New Mexico Supercomputing Challenge

Traffic Simulation

Team: 65

School: Los Alamos High

Area of Science: Civil Engineering


Interim: Problem Definition:
Civil engineering, being the engineering discipline of creating roadways and traffic systems, is clearly of real use today. As roadways expand and old cities crumble, the innovation of new designs in communities is necessary. Yet, new traffic conditions, populations, and varying employment centers mean that we cannot keep recreating the same structures when they fall over. When the computer world takes leaps and bounds, we can easily see applications of using computer models to reflect traffic conditions as they change. Day to day, computer modeling can reflect the parameters of traffic centers and their intersections.

The goal of this project is to create a functional traffic simulator that will consist of multiple agents representing cars. These cars will each have a given starting point and destination, and, by means of a simple A.I., will select their own route to each point. Maps will be hard coded so as to reflect actual needs of traffic, and be loosely based on existing road maps. While the cars advance in a turn-based succession, the computer will observe the ability of certain traffic features to cope with the burden of traffic on that spot. Specifically, the location of a 4-way intersection will be observed, and we will vary the traffic function used there to see which one best handles the incoming cars under what circumstances.

Problem Solution:
As stated, the computer will be responsible for plotting the paths of several cars moving on the map simultaneously. The computer will run the simulation multiple times, with different combinations of traffic functions throughout the map. For example, the first trial may show every 4-way intersection with a stop light, whereas the next will use a mixture of 4-way stops and roundabouts. The efficiency with which the intersection disperses cars to their location will show its strengths and weaknesses relative to its conditions over a range of trials. The cumulative result of the data will show the circumstances where each traffic structure succeeds, solving the original problem.

Progress to Date:
Previously, we have created a simple A.I. for the cars that shows preference to larger roads and always advances in the general direction of its destination. We also have designed a system of input for the map, although it has not been drafted yet. We have researched the sorts of population influx seen on certain roads so as to best design a map that reflects a wide variety of traffic conditions, i.e. tight streets; long, open roads; big city travel.

Predicted Results:
This model should eventually allow us to see patterns in traffic caused by different types of intersection structures. Its use will be in no way limited by the map, as they will be uploadable, and the cars and other parts of the simulation will work regardless. The model will serve beyond our original research inclinations, allowing a user to see how best to design an intersection. Our original results will still stand, however. The model should demonstrate general conditions under which one might expect to see the greatest return from implemented traffic structures.


Team Members:

  Phillip Heikoop
  Nathan Delgado
  Michael Mallett

Sponsoring Teacher: Lee Goodwin

Mail the entire Team

For questions about the Supercomputing Challenge, a 501(c)3 organization, contact us at: consult1314 @ supercomputingchallenge.org

New Mexico Supercomputing Challenge, Inc.
Post Office Box 30102
Albuquerque, New Mexico 87190
(505) 667-2864

Supercomputing Challenge Board of Directors
Board page listing meetings and agendas
If you have volunteered for the Challenge, please fill out our In Kind form.