Systems Dynamics

Dynamic stock and flow diagram of New product adoption model. Wikipedia

The emphasis within the systems and complexity paradigm is on local interactions and how this gives rise to emergent macro level phenomena of the state to the whole organization. System dynamics is one modeling language from systems theory that is used in the management of large complex systems in order to try and capture these local causal links and how their interactions over time give rise to long-term patterns of behavior. System dynamics is a methodology and mathematical modeling technique to frame, understand, and discuss complex issues and problems. Originally developed in the 1950s to help managers improve their understanding of industrial processes, it is now being used throughout the public and private sector for dealing with large complex projects, such as for macroeconomic analyses, military interventions or environmental policy modeling.

System dynamics is again a holistic approach in nature. In our analysis, we often focus on discrete events, like watching the news every night where the presenter simply describes what happened. But of course, if we really want to understand something like why a dictatorship fell, or why the price of biogas is going up, or why a large organization like Nokia was incapable of responding to the changing market, we can’t simply do this by describing what happens. We need to understand the dynamics behind these events; what is really driving the system. And this is what systems thinking is about, trying to get behind events to see what is truly causing them.

Systems Behaviour

All complex organizations have a typical behaviour. We see recurring patterns, like with the example of Nokia failing to innovate and adapt. We see this with many large organizations that are incapable of adapting to changes and that eventually get disrupted. These typical recurring behaviours within a system are called system archetypes.

System archetypes are reoccurring patterns of behavior within a system. We see a child playing away happily and then some small event happens and they start crying. The first time we see this happen, we would think that it was the small event that caused the child to cry, but if we were the child’s parent and saw this pattern of behavior happening time after time, we would start to realize that in fact, it is not so much the event itself but some underlying dynamic that is creating this behaviour.

Archetypes

Thus, systems exhibit certain behavior. But under this, is a deep system structure; a set of underlying dynamics that are really driving the system and it is these drivers that system dynamics tries to capture. In large complex systems, it is these dynamics that hold the system within a certain configuration or state. Therefore, it is necessary to understand these dynamics before we can try and effect the system. If we don’t understand the underlying dynamics, we will likely get unintended consequences or no effect at all. For example, we can think about the Iraq war. The Iraq invasion was a direct intervention into a very complex system, that of the whole social, cultural and political system of a nation to which the invaders had very little understanding of the dynamics. By altering those dynamics that had held the system in its previous state, it essentially went out of control and eventually collapsed into chaos. Sufficed to say, you can’t really manage a system until you understand its dynamics and by understanding those dynamics you can find the leverage points that enable effective changes in that system.

Interdependency

Complex systems display high levels of interdependence among their members

The basic idea of systems thinking is that of interdependence. For many people, this is what defines the idea of a system – the fact that the parts are interdependent. Therefore, every action within the organization is seen to trigger a reaction. In system dynamics, this reaction is called feedback. In these highly interconnected systems, nothing exists in a vacuum, every action feeds back to its source over time. The fuel exhaust coming out of your car engine may look like it simply disappears having no effect. But that is just because you are isolating one component in a much larger system. Of course, sooner or later, all that exhaust feeds back to affect all of us drivers. And that is the nature of managing complex systems. You have to look at the whole in order to see these feedback loops that are driving it.

Feedback Loops

Feedback loops describe a state of interdependence between two or more elements within a system and there are just two types of feedback. Positive feedback which is a self-reinforcing loop and negative feedback which is a balancing loop.

Reinforcing feedback

Reinforcing feedback (or amplifying feedback) accelerates the given trend of a process. If the trend is ascending, the reinforcing (positive) feedback will accelerate the growth. If the trend is descending, it will accelerate the decline. For example, the falling of an avalanche is a self-reinforcing feedback process. The more material that falls, the more momentum it adds to the avalanche, thus dislodging more material which feeds back to create greater momentum and so on.

Another example might be minority group prejudice within a society. Prejudice tends to generate discrimination against the minority, which tends to limit their opportunities, which feeds into limit their achievements, which in turn feeds back to validate the majority that the minority are not as good at them. Thus, once again reinforcing the chain of events. This downward spiral that may be caused by a self-reinforcing feedback loop is called a vicious cycle. But we can also have virtuous cycles. For example, having a good reputation as an organization means others will talk about you favorably, which will feedback to motivate your organization to live up to that expectation and excel, which again will affect people’s perception favorably and so on.

Balancing feedback

Balancing feedback, or stabilizing feedback, involves two or more elements that are counterbalancing each other. If we get more of one, this creates less of another, which feeds back to reduce the first. For example, the market mechanism that balances supply and demand is a negative feedback loop. If supply goes up, the price goes down, which promotes more buying which feeds back to reduce the stock thus bring it back to a balanced equilibrium. Maintaining your balance while riding a bicycle is an example of a negative feedback loop. The more you move off in one direction, the strong you react by pulling yourself back on course. Thus, we can see how negative feedback works as a control system to maintain and regulate the organization within a set of desired parameters.

Whereas positive feedback tends to lead to instability via exponential growth, oscillation or chaotic behavior, negative feedback generally promotes stability. Negative feedback tends to promote a settling to equilibrium and reduces the effects of perturbations. Negative feedback loops in which just the right amount of correction is applied – with optimum timing – can be very stable, accurate, and responsive.

Dynamics

These feedback loops then create a certain reoccurring pattern to the system as it changes over time. For example, most feedback does not occur instantly, especially when we are dealing with a large organization. There is instead a time delay. So if we have a system dominated by negative feedback with time delay, we get an oscillatory pattern over time as we have one effect that takes it in one direction before the counterbalancing feedback gradually dampens it down and brings it back in the other direction. We might see this dynamic in countercyclical fiscal policy by a government as it tries to counterbalance the business cycle.

Inversely, positive reinforcing feedback loops create a dynamic of rapid exponential growth or decay that often leads to collapse when some maximum level is reached. The so-called tragedy of the commons may be an example of this archetype. Agents use the common limited resource to profit individually. As the use of the resource is not controlled, the agents would like to continuously raise their personal benefits. The resource is therefore used more and more and the revenue to the agents starts to decrease as the overall resource becomes depleted. The more an actor sees others over-exploiting the resource, the more they are likely to intensify their activity knowing that it will be gone soon. The agents intensify their exploitation until the resource is completely used up or seriously damaged leading to collapse. A real instance of this can be seen in overfishing.

Leverage Points

By understanding these dynamics that hold the system in a particular configuration or generate a particular behavior over time, we can begin to identify points within the linkage of effects that might alter its behavior towards the desirable outcome. Traditionally, we focus on changing things by changing the components in the system. If a company is doing badly, we fire the CEO, but this is the lowest point of leverage. By understanding the dynamics, we can see that it is usually not the members of the organization that are creating systemic failure. It is the system itself, the way it is arranged, that creates systemic failures. If you are in systemic political deadlock, electing a new president is unlikely to change much, because you are simply placing a component in a broader system that is setup to create the same behavior again.

It is only at this systems level that you have the opportunity to really change things through leverage points. These “leverage points” are places within a complex system, such as a corporation, an economy, a city, or an ecosystem, where a small shift in one thing can produce big changes in everything. The founder of system dynamics talked about complex systems as being counterintuitive and leverage points being likewise. In our traditional approach, we focus on the component parts to an organization. We think it is the president of a country that has the greatest leverage. In fact, the component parts are the lowest point of leverage within the system. The higher points of leverage are in changing the connections within the organization – changing the flow of information and, most of all, changing the model that the organization uses to interpret that information.

For example, let’s think about trying to get people to smoke less. We can try to directly affect them by increasing the cost of cigarettes, but also we can indirectly influence them by altering the feedback loops. Most people who smoke try to ignore or downplay the negative effect it is having on their body, simply hoping that they don’t get cancer. This is an open loop system with negative externalities. They are consuming the cigarette and shifting the negative effects away by denying them in some way. Now, when it comes to managing this system, instead of focusing on the parts, we try to alter this feedback loop. We place a big graphic image or sign on the face of the pack that connects them with the consequences of their actions every time they smoke. Thus, trying to close the feedback loop to reduce their consumption.

As another example of a counterintuitive leverage point, we could cite a recent piece of research done analyzing the social network of a company in Hungary. The analysis showed that the network had one dominant node in it, which had a much higher level of connectivity than all of the senior executives. But this node was not the CEO. In fact, it was an external health practitioner who, during his job, got to travel around the different departments and make contact with almost all the employees, thus serving as an important point of information exchange within the organization.

Paradigm shifts, such as sustainability, are the greatest leverage points for an organization to achieve a certain outcome

But, ultimately, the highest leverage point in any organization is in changing the model that they use to interpret events. Changing the paradigm can change the whole organization’s understanding of itself and its function, which can ultimately result in a complete transformation. As an example, we can think about the rise of the paradigm of sustainability over the past few decades. Before this change in paradigm, talking about the environment, trying to conserve nature, or effect any positive environment impact was largely a lost cause. Environmental concerns were marginalized and generally bulldozed by economic interests. But, given this paradigm shift, we now see sustainability moving to central stage as it becomes part of what organizations are and do. Sustainability is now starting to be seen as part of the marketing, value proposition and competitive advantage of many organizations. This change in paradigm is then generating real outcomes that work and change the overall macro state to this complex organization. As the saying goes, there is nothing as practical as a good theory, and the paradigms and models an organization uses are by far the highest leverage point for effecting the overall state of the organization.

2017-08-15T15:24:12+00:00