exogenous time series). It integrates multiple stakeholder perspectives and its key aim is to design better policies so individuals and groups can learn to take effective action.


Representing the dynamics of systems

In SD we specifically add the dynamic dimension to the systemic perspective, with a focus on feedback and introduce the concept of stocks and flows.

Stocks and Flows

A stock is an accumulation of material or information that has built up in a system over time. A stock is an amount or a level, like the amount or level of water in a bathtub (eg 150 litres) or the number of people with a disease (Prevalence). A flow is material, energy or information that enters or leaves a stock over a period of time. Flows are rates per unit time, like the rate of water flowing into a bath through a tap (eg 2 litres per minute) or the amount of water flowing out of a bath through a drain or plug-hole (say 1 litre per minute) or the number of new cases of a disease per year (incidence rate).

A stock is the memory of the history of changing flows within the system. Stocks allow inflows and outflows to be decoupled and to be independent and temporarily out of balance with each other. Therefore we do not need to assume a steady state of stability, where inflows equal outflows.

Information Feedback and Circular Causation

We perceive states of the world, and we act on this information based on our beliefs or mental models, including our understanding of causes and effects. The logic people use to make decisions (converting information into action) that make sense in one part of a system may not be reasonable or desirable within a broader context or when seen as part of the wider system. So the bounded rationality of each actor in a system may not lead to decisions that further the welfare of the system as a whole. The system dynamics method aims to avoiding these unintended consequences of clinical policy and management interventions due to their feedback effects.

These unintended feedback effects are sometimes described as vicious or virtuous circles or cycles.

Negative Feedback

A feedback loop is a closed chain of causal connections from a stock, through a set of decisions or rules or physical laws or actions that are dependent on the level of the stock and back again through a stock to change the stock. Balancing or negative feedback loops are equilibrating or goal-seeking structures in systems and are both sources of stability and sources of resistance to change.

This explains why large interventions can have small effects, particularly in social systems. This gridlock is called policy resistance. In the human body this resistance to change is described as homeostasis, for example, when blood pressure or blood sugar level is kept within a narrow range despite large fluctuations in food intake and exercise.

Positive Feedback

Conversely small interventions can have large effects. This is described as a leverage point in the systems. A common example is the trim-tab rudder in a boat, where a large rudder is made easier to move by using a small rudder.

These positive feedback effects are called runaway loops or reinforcing feedback. Reinforcing feedback loops are self enhancing leading to exponential growth or to runaway collapses over time. A snowball running down a hill causing an avalanche or collapse of a corporation like Enron into a death spiral are examples, and the point of no return is called a tipping point. Adam Smith wrote about the health and wealth of nations. Does health determine wealth? Or does wealth determine health? Both beliefs are justified. This is an example of reinforcing circular causation. Another lifestyle system of circular causation is succinctly described by Honore Balzac.

“I’d need rest to refresh my brain, and to rest it’s necessary to travel, and to travel one must have money, and in order to get money you have to work…I am in a vicious circle…from which it is impossible to escape.” Primer p30

We will revisit this effect later in Jack Homer’s model of worker burnout.


The information delivered by a feedback loop can only affect future behaviour…things take time

Delays make a system likely to oscillate. Eg health workforce or building boom bust cycles Changing the length of a delay may make a large change in the behaviour of the system

Where there are long delays in feedback loops some sort of foresight is essential. As Wayne Gretsky described his success in ice hockey, “You have to skate to where the puck is going to be.”

Putting it all together

  • Complex behaviours of systems often arise as relative strengths of feedback loops shift, causing first one loop and then another to dominate behaviour. This is described as feedback loop dominance.
  • Accumulations delays and feedback structures cause nonlinear behaviour.
  • Non-linear relationships exist between causes and interventions and effects, where the cause does not produce a proportional effect.
  • System dynamics models explore possible futures and ask “what-if” questions.
  • The usefulness of a model depends on relevance and plausibility.
  • Confidence in the model is based more on structure than coefficient accuracy.


Questions & Comments to Geoff McDonnell
Personal tools