HOME

Building the model: things + communications + contexts

Once you have a clear idea of the situation ... once you have reviewed the data and defined the terms that you might use in your model, then you are ready to start putting it all together.

The definitions will turn out to fall into three basic categories:

I use some of these terms interchangeably: thing, object, node ... all of these are the same. Ditto with relationship, communication, connection. I do this because:

  1. Different books on systems thinking use different terminology. You might as well get used to seeing all of them.
  2. You should not imagine that the system terminology has any special power. It simply does not matter whether you call it a relationship, a connection or a communication ... the important thing is that it enables one of the items in a system to interact with the rest of the system.

Building the Model

In the previous tutorial I started building an example system from a recipe for pasta sauce. Here are the definitions:

Things

Relations

Context

So What? All You've Done Is Make Up a Bunch of Categories

Yep; that's true. Nonetheless, what you see here is the result of a lot of thinking about the situation. The lists of items and connections and contexts is there to remind me of my thinking ... and so is the diagram. But it's the thinking that matters.

I can't emphasize that enough. As opposed to aesthetics ... in which you appreciate and get in touch with how you feel about the situation ... or criticism ... where you build a logical argument to prove a point ... or science ... where you dig and dig and dig to gather facts and details ... systems thinking is about trying to figure out what is going on here.

The lists of things, relations and environments ... and the diagrams ... and the model that you will build from these ... is there to force you to prove to yourself, and to others, that you have actually thought the problem all the way through.

Don't get fixated on the lists and the diagrams and the model ... it's the thinking that matters.

OK. What Next?

In reading through the relations list, I realize that what's missing from my first try at the model is time. Clearly some of the items need to be prepared before others. And this means that I cannot just add everything to the pot and turn on the heat.

Browning: onions need to be browned, and herbs can be browned. Apparently, this needs to happen first, and in the oil. I figure this out because once I add the liquid (in the tomatoes and wine) the mixture will be boiling ... and browning will no longer be possible. Also spices, although they are similar to herbs, can burn if browned too much ... so I'll want to add the spices after the onions and herbs have started to brown.

Boiling: this needs to be done, at least for awhile, to get the best taste. Since adding the tomatoes will cool the liquid, there will need to be more heat at that point.

In What Order?: it sounds like I want to heat the oil first, then brown the onions and herbs, then add the spices followed by the wine, then add the tomatoes and turn up the heat until everything is boiling.

Cautions: in addition to the timing relations, I need to be careful not to leave the heat on high too long. And I need to not cover the pot too tightly. If I'm not careful in doing this, the pot will "boil over" into its environment.

So now my model looks something like this — a three-step process with some cautions:

SAUCE

Note: the single pointed arrows mean "in" ... "the herbs brown in the oil". The double pointed arrows mean "together" ... in the third box "all of the elements cook together".

By the way, notice that the spices are included in "boil." They are weird ... they "can" be browned, but they don't have to be. And they might burn if browned too long. So, really, I want to add them between step 1 and step 2 ... but it doesn't seem to matter all that much, really, so I just moved them to step 2. If this later turns out to be a problem, I'll have to create a step 2.5 just for the spices.

What This Is NOT

This is not a recipe for tomato sauce. Possibly you could use it to make a pot of sauce ... but that's not what it is.

Nor is it a scientific equation. It does not contain specific amounts or times or any of the small details that you would need to do a scientific analysis of tomato sauce.

Nor is it a story. I am not trying to tell the wonderful tale of how my grandmother came to develop this pasta sauce and pass it on to her grandchildren.

In fact it isn't any of the normal things that you probably expect it to be.

What This IS

It's an idea. When I load it into my brain, it helps me understand what goes on when I cook up a batch of this sauce. It doesn't give me full details, and it isn't a recipe ... but it shows me all of the items that need to be in the pot, it shows me how they relate to one another, and it gives me the sequence of events that I go through when I make the sauce.

With this understanding, I can answer questions about the process of making the sauce. If I pass the recipe on to my sister, for example, and she call me up and says, "my sauce tastes burned", I can respond: "are you adding the spices with the herbs ... because the spices can burn in the hot oil."

So What?

You're right, no one in their right mind would build a systems model for the process of making pasta sauce. At least not for a sauce made in a normal kitchen.

But suppose you are a global food corporation with a factory that needs to bottle 10,000 cans of pasta sauce every day? The factory manager cannot simply open a recipe book and start cooking. In fact the factory will probably have a number of different rooms, each containing vats and storage bins, through which the product must flow as it moves from raw materials to finished tomato sauce.

Our model says that the factory needs a browning room, a boiling room and a cooking room ... at least. And it gives a timing pattern for how the food will pass through these rooms. Once the details are worked out ... one ton of onions, 100 pounds of herbs, 50 gallons of oil, and so on ... the method laid out in the system model can be applied.

On a smaller scale, the same model might be applied to a cafeteria that only needed to produce 50 gallons of tomato sauce for dinner one night. In fact, the "recipe book" for a cafeteria might consist of a collection of such models.

This is one of the advantages of the systems approach. Once you have done the thinking required to build the model, it is usually not difficult to scale the model to fit new dimensions.

Another way to put this is to note that: models are portable.

A problem statement, along with the data collected, the set of definitions, the model and its test results, provides an easily communicated, concise statement that others can use to understand your problems and perhaps apply your solutions to their own.

Learning to Organize Data into Models

Systems thinking lies part way between the scientific method of collecting and studying data, and the critical method of debating arguments. In the pasta sauce situation I first collected data about sauce making by observing someone make a batch of sauce. Next, I laid out a model that I thought fit the data. Then I argued with other people about whether or not the model I had build was good enough.

Some people do model building in a more mathematical way, others do it more like they were making a painting or a sculpture. Some do it as if they were writing a story. There is no, simple, one-size-fits-all way of doing it. It is, basically, thinking ... and each of us thinks in our own unique way.

Personally, I treat it as a kind of puzzle. I know certain facts, and I have a certain problem to solve, and so I want to arrange the facts in ways that help me solve the problem. So I try out a particular pattern and see if it works. If not, I try again. Sometimes I decide that I don't have enough facts, so I gather more and start over. Most of the time I find the effort interesting, even fun; but other people would find this approach rather tedious.

Whatever approach you decide to use, certain aspects of systems thinking will almost always be true:

Systems thinking is probably best thought of as a craft. It requires the integration of various mental skills: observation, logic, imagination, and your abilities improve through practice.

I've included some extensive examples of systems thinking in these tutorials, and many others are available on the links page. Reading explanations of how other people do systems thinking may help you. However, in the end you will have to jump in and try it for yourself.

(Systems thinking is fairly easy to learn through practice, but difficult to teach via theory. College courses in systems theory and thinking exist, but as far as I know, there are no academic degrees in the subject. If you look at the biographies of the authors of the books on the links page, you will find biologists, economists, sociologists, mathematicians, and a few people who are simply "consultants." This is typical of people who work with systems ... they studied something else in college and gradually learned systems thinking as they solved problems related to their work. Eventually, they left their original fields and just worked at solving problems in general.)

 

Examples

The following examples illustrate the contents of this tutorial:

Example 1. College Life

Example 2. Fossil Fuels

 

HOME – Next Tutorial