Tuesday, February 6, 2018

Systems (updated)

I had to update this mental model introduction, because the former text wasn´t good enough. 


System is defined as: ”A set of things working together as parts of a mechanism or an interconnecting network; a complex whole or ”A set of principles or procedures according to which something is done; an organized scheme or method.


Universe is a system. Our bodies, habits and skills are systems. Most of the things we do are systems. Some systems are consisted on smaller subsystems like our bodies have skin, organs, etc. Latticework of mental models is an example of a thinking system.

Systems basics

All the systems have elements, interconnections and a function or a purpose. Elements are not necessarily physical things. For example, your individual characteristics like self-confidence is a parts of the system called you. Anything with different elements is not a system. You have to ask yourself couple questions: Can you identify parts? Do they affect each other? Do they produce a behavior together that is different from the behavior with separate individual elements? And does the behavior over time stay the same in different circumstances? The interconnections are the relationships that keep the elements together. Interconnections can be physical and informational flows. Physical flows can be water going through the pipes, etc. A signal of measured temperature of the room may change the electric power used for heating. It is an informational flow. You should always think about the behavior of the whole system, even though you are concentrating on one element or interconnection.

A function or a purpose is not necessarily known consciously. It can be found by examining the system behavior for a while. You cannot really understand a purpose or a function if you believe in its stated purpose. This part is mostly the most crucial part of the system. The system is changed, when its purpose or function changes without any changes in elements or interconnections. Changing the purpose or a function of a system is one of the best ways to make it better. Depending on the system, it can be one of the hardest things to do. Change in the interconnections change the system too. It can become unrecognizable. Changing the elements of the system do not create big changes in the system, unless they do not change the purpose or the interconnections of it.

The element of the system you can feel, see or measure at any given time is a stock. It can be a the temperature in the room, a group of people or the amount of cars in the inventory. Flows are the actions that create the change over time for stocks. Flows change the amount of stocks within the system. If you understand the behavior of stocks and flows over time, you understand a lot about the behavior of the system.

Think about the room temperature when there is cold weather outside. The heating system creates inflows and the bad insulation creates outflows of the heat in the room. Change in the heating power changes the temperature inside as does the change in the temperature outside the room. When the inflow is equal to the outflow, the temperature stays the same. In other words, a temperature can be increased by decreasing its outflow rate as well as by increasing the inflow rate. These changes are mostly slow.

Stocks work as delays or buffers of the system. The bigger the stocks, the responses to the changes are slower. These changes happen mostly gradually. The speed of changes in systems are set by the changes in stocks. Stocks help inflows and outflows stay independent from each other. Flows can also be temporarily ouf of balance, because stocks work as buffers. The volumes of stocks can be changed by changing the volumes of flows. Keeping the stocks in the acceptable ranges, we need feedback processes that manipulate the volumes of flows.

Self-reinforcing, self-stabilizing feedback loops and delays

If you see a system behave consistently over time, there is likely to be a mechanism called feedback loop. There is a feedback loop, when changes in stock affect the flows into or out of the same stock. Feedback loops can be either self-reinforcing or self-stabilizing. Universe is always expanding. The interest on interest compounds the debt. In self-reinforcing loops, the previous outflows of the stock change the inflows coming into the stock and amplify the outflows coming out of the stock. What happens is that eventually self-reinforcing stock gets to the point where the amplifying destroys the whole system. The other possibility is that the limit of growth comes to the point where the growth stops, slows, diverts or reverses. When this happens, a self-stabilizing feedback loop is formed.

This self-correction keeps the system working without exploding. Every natural system has an optimal growth rate. We should use it into our advantage. Our skin keeps the temperature of our body from overheating by sweating if we are in an environment in which the temperature is too high or we are exercising. It can be hard to notice these stabilizing processes, even though they are mostly necessary. We should keep the self and self-stabilizing loops and their interaction in balance. We should always limit the effect of self-stabilizing processes on self-reinforcing processes in balance. Most of the systems have many interacting feedback loops.

All the systems have delays. Feedback loops always have some delays. People have a natural tendency in concentrating on the consequences we see right after we have done something. The second and third order consequences may come after a long delay. In complex systems, consequences can come after years. We may concentrate on the symptoms, rather than the solutions because of the delays. Some system structures have unrecognized delays and they can lead to wrong solutions. You should have better awareness about the delays. And we should also remember that causes and effects are not always close in time. When you notice a delay in the system, you have to find out, whether the length of the delay is too short or too long. When the delay is too short, it leads to too much variation for the outflow of the system. When the delay is too long the system can become too stable and ineffective.

Small changes in systems can create big results

This happens in individual circumstances as well as in systems. Well-focused small actions aimed for changing the system can create surprisingly big results. Solving a difficult problem can be a matter of finding the system structure in which the small change delivers the big and lasting results. Most of the times, finding these structures is hard. You need to understand the system and how its parts interact with each other. Otherwise you will never find the right structures for these high leverage improvements. You cannot do this without understanding the system as a whole.

Reacting to the change in the system is a lot easier than truly understanding its cause-effect relationships and ways to improve it with small changes. Reacting to the change can be a bad thing. It may cause small changes to the system structure creating bigger problems through self-reinforcing feedback loops. Changing a system structure can have different consequences in the short run and in the long run. Different parts of the systems can also have different consequences caused by the small changes. You also have to think about interactions of the parts before changing anything in one part of the system.

Most of the results in your life depend on the quality of your systems

You are mostly focusing on the different parts of your systems in a particular moment without thinking about the system as a whole or any long-term consequences. Your systems can deliver results that exceed the sums of their parts by a large margin. They can also deliver results that are not even close to the sums of their parts. People using the same systems tend to produce similar results. When you do not change your systems you cannot expect different results. Most of the time, the results you get are not caused by other people, some particular conditions, bad luck or some other explanations. The reason for bad results are the systems or their implementation or your understanding of them. You should think a lot more about the quality of our systems or their subsystems. Then you can get better results. You cannot really improve yourself without improving your systems.

Have a nice week!


No comments:

Post a Comment