Simulating Emergency Evacuation Events

Anna Lee

Author: Anna Lee | Major: Industrial Engineering

My name is Anna Lee, and I am majoring in Industrial Engineering within the College of Engineering. During the spring semester of 2022, I performed research under the advisement of Dr. Manuel D. Rossetti. I hope to continue this research throughout the summer and complete it at the end of the fall semester of 2022.

My research aims at creating a simulation model of an emergency evacuation in a large auditorium space. This semester I was able to complete background research on similar simulation models and different approaches I could take for my own model. My goal for this research is to build a model that provides insight into how quickly people are able to get to safety and possible reasons the process is slowed down.

One of the courses I had to take to complete my degree was Simulation, which was taught by my research advisor, Dr. Manuel D. Rossetti. During the course, my interest was sparked by the large number of possibilities that simulation provides in a variety of fields. I was particularly interested in the simulation of evacuations and their relevance to the world today. Evacuation plans are very important to any enclosed space, so I was not surprised to find that the topic of simulating evacuations was highly researched. Through my background research, I was also able to understand that there is a multitude of different simulation software packages and many different simulation types that could be used to build a model. The simulation type I found to be most applicable to human-centered processes is agent-based modeling. I spent time in the previous semester learning about agent-based modeling and how it is used within two different simulation software applications with the goal of selecting the best software to use for my eventual model. Through this, I discovered how unique agent-based modeling is to other model types and the challenges that it would bring with it. Agent-based modeling is different from the other modeling techniques I learned in my course, such as discrete event or Monte Carlo simulation. It focuses on singular agents and the impact they have on a system, which most closely describes an evacuation event.

My faculty mentor has been able to provide a large amount of support and guidance due to his expertise within the field of simulation. Through his advisement, I decided to build my model using Arena Simulation Software. This will give me the best opportunity for success due to his previous work in the area and my knowledge gained from taking his course.

Going into the summer and fall semester I hope to continue building a list of input parameters, such as the size of the auditorium space and the number of people or agents involved, to have as accurate a model as possible. I plan on simulating a space similar to Hillside Auditorium at the University of Arkansas campus because I’d like to understand evacuations in a general auditorium space as well as lecture halls specifically. I’m excited to begin work on building this model and analyzing its output to draw meaningful conclusions on the safety of enclosed spaces with high-density populations.