Educators Online |
WebED Education, Outreach and TrainingA Short Course: Using Modeling and StatisticsWhat is Modeling? Dr. Steven I. Gordon Modeling is actually something that people do all of the time without realizing it. What is modeling? A basic definition is any simplification of a complex system that is used to understand that system or make decisions about your behavior in that system. To illustrate, let's start with an everyday example. Let's say you need to decide the best route to take to work To make this decision you will need to formulate at least one objective for your trip. Do that now before you click on this link to see a list of possible objectives. If we focus just on the first objective on the above list, we need to decide what will effect that objective. To do this, we must create an abstraction of what is a complex system and set of decisions. We have a finite but potentially large number of routes we can take. Each section of roadway has a particular speed limit. Each intersection might cause us to slow down because of a stop sign or a traffic light. The amount of traffic on each section of roadway will affect how fast we can drive. This varies dynamically by time of day -- at peak rush hour there is more traffic and it moves slower. What time of day is your work trip compared to this peak? Other things that can slow us down are construction, the occasional accident, or changes in the weather. If the weather is bad, we might change the path we take. In snow, the main streets are more often plowed and salted and thus move much faster than the side streets we might take on other days.
Based on our own experiences, we abstract this complex reality to create a mental model of what we perceive to the best that route that meets our objective. Every day we take this route and this gives us real data that may or may not match our original expectations. For example, we may have omitted information about intersections in our original model but find that one particular traffic light (where we have to make a left turn) becomes a major problem during congested times; it may take three or four light changes to get around the corner. In essence, we are comparing real data to our model, and adjusting the model and our route accordingly until they get us closer to our objectives. If we formalize this process it might look like the figure below.
Models are an abstraction of reality designed to allow their creators to gain insights into how systems work, how to manage them, and how to make informed decisions. In science, they are way of conducting inquiries about complex systems to address these questions.
Please contact Al-Azad Iqbal or Steve Gordon for Questions and Comments - Updated 10/2/07 |
 |
|
|
|
|
|
|
|
|
|
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
 |
|
||||||||
