2011 Complexity and Modeling Program Project Page

From Santa Fe Institute Events Wiki

Complexity and Modeling Program Navigation

CAMP Participant's Final Projects

Gavin's Final Project: I plan to use this model to simulate the spread of Swine Flu via multiple methods of transmission through individual smaller, mostly isolated human populations, all of which are part of a larger network of similar populations. The model will demonstrate the effects of practices such as hand washing, quarantining, and school-closures across both the small and large populations. The model will demonstrate emergent effects of such practices over the general population, allowing the user to learn about the methods in which a virus spreads among a small isolated community, and then breaks out into a larger population. All variables will be dynamic through the use of various sliders, allowing total control over the spread of the virus. The model will incorporate aspects of acquired immunity, varied transmissibility over time, and varied susceptibility to infection.

Ben's Final Project: My NetLogo model is intended to demonstrate the effects of factory emissions on the health of the neighboring population. If a certain factory increases their emissions of harmful gases by a specific percentage, what is the effect on the expected lifespan of nearby inhabitants? If a large number of trees in the area are harvested, what will the decrease in carbon absorption do to the atmosphere in the vicinity? I hope to answer these question with my model.

Nico's Final Project: My final project will be based on the squirrel ecology lab. Squirrels will search for food but tend to remain in a certain perimeter around their tree. This is because: 1. The chance of predation increases as you get out into the open. 2. Food closer to the tree comes with a competition factor. 3. It takes energy to cross the distance to food. As a Netlogo model, this experiment will include these three factors. Depending on the hunger of the squirrel what food source will pose less of an overall cost? I will answer this question by using code from the ant experiment. When the squirrel gets the food, it needs to take it back to the tree to eat it, and when it does it will gain energy. If the squirrel doesn’t get food before its energy drops to zero, it will die. I will also include a predator that will kill the squirrel if it gets too far away from the tree.

Ken's Final Project: I would like to model my final project on a popular sci-fi theme, “Humans Vs Aliens”. In the model, I’ll have humans and aliens fight it out to answer the very basic question of “who will win?” I’ll set up different variables, such as attack power, initial population, reproduction rate, etc, to see different situations. Hopefully, through different combination of variables, I’ll see some interesting results in who wins and who loses under certain circumstances.

Tohanh's Final Project:

Since we're living in a melting pot where people are from all over the world, I think it would be beneficial to go deeper into people's behavior. Thus, my project will concentrate on people of different races and their "home-location" preference. This project is mainly inspired by the model "Ethnocentrism," whose idea is to have agents treat agents within their group more beneficially than those outside their group. The model will involve the Prisoner Dilemma's interaction, which means agents will gain or lose energy depends on their cooperation with others. There will be a death-rate to control the population.

The model has four types of agents:

1/ one that only settle next to agents of their trait

2/ one that only settle next to agents that doesn't belong to their trait

3/ one that settle next to any agent

4/ one that do not settle next to anyone.

From these distinct behaviors, the model will ideally generate interesting behavior within agents. Which type of behavior will get agents to expand their trait? Which behavior would make it difficult for agents to settle down? Is there any relation between the model's observation and real life behavior? My goal is to have the (partial) answers for those questions by this Friday through this project.

Henry's Final Project:

My Project is a model of the Amazon rainforest ecosystem. There will be 6-10 species of plants and animals that all interact with each other to make the ecosystem stable. But there will also be a function to clear-cut an area of rainforest or to add poachers that target a specific species. The purpose of the model is to show what can happen when a specific species or area within an ecosystem is targeted.

Sara's Final Project

My plan for my final project is to create an epidemiological and ecological model depicting the spread of White-nose Syndrome (WNS) in bat populations, particularly the little brown bat in the northeastern United States. The WNS is characterized by a whitish fungal growth around the faces of infected bats and acts by disrupting the hibernation behavior of bats, causing them to leave the cave and die from exposure and starvation. It is known that the fungus thrives in cold environments, but the method of convection for this disease is not fully understood. It is theorized that spores of the fungus are spread from bat to bat. I would like to examine the specialized flocking and roosting habits of bats and possibly the environmental factor of temperature and what implications these things may have on the spread WNS.

Jason's Final Project

For my final project, I want to model the spread of a virus through a school environment. I will have student-turtles gather in classes, walk between classes during passing time and maybe run around outside during recess. The model will run over a period of several days, with sick students being kept at home. I will see what happens if I change the infectivity of the virus, the number of students, or the length of the time period before students start exhibiting symptoms and get sick. I also want to make it possible to change the method by which the virus is transmitted. In addition, I want to see if the passing time length or recess length have an effect on the virus spread. I hope to see some interesting results by Friday.

Emma's Final Project

I intend on using my final project as a basis to learn how to program in NetLogo. My idea is to simulate a game of capture the flag, where agents are divided into teams and given the mission to bring back the opposing team's flag onto their own territory. When an agent contacts a member of the opposing team on their own territory, the opposing member is sent to "jail". Those in jail are freed when contact is made with a member of their own team. They then return to their own field uninterrupted, then resume play. If my programming ability suffices, I will test different strategies for the agents; have an offensive team play a defensive team, etc. My personal goal is merely to become more experienced with the programming mindset, and learn how to function in NetLogo.

Hugo's Final Project
I would like to make a flight simulator, the AI will be sophisticated, and I hope to make cool animations/graphics.

Ideas; I'll probably do:
k nearest neighbors for avoidance
switch between 3d/2d
animations, custom shapes
user controlled plane
teams will align in formations

I would like to see how planes interact with different settings. A possible application is safe traffic managment in an airport.

Alexis and Teddys' Final Project

Our project will be a model of drug traffic control in New York City. The agents will be defined as the policemen and shipments. The police search the shipments, each of which has a given probability of containing narcotics. Unfortunately, the system isn't perfect. Some shipments containing narcotics will never be searched. Others may be “false negatives” and pass inspection even if containing illegal goods. Meanwhile, the drugs that DO get through this system will be introduced to the relatively small-volume trade on the streets. Under the same budget, the user can choose to direct some of the law enforcement towards catching these criminals; that is, if the police “wiggle” within a certain radius of the criminals, there will be a given probability of a successful arrest. The model will display the volume of narcotics confiscated by law enforcement. Thus, ideally, our model will provide information on how to best direct the budget between large-volume smuggling at the ports and small-volume trade on the streets.

  • CAMP Participant's Models can be seen on the Santa Fe Institute Website