1. What is a model?
A computer model is a simulation of a real situation.
Computer models are used to simulate the weather. The model is then used to predict the weather in the future.
Financial models are used in business. These models let business people see in advance what the effect of making business decisions will be. For instance, a financial model might tell a business how monthly profits will change if the business spends $1000 on an advertising campaign on the radio.
Whole countries use economic models to try and predict future trends for metrics such as the unemployment rate, gross domestic product and the rate of inflation. The model is used by economists to help them to understand the effects of making decisions like "raising interest rates".
A road model can be used to experiment with the timing of traffic lights at an intersection. The timing can be worked out so that disruption to motorists is minimized.
Pilots use simulators to practice procedures for handling emergencies without the need to practice procedures like "crash landings" using a real airplane.
2. Why are models important?
Models let us experiment with model parameters. They help us to understand a problem better. They help us to predict the future. The cost of changing model parameters is small when compared with the cost of changing the real environment. Imagine that we build a road, and have a guess at how wide to make the road. What happens when we find out at a later stage that we made a mistake and that the road was made too small for the number of people who want to travel on it. This would cause a lot of inconvenience and waste a lot of money. The road might need to be rebuilt or another road built. What happens if we build a roundabout and then find out that there are many accidents that occur there. A model of the roundabout might have predicted that a roundabout design was not going to be very good.
3. What is an advantage of making models?
Models let us test a design before having to manufacture it. We could make a model of a racing car and then test it in a wind tunnel. If the model works out to have good aerodynamics then we could manufacture the real racing car.
4. Name four different types of computer model.
(i) Financial models let us simulate a business. (ii) Flight simulator models let us model the behavior of an airplane. (iii) A 3D model of a house can be used to show us what a finished house might look like (iv) a population growth model can be used to predict future trends and then help us to decide where to build housing estates. (v) Expert systems model human behavior. They are able to provide decisions (using an inference engine) that are based on a large knowledge base. Expert systems "learn" and get more expert as time goes on…
5. What are the stages to be complete to build a model?
The validity of a model depends entirely upon how well the model represents reality. So, it is very important to understand the reality very well. To help to understand this it is common to (i) collect data from the environment to be modeled. (ii) A model is created. This can be created after analyzing the data. (iii) The model is tested and refined until the model predicts accurately enough for our needs. (iv) The model becomes a useful tool for prediction.
For example, to make a road intersection model we would need to know the number, speed and frequency of cars approaching an intersection. We could then make a model (based on a branch of mathematics called queuing theory). We could test the model with data we have collected by observing the intersection. We can then add a set of lights and try to adjust the timing parameters to minimize inconvenience to drivers.
6. What is virtual reality?
This is a computer representation of reality. A flight simulator is an example of a virtual reality environment. Some virtual reality systems require a person to wear a headset and special glasses.
7. What is an expert system?
An expert system is a model of a human in the sense that the computer is programmed to "think". The computer is able to make decisions based on its "experience" and the decisions it has made in the past. The model is very good for managing financial situations such as when to buy and sell shares on the stock market.
8. Describe how a model would be useful for planning the timing for the traffic lights at a busy intersection.
To make a road intersection model we would need to know the number, speed and frequency of cars approaching an intersection. We could then make a model (based on a branch of mathematics called queuing theory). We could test the model with data we have collected by observing the intersection. We can then add a set of lights and try to adjust the timing parameters to minimize inconvenience to drivers.
9. What are the main parts of an expert system?
An inference engine that makes decisions based on "experience" – which is stored as facts in a knowledge base.
10. Describe a real time model.
A flight simulator is a real time model. It provides feedback to a pilot so that he can learn how to carry out procedures in the event that an airplane has a problem.
11. Describe a simulation model.
A model for managing the macro economics of a country is a simulation model. This model is used to see the effect of making financial decisions. Governments need to make good decisions so that people can get more wealthy and the number of people who are unemployed is reduced.
