New Mexico Supercomputing Challenge
 I About I Resources I Help I Archive I News Flash I

Registration

Proposals
Interims
Final Reports

Dates

Kickoff
Evaluations
Expo
STI
Wiki

School Map

Mail

Challenge
Technical Guide

Past Participant Survey

GUTS

Challenge Team Interim Report

 Team Number: 056 School Name: GODDARD HIGH SCHOOL Area of Science: TRAFFIC MODELING - EMERGENCY EGRESS Project Title: Simulating Emergency Egress
 Abstract Interim Final Report

Problem Definition:
Emergency evacuation or emergency egress is a major concern in the construction and design of buildings. It is difficult to know whether existing evacuation plans are adequate to evacuate a large group of people out of a building in an emergency. Physical experimentation, with human volunteers, will provide much of this information, but it presents ethical, practical and financial problems. Human volunteers are subject to injuries even though there is no real emergency. Another problem with physical experimentation is the lack of realism. The volunteers know that they will be evacuating the building after a certain signal, so the situation lacks the surprise factor and the volunteers know that there is no real danger. Therefore, a computer model of an evacuating crowd could give a better picture of the real situation.
Problem Solution:
In our research, we have found two methods for modeling emergency egress, macroscopic and microscopic. In a macroscopic approach, a crowd is no longer looked upon as individuals but rather a single object with density and velocity. In the microscopic approach, the same crowd is made up of individuals each with specific characteristics. Because we feel it would be more realistic and interesting, we have chosen the microscopic approach. Our research also indicates that the microscopic method seems more popular among professionals in the egress-modeling field. Our microscopic model will create a class to represent people, with each object of the class representing one person. We have not yet decided if a second class (for physical space) will be created. Physical space may be addressed without a specific class defined. Our model will make assumptions and simplifications for individuals and physical space based on our abilities, and our desire to model our school's emergency egress reality.
Progress to Date:
Our team has extensively researched egress and human behavior in urgent situations. From this, we learned that some of our original ideas were too simple, or too complex. Since then, we have not only had the opportunity to interview the local fire inspector, but also to contact a few experts on the subject. Although we have received no response from the experts, we have uncovered numerous reports that are invaluable to the work we are doing. Most of the reports we have found have explained behavioral programming and helped us to better understand how parameters and algorithms will make our program work. We have discovered the existence of emergency egress modeling software that is quite elaborate and sophisticated. We had to make the decision whether to attempt to use this existing software or create the model by ourselves. Developing the model from scratch is the path that we have decided to follow.
Expected Results:
We expect our results to indicate problem areas during emergency egress based on given conditions. The conditions may include population density, level of panic, time allowed for evacuation, obstacles, and other factors. Our model will allow us to modify conditions thereby permitting us to determine critical factors that relate to efficient evacuations. Time and programming capabilities may limit our model so that we may not have as much detail as we would like. Therefore, our model will describe general patterns of egress and will match our school environment as well as we can knowing that it must be simplified.

Team Members

Team Mail