Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems
Swetha Sree - Bowie State University
Co-Author(s): Sharad Sharma, Bowie State University, Bowie, MD
Evacuation drills range from small scale evacuation of a building due to fire to the large scale evacuation of a city because of a flood, bombardment or approaching weather system. It is very expensive and time consuming to perform emergency evacuation drills in real time for a building. Our hypothesis is to have a goal-finding evacuation simulation application to aid in running several evacuation drills and theoretical situations. This project utilizes fuzzy logic to model stress and panic in a goal finding application. Human behaviors like stress and panic during emergency evacuation play important role in decision making. Also physical factors like body mass of individuals, their distance from exit and also obstacles or path stoppers on the way to exit affect people in responding to the situations and to take decisions. Our methodology includes implementing a fuzzy logic based engine which is capable of taking the internal and physical factors as inputs and calculate speed for each agent in the application. The planned application will help in executing numerous evacuation drills for “what if” scenarios by integrating agent characteristics. Our proposed multi-agents system is build using C# programming language. We have also conducted a real-time evacuation drill and compared our results with it. Our results show that the inclusion of fuzzy attributes made the evacuation time of the agents closer to the real time drills.
Not SubmittedFunder Acknowledgement(s): The authors would like to thank the National Science Foundation for supporting the project. This work is funded by the Grant Award number HRD-1238784. The authors would also like to acknowledge the support of the Virtual Reality Laboratory at Bowie State University.
Faculty Advisor: Sharad Sharma, ssharma@bowiestate.edu