Computer Science Professor Amelia Regan is passionate about her work. She served as director of the Transportation Science master’s and Ph.D. programs from 2015-2018 and was the founding director of the Master of Computer Science program. Her research interests span everything from cyber physical transportation to technology adoption in transportation and, more recently, technologies to improve the safety, comfort and convenience of pedestrians. Here, she talks about how her love of trucks put her on a path that eventually landed her in the family profession.
Can you tell us about your background and what brought you to UCI?
Who knows why, but I have always loved trucks! I studied mechanical engineering, thinking I wanted to do automotive engineering at the University of Pennsylvania, but it didn’t grab me. So I switched into systems engineering, which is a mix of applied math, civil and industrial engineering and, to a certain extent, computer science. When I graduated, I took a job in Washington, D.C. with the Association of American Railroads, working on large-scale network modeling of the North American freight rail system — it was big data long before the idea of big data. But then the United Parcel Service (UPS) started an operations research group to develop the first generation of their automated routing and scheduling systems and, because I had interned at UPS when I was at Penn, they invited me to join that group. It was a fascinating time to be doing that kind of work, but in order to get promoted, I needed a graduate degree. So I picked up a master’s degree in applied math — effectively, operations research – at Johns Hopkins University because it was local.
All of a sudden academics became more interesting — I was learning things that I was applying, and that made me understand how much I didn’t know. My twin sister was earning her Ph.D. at the time — she is now the associate university librarian for public services here at UCI — and my parents were professors, so I gave in to the family profession. I decided to go to the University of Texas where my sister was getting a Ph.D. in English, to get a Ph.D. in civil engineering.
I became an assistant professor of civil engineering here at UCI in 1997, and about 10 years later I moved to computer science, which has actually been a better fit. My engineering colleagues were great, and we still work together quite a bit, but I have more in common with the computer science undergraduates and to a certain extent the graduate students. It’s just the perfect time to be in transportation and computer science, because there’s so much going on with automation and new fuel and vehicular technologies.
So when will we all have self-driving cars?
One thing I’ve observed in the transportation industry is that everything happens more slowly than you think. People will say, “Oh, we’ll all have self-driving cars by 2025.” Heck, no. Before we have full, automated driving, we’ll have connected vehicles, communicating all the time and avoiding accidents and improving efficiencies. Ad-hoc vehicular networks have been part of my research for nearly 20 years, so it’s nice to finally see some applications in the field.
For automated driving, we’ll need extremely detailed maps on all the freeways, because you need to know everything about the road network. Where does the road dip off? Is there a shoulder? Is it a hard or soft shoulder? Are there skyscrapers that are going to interfere with the communication systems? These are all things that have to be worked out and measured, which is going to take a lot of time and money. And then who gets the data? Who owns it and shares it? These are important issues, and while automotive companies are working on automated driving systems, it’s going to be a while.
We will have platoons of automated trucks before we have automated cars, because you don’t have to constantly worry about changing origin-destination pairs — there are well-established transportation routes that have to be covered every day. Most of my interesting algorithm development work has been for the optimization of things that relate to freight transportation systems, and the most important work I’ve ever done is on combinatorial auctions, which are the methods by which companies get their long-term trucking contracts.
We have the technologies to make platooning on freeways and highways relatively safe, and there are huge economic and environmental advantages. Trucking companies respond to economics. Training and keeping truck drivers is very hard, and there are safety implications. So they’re going to save lives and money, because they’re going to have fewer accidents. There are labor shortages right now in the U.S., so it’s difficult to find people to do long-haul truck driving. It’s a career that’s not very attractive to anyone with a family.
Any other research interests?
I have been blessed to have outstanding Ph.D. students with varied interests — in transportation and in more general topics in computing. I can get interested in almost anything if I have the right research partners, and here at UCI and internationally, I have found no shortage of those.
Anything else you’d like to add?
Over my career, from high school and undergraduate school to UPS, the Ph.D. program and here at UCI, I have had the most incredible mentors and support from my peers. The single most important thing to me as an academic is to give that support back to students.
— Shani Murray