I have come to believe that the most important learning goes on
during the process of constructing the model. Constructing a model
forces you to identify the relationships between system components,
establishing what the controls are, what the connections are, what
things you do understand and what things you don't understand. In
other words, this process generates questions and those questions
motivate the search for answers. This means that even if you are
unable to find all of the data you need to make a reasonable model,
one that generates acceptable results, you can still learn a great
deal about the real-world system that you're working on.
RETURN TO MAIN PAGE