Analyzing Vulnerabilities in Highway Networks Using Optimization

Brandon Jerome

Author: Brandon Jerome | Major: Industrial Engineering

Hi, my name is Brandon Jerome and throughout the Spring 2022 term I had the opportunity to work on my honors thesis in the Industrial Engineering department, within the college of Engineering, with Dr. Kelly Sullivan implementing optimization research methods to analyze vulnerabilities in the Northwest Arkansas highway network. Following graduation, I will be joining Samsung Austin Semiconductor in an engineering rotational program and I hope to continue to work within the technology space with the goal of becoming a project manager or finding myself in a management position helping to run teams that implement big and innovative projects or departments.

My research utilized geospatial data on the Northwest Arkansas road network to create a network that we ran tests on via mathematical optimization models to tell us which areas of the road network are the most important, and therefore most vulnerable. Our optimization models basically play a game between a follower and a leader the follower runs a network with the goal of maximizing or minimizing some function while the leader strives to inhibit the objective from occurring by interdicting (i.e., damaging or removing) arcs in the followers’ network. In our instance the follower is an individual travelling along the NWA highway network the goal is then for the leader to find the most destructive set of arcs to damage to increase the followers travel time between some set origin and destination. We were able to run different experiments and versions of our optimization models which gave us copious amounts of data to build visualizations that detailed the sections of highways that our model found to be the most important. In a real world setting this research could be used by states, counties, or cities to aid in their funding and planning processes for new roads.

When I first began the process of completing my honors thesis, I enrolled in two classes taught by Dr. Kelly Sullivan that were aimed at helping students pick a mentor, topic, and write a grant proposal to secure funding. To pick our mentor our class first heard a presentation from each professor in our department that detailed what that professor performs research on and if they have any openings or ideas. After emailing Dr. Sullivan we met and talked about what areas of his research I was most interested in and began thinking about areas we could extend these topics or perform new analysis. My first idea for research that we pursued was in the wildfire space. This was the one part of my research that gave me the most challenges as I spent almost an entire semester performing background research but realized towards the end that it would be difficult to find relevant data on wildfires to meaningfully perform research. Pivoting to a new topic took a few long weeks of work to make up for all the lost time, so I would say to someone looking to do an honors thesis that creativity in your thesis topic is obviously important but be careful on how wild your ideas get. If I were to do my thesis over again and try to avoid switching topics I would have asked to extend or work on some current research rather than try to create a whole new topic. I would also say that in general if there are any classes you can take that are built to jumpstart your thesis process take them as they will put you miles ahead than where you would be without instruction, and they can help to apply for funding or grants which are an added bonus.