Systems Thinking and Simulations: Part II

2 comments

Posted on 19th February 2010 by admin in 3D Virtual Worlds |Experiential Learning |Simulations

“We learn best from experience, but we never experience the consequences or our most important decisions. How then can we learn? (Senge, 1990)”

We all work and live in complex systems. However we learn how to interact with segments or parts of these systems as opposed to learning the whole. For example in an end-to-end product development system, the end user or client places the demand on the system. Clients are spread out over large areas and their reasons for placing demands on the system vary. Product development systems normally are fragmented and spread out over the globe in some cases.

Consider a modern development process where the marketing research occurs in one location, design and prototyping occur in another location, production at an over seas manufacturing facility and finally it enters the global sales and distribution network. Normally the way we learn about these systems is with information centric “system overview” presentations followed by each team or group learning their respective roles in segmented training courses or curricula. There is a problem with this approach. What people do not learn, as Senge (1990) suggests, is how all the parts and processes interact with one another over time and distance.

Most of the systems that we work in are now very global. As a result of this larger distribution of the segments, the system becomes ever more complex and it becomes more difficult to understand how our actions will impact some other part of the system. Some of these added complexities include: local culture, language, policies and regulations. To make matters worse, the systems that we work in also include environmental and global economic impacts. So how can someone learn how the entire system works and how their small actions impact the larger system? The answer suggested by Senge (1990) is through the use of simulations which he refers to as “microworlds”.

“Microworlds will, I believe, prove to be a critical technology for implementing the disciplines of the learning organization. And they will accomplish this by helping us rediscover the power of learning through play. (Senge, 1990, p. 315)”

In 1990, Senge presented screen shots of simulations developed on a Macintosh, probably developed in Pascal. Pascal, named after the scientist and philospher Blaise Pascal, was a very early object oriented programming language which is needed to develop object based simulations. What do I mean by object based simulations? Let’s consider a very simple system like the heating system in your home.

The system has a thermostat used to adjust the desired temperature which is the setpoint. The thermostat also has a temperature sensor to receive feedback from the system, your house, in terms of current temperature. What the thermostat does is determine if there is a gap between the current temperature and the desired temperature (setpoint). If the temperature is below the setpoint, a switch is closed and your heater motor starts up, blowing hot air throughout the house. Once the air temperature around the thermostat reaches setpoint, the thermostat switch opens and shuts down the motor. If we were going to develop an object based simulation for this system, we would create several objects in a programming language.

  • House
  • Windows
  • Air
  • Thermostat
  • Switch
  • Blower Motor
  • Person 1
  • Person 2

What the programmer would do is create a representation of each one of those objects and program into the object the variables and functionality for each. The house might be the object that connects all the other objects. If a window is open, then temperature of the air in the house goes down, if the air outside the house is colder then the temperature inside. Why did I select two persons as objects? Because we all know that people have different preferences for what is a normal temperature in the house and person 1 might walk up and drop the setpoint on the thermostat or open a window if he or she is too warm. Now once you have your single house simulation up and running you can make duplicates or instances of it and change some of the variable on sets of house instances. For example you can set the variable for 1000 homes that have poor seals on their windows. Make 100,000 instances of the homes and provide outputs of what is going on in the 100,000 homes to the power generation system. Now you can start to see what is going on in your system. For anyone that has every played SimCity, this might sound familiar. However SimCity is missing a critical component, live social interaction. Human nature is extremely complex and nearly impossible to fully simulate. So what if we take complex simulations like SimCity and now add live people into the environment and let them make decisions and take actions at a micro-level as well as a macro-level.

One of my favorite science fiction shows, Stargate Atlantis played out this scenario in one of their episodes. In Season 3, Episode 16 called “The Game”, two of the shows characters, Sheppard and McKay discover what they believe to be a simulation game on the Atlantis computers of two societies. McKay controls the one society and Sheppard controls the other. What they find out is that the game that they have been playing is connected to a real world with real people on it. What they learn when they finally meet the people on these worlds is that the game gets a bit more complex when real people are involved.

3D virtual live environments like Second Life have the potential to provide for these sorts of complex simulated worlds with the added dimension of real people with all the complexities that humans bring into a complex system including emotion and complex cognitive processing within the human mind.

So why would this sort of simulation be important? It is important not only for systemic learning but large scale decision making.

Let’s say that congress is considering ways to save energy and in turn save the environment. Someone in congress suggests a tax break for replacing old doors and windows. This seems like a good program but how will it actually impact the system? How will a law like that impact the economy and the environment. Only with complex interconnected system simulations can we begin to understand the impact of a decision like that.

Or what if a CEO and her executive staff are considering moving the manufacturing division to another country. How can they really know what is going to happen in year, two years or even ten years forward in time. Simulations could help to drive this kind of visioning. In fact if the team wanted to understand different impacts from how different cultures react to the system, they could place people from the different locations around the globe into the simulation and allow them to act as they normally would act when presented with a complex problem such as a spike or steep decline in demand.

Now what we need is to gradually build objects on a standard platform and gradually begin to interconnect them until we have a mirror image of the Earth’s systems. This would allow government and corporate leaders to “play” things out before making decisions to develop a product, outsource, go to war, or change a major policy. It’s an intriguing and also scary thought.

What do you think?

References

Senge, Peter M. (1994) The Fifth Discipline. New York: Currency. Print.

System Thinking & Simulations: Part 1

0 comments

Posted on 15th February 2010 by admin in Experiential Learning |Simulations

When learning professionals consider Senge’s book (1990), “The Fifth Discipline”, the first thing that comes to mind are the terms, “systems thinking” and the “learning organization”. However for anyone who has made and attempt to read the entire book, you probably walked away with the thought, how do you implement this grand vision? Senge (1990) suggests that computer based simulations are an excellent way to provide the type of experiential learning needed to develop the disciplines or the competencies outlined in his book.

Peter Senge (1990) presents five disciplines or competencies that he suggests are required in order for an organization to truly become a “learning organization.”

The disciplines include:

  • “Personal Mastery” which Senge (1990) defines as, “approaching one’s life as a creative work, living  life from a creative as opposed to reactive viewpoint.” In essence we need to accept that we are constantly learning and developing. We need to take an introspective look at ourselves from time to time, look at where we are and where we want to go and constantly adjust the path to get there. This does not refer to gaining more knowledge although it is part of the process. Personal mastery refers to improving ourselves, and aligning our behaviors with our personal vision. Personal mastery includes “personal vision” which Senge (1990) explains is much more than goals and objectives. A vision comes from deep within. It’s something you are truly passionate about. Guy Kawasaki (2004) refers to this as having a desire to “make meaning”, which Kawasaki suggests is a critical attribute for successful entrepreneurs.  Personal vision is not driven by the desire to gain money or power. It’s the sort of thing that drives social entrepreneurs.
  • Second, we need to learn how to manage the “Mental Models” that we all have. A mental model is the way in which we view the world. It is developed throughout our life and is frequently a result of cultural evolution, handed down through generations. Organizations have mental models that are also developed over time. It is what many organizations refer to as it’s corporate culture. Social learning theories suggest that our culture evolves through the generations but with much greater speed then genetic changes. (Flinn, 1997) We need to first recognize that these deep rooted ways of looking at the world exist in an organization so that we can address it when developing learning and change management initiatives. Mental models are generally based on assumptions. An organization that is not willing to question those assumptions can be crippled from learning.
  • The third discipline is “Shared Vision.” When I think about the 1980 U.S. Olympic hockey team and why they won, I believe it had a great deal to do with achieving a shared vision. The team members first had to accept that they were not part of distinct colleges but were now part of single team. In fact the shared vision that was established by their coach, was not only about winning the gold medal, but it was much about becoming one. It was more then talent and skills. If you have never seen the movie, please watch the 8 minute clip below. It serves as an excellent metaphor or developing shared vision.
  • The fourth discipline is “Team Learning“. Probably the best example for team learning is a symphony orchestra. Each musician needs to master their own instrument but they also need to play in unison with all of the other musicians. If all of the musicians do not play in unison the music is not complete or full. All of the musicians need to learn together as a team. You can’t have one stand out. Another important aspect of team learning is recognizing that diversity in team members, strengthens the outcome of the teams. (Leonard, D., & Straus, S. 1997)

You need to watch the entire 8 minutes to fully understand the concept of shared vision as well as team learning. This is a true story and it was a culminating moment for this team. Think about the messages that the coach gives the team members with every suicide run. In U.S. team sports, we call these suicides. You will understand why when you watch the video.


Systems Thinking

All of these disciplines, Senge (1990) argues, must be integrated with the fifth discipline of systems thinking. However there is a challenge with learning systems thinking. Not only do you need to understand how the system is functioning and what feedback loops exist, but you need to learning through experience how your actions and behaviors impact the system. This is very difficult because our actions and their subsequent impact can be separated but significant distance and time. Senge in 1990 introduced the concept of “microworlds” which are essentially computer based simulations that allow one to change the variables in the environment and see what the impact will be on the system. Remember this was 1990. The screen shots in Senge’s book are from a Macintosh, probably developed in either Basic or Pascal. In my next post we will explore different cognitive technologies both old and new for addressing the challenge of teaching systems thinking and integrating it with the other four disciplines.

References

Flinn, M. (1997). Culture and Evolution of Social Learning. Evolution and Human Behavior, 18, 23-67.

Kawasaki, G. (2004). The Art of the Start The Time-Tested, Battle-Hardened Guide for Anyone Starting Anything. New York: Portfolio Hardcover.

Leonard, D., & Straus, S. (1997). Putting your Whole Company’s Brain to Work. Harvard Business Review, 75(4), 110-121.

Senge, P. M. (1994). The Fifth Discipline. New York: Currency.