Exploring a Simple Disease Model

The disease model can be found at this address:

http://www.shodor.org/talks/ncsi/agentsheets/SimpleSickApplet/index.html

The model is an agent based model showing one sick person (red with a sad face) surrounded by well people (blue with a happy face). The model is set so that all of the people move randomly on the model surface. If a well person comes into contact with a sick person, there is a probability that they will also become sick. All sick people can cause well people to become sick with the same probability.

You can control the simulation in several ways. At the bottom left are buttons to start and stop the application and proceed one step at a time. In the center is a slider bar that controls that speed at which it runs. When you first run it, you may want to move the speed down halfway so that you can more easily observe what happens. At the right hand side is a reset button which sets everyone back into the same starting positions.

You can also control some of the parameters and create a graph of any parameter. You can do this by clicking on the arrow in the upper right hand corner and choosing Simulation Property Editor. The infectionrate is the probability that a sick person will make a well person become sick. You can change this number and see what impact it has on the rate at which people become sick.

You can also create a graphic plot of any of the variables. For example, to create a graph of the number of sick people over time, click on the sick line and then the Plot button. Check the box to turn on plotting. When you run the model, a small graph will appear showing the number of sick people over time.

One way you could explore the system is to change the probability that a well person will become sick when they come in contact with a sick person. You can then plot the number of sick people and see what the impact is on the time it takes to make everyone sick. Try this by doubling and tripling the probability and re-running the model. Is the resulting change in the time to get everyone sick linear or non-linear? Do you get the same answer every time you run it with the same parameters? Why do you think this is the case?