Understanding Complex Systems? Start with System Dynamics

Complex systems require a shift of perspective from event-based to systems-based thinking.

Photo by Zhijian Lyu on Unsplash

When we talk about society, we cannot just ignore the interactions among the different groups of people and how they behave according to the law and regulations that put an order in our chaotic world. In development management, we were taught to think in causal loops — in systems, to be specific. As an engineer, thinking in circles is the most bizarre thing, especially when we were taught that machines could be broken down and analyzed by component. The approach is popularly known in undergrad as Finite Element Analysis (FEA), or in some fields, mean-field theory.

FEA works for simple cases, where components can be isolated from the rest, like parts of a chemical reactor. However, for complicated systems, where members cannot be completely isolated, and interdependencies exist among components, like in chemical plants where each piece of equipment is part of multiple process lines, the FEA approach may not work as one may expect.

The interdependence or co-dependence of components in a system requires us to adopt a new thinking mode when analyzing the problem. By mapping components in causal loops, we unravel relationships we initially thought do not exist.

Systems Thinking. It shows how a system perspective on issues allows us to break down problems into structures and mental models , which is not feasible with an event-based/ FEA approach—photos from Google.

For the example below, we realize that constructing wastewater treatment plants to restore river biodiversity only comprise a small part of the solution. Through a system analysis, we uncover that “public awareness (of the mining complex)” would pressure the company to reduce dumped tailings, which consequently improves biodiversity. Moreover, “public awareness (of wastewater treatment plants)” would also drive the flow of capital towards the construction of these facilities. What’s more interesting, there is a time lag (or delay) in public awareness, i.e., the public only noticed the effects of mining companies and treatment plants because river biodiversity started to deteriorate.

Example of a Complex System. Illustrates how understanding systems dynamics would allow us to position solutions better in the entire system. Each arrow represents a cause-effect relationship with signs signifying the correlation, i.e., ‘+’ and ‘-’ for positively and negatively correlated items, respectively.

The causal loop also suggests that over time, as the issue is resolved, we could expect less effort from the mining company to reduce its tailings and waste due to low public clamor, reviving the problem all over again — the cycle continues!

Systems dynamics (or thinking, I use them interchangeably) is a crucial part of a wider branch of Complex Science, which is about studying complex systems. On the other hand, complex systems refer to systems with many moving and interdependent parts — similar to the example above — resulting in new behaviors unexpected from individual parts (emergence).

Dynamic stock and flow diagram of the model. New product adoption (model from an article by John Sterman 2001). By Patrhoue — Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=7249775

Through my examples, I have illustrated three properties of complex systems: interactions (how component X relates to component B), dynamics (how the state of each component changes over time, in response to the environment, and to its related features), and emergence (how the system behaves differently as a whole). Let me just add one more property that I personally find most interesting: Adaptation and Self-Organization.

Swarn Behaviors. Birds flocking, an example of self-organization in biology. By Christoffer A Rasmussen (Rasmussen29892 at da.wikipedia) — Own work, Public Domain, https://commons.wikimedia.org/w/index.php?curid=6736876

Adaptation refers to the ability of a system to evolve and become more complex over time. In the river biodiversity example, we can interpret “public awareness” as a form of adaptation. Meanwhile, Self-organization refers to the ability of complex systems to organize without an external blueprint or help.

Let’s assume that the community just came into the picture, all things remain the same. Over time, the system adjusted/ adapted to the new element introduced into the system (the community) and it generated new connections that made the system even more complex. Moreover, nobody instructed the system to behave this way when the external factor was introduced (self-organization)

This kind of behavior is evident in most biological systems, like how a dam construction would gravely affect the biodiversity in an area since the structure would introduce temperature fluctuations and affect the flow of nutrients in the water system. As we continue to expose the system to external stimuli, we could expect that the system will continuously evolve. In fact, maybe the causal loop diagram that I provided above does not hold true anymore considering that 5 years have passed since the authors published their work.

As I conclude, understanding systems dynamics in engineering and development management birthed my fascination with Complex Systems. Complex systems are complex, and the key to understanding them is thinking in systems — or causal loops. While it may seem hard to grasp at first for an engineering graduate, consistent exposure to dynamics and a conscious attempt to understand the workings of society would eventually lead anyone to find connections among components forming the entire system easily. Once we identify those connections, it would be easier to uncover new relationships, dynamics, and characteristics from these systems allowing us to form more holistic solutions matched to the problem we are trying to address.

Like what they say, “modern (complex) problems require modern (complex) solutions.”