5. Check your model for compliance with any appropriate physical, economic, or
other laws; for examples, the conservation of mass, energy, value; any continuity requirements. Also, check for consistency of units. Look for the possibilities of division by zero, negative volumes, or prices, etc. Use conditional
statements if necessary to avoid these violations. Fully document your parameters, initial values, the units of all variables, assumptions, and equations before
going on.
6. Choose the time and space horizon over which you intend to examine the
dynamic behavior of the model. Choose the length of each time interval for
which state variables are being updated by reference to the space over which
the dynamics occur, and mainly by reference to the fastest rate of change you
expect in your model. Then choose the numerical computation procedure by
which flows are calculated. Set up a graph showing the most important variables and guess their variation before running the model.
7. Run the model. Are your results reasonable? Are your questions answered?
Choose alternative lengths of each time interval for which state variables are
updated. Choose alternative integration techniques. Explain any differences.
8. Do a sensitivity analysis of the parameters and initial values in the model. Try
out these small changes singly and collectively within their reasonable
extremes and see if the results in the graph still make sense. Revise the
model to repair errors and anomalies.
9. Compare the results to experimental data. This may mean shutting off parts of
your model to mimic a lab experiment, for example.
10. Revise the parameters, perhaps even the model structure to reflect greater
complexity and to meet exceptions to the experimental results, repeating
steps 1–10. Do the results of this model suggest a new set of questions? They
should.
1.7 Why Model?
Now that we introduced you to modeling, the software, and general principles of
modeling, it is time to step back and ask ourselves again an important question:
Why, and for what purposes, do we develop models? Dynamic modeling has four
possible general uses:
• First, you can experiment with models. A good model of a system enables you to
compare your result to those available from the real system and to change the
model components in order to see how these changes affect the rest of the
system. You can experiment, form and run scenarios, and bypass the inherent
risk aversion to making changes in a real system.
• Second, a good model enables prediction of the future course of a dynamic
system. Some modelers want only to explain what is going on, others aspire to
a higher and more difficult (and dangerous) calling: forecasting. A good
model will highlight gaps in what we know about the system we are studying.
1.7 Why Model?
23
other laws; for examples, the conservation of mass, energy, value; any continuity requirements. Also, check for consistency of units. Look for the possibilities of division by zero, negative volumes, or prices, etc. Use conditional
statements if necessary to avoid these violations. Fully document your parameters, initial values, the units of all variables, assumptions, and equations before
going on.
6. Choose the time and space horizon over which you intend to examine the
dynamic behavior of the model. Choose the length of each time interval for
which state variables are being updated by reference to the space over which
the dynamics occur, and mainly by reference to the fastest rate of change you
expect in your model. Then choose the numerical computation procedure by
which flows are calculated. Set up a graph showing the most important variables and guess their variation before running the model.
7. Run the model. Are your results reasonable? Are your questions answered?
Choose alternative lengths of each time interval for which state variables are
updated. Choose alternative integration techniques. Explain any differences.
8. Do a sensitivity analysis of the parameters and initial values in the model. Try
out these small changes singly and collectively within their reasonable
extremes and see if the results in the graph still make sense. Revise the
model to repair errors and anomalies.
9. Compare the results to experimental data. This may mean shutting off parts of
your model to mimic a lab experiment, for example.
10. Revise the parameters, perhaps even the model structure to reflect greater
complexity and to meet exceptions to the experimental results, repeating
steps 1–10. Do the results of this model suggest a new set of questions? They
should.
1.7 Why Model?
Now that we introduced you to modeling, the software, and general principles of
modeling, it is time to step back and ask ourselves again an important question:
Why, and for what purposes, do we develop models? Dynamic modeling has four
possible general uses:
• First, you can experiment with models. A good model of a system enables you to
compare your result to those available from the real system and to change the
model components in order to see how these changes affect the rest of the
system. You can experiment, form and run scenarios, and bypass the inherent
risk aversion to making changes in a real system.
• Second, a good model enables prediction of the future course of a dynamic
system. Some modelers want only to explain what is going on, others aspire to
a higher and more difficult (and dangerous) calling: forecasting. A good
model will highlight gaps in what we know about the system we are studying.
1.7 Why Model?
23