The Interconnections in and between systems

Before going to this text, you should probably read this, unless you are familiar with systems and its components.

Most things are interconnected. You cannot look only separate elements and their inputs and outputs and find out how the system works as a whole. You have to see how independent elements interact together. You have to understand what happens when an element A interacts with an element B. You have to understand what happens what happens to the element B when the element A changes its output or what happens to an element C, when the elements A and B change. Sometimes these interconnections are just information flows and sometimes they are feedback loops.

Interconnections can be loosely or tightly connected. Some of the elements can be isolated, but they are rare. It is also hard to find some systems that are isolated from others. Your body is one example of systems within the system. Some of its elements are loosely connected with others and some of them are tightly connected. Tight connections between different elements or systems are more efficient than loose connections. There are opportunity costs between the efficiency and the probability of not having a system failure. The more tightly connected the elements or systems are, the bigger the probability of system failure. When the failure hits the tightly connected element, the whole system will probably fail. In these cases, there is often only one path to a successful outcome. Systems can be loosely or tightly coupled as a whole.

Interconnectivity can be divided into four degrees of connectivity

Interconnectivity can be divided into four different degrees of connectivity:

1. Underconnected state: primitive cultures and undeveloped countries may be underconnected. This basically means that their environment may change without them knowing it. And they cannot do anything to react to the change.
2. Interconnected state: When the environment changes gradually, businesses, economic systems and governments are capable of keeping up with the change. For example, evolution works most often in the interconnected state.
3. Highly connected state: In this state, businesses, economic systems and governments are driving change. Capitalism, in its purest sense, works in a highly connected state. Sometimes it becomes overconnected.
4. Overconnected state: In this state, institutions, change so quickly that the environment in which they are embedded are unable to cope. Or the other way around: environment changes so quickly, because of the increase in interconnections that the institutions are unable to cope. Overconnected states provide disruption in one way or another. The crash of 2007-2009 was one of the best examples of a system in an overconnected state. The system failure was very close.

System safety and interconnections

Negative feedback loops mostly help make systems safer, but positive loops can cause serious problems. Positive feedback reinforces and amplifies change, accelerating it and causing the environment getting more unstable. In many tightly interonnected environments, initial stimulus works its way through the environment and back through feedback loops to provide more stimulus to the system. This drives the system faster in the same direction and causes rapid change. This rapid change causes more rapid change with increasing speed until the system will brake down. This is why you need to have balancing feedback loops in the system. They protect the system from rapid and unpredictable changes. These changes are the most dangerous ones. There can be long delays before these positive feedback loops start showing signs of trouble. Then it can be too late.

Information flows can also cause some problems in systems. They are always somewhat delayed. When the flow of information is delayed too much, the system can get into trouble. For example, when the balancing feedback loop has too big of a delay, the reinforcing feedback loop changes the system too much that it can cope with the change. Delays are hard to time properly. And they are also hard to notice sometimes.

Changing and removing interconnections can make a system safer too. Making looser interconnections can help systems become less delay dependent. They become less depended on the timing of the delay. Sometimes looser interconnections can help you to process things in parallel. Then you can have multiple actions in the same time. A system with tight interconnections may not be able to achieve this. You can also make a system safer by removing some of its interconnections. The more interconnections and the more tighter they are, the more probable it is that the system will fail. You should think about your interconnections in your systems and start removing all the unnecessary ones.

Sources:

Overconnected, William H. Davidow
Thinking in Systems Donella H Meadows

-TT