Civil engineering grad student focuses on infrastructure, advocacy
It’s easy to forget that consequences, even unintended ones, sometimes can be positive.
After Hurricane Katrina closed universities in New Orleans, UC officials offered students a chance to spend fall semester of 2005 at various UC campuses, including Berkeley. Their only motive was humanitarian. But the gesture affected students, including at least one at UC Berkeley, in unanticipated ways.
Iris Tien was an undergraduate then. As a resident assistant, responsible for two floors of an eight-floor dormitory, she was responsible for converting dormitory lounges into bedrooms and hosting events to help the New Orleans students feel at home.
Fast forward seven years. As a 24-year-old civil and environmental engineering graduate student in UC Berkeley’s Civil Systems Program, Tien is modeling complex infrastructure — research that might someday be used to determine weak spots in bridges, highways and water systems, including the kinds of levees that broke under the onslaught of Hurricane Katrina.
As the U.S. grapples with declining infrastructure and tight budgets, Tien’s work could prove particularly useful. Last year, the American Society of Civil Engineers reported that the poor condition of our highways, railroads, bridges and transit systems cost $130 billion in 2010 alone. To bring infrastructure back to minimum standards, the U.S. would need to invest $846 billion over nine years, or $94 billion per year. Yet partisan politics is keeping investment at a minimum.
It’s no wonder that Tien’s research on infrastructure cost-saving is getting attention. In March at UC ‘s graduate research advocacy day in Sacramento, her work drew much interest from the office of state Assemblyman Jerry Hill, who represents San Bruno, where a natural gas pipeline burst in 2010, causing an explosion and fire that killed eight people and destroyed nearly 40 homes.
Tien grew up in Silicon Valley, where engineering is practically a religion. She spent a summer working for Hewlett-Packard as a National Science Foundation Fellow and received a Chancellor's Fellowship for Graduate Study. But she’s hardly an engineering nerd. She’s not only a basketball junkie; Tien also steeped herself in the classics at Philips Academy in Andover, Massachusetts, before choosing to return to California for college. In literature classes at UC Berkeley, Tien found a mirror to her engineering studies in the writing of 18th century poet John Milton, who grappled with free will and predestination in “Paradise Lost.”
Tien talked about her research in a recent interview. She was sitting down at the time, thanks to the unintended consequence of spraining her ankle in a basketball game.
Question:
How did you end up in the sewer, so to speak?
Answer:
I started off in medical research. In my last year as an undergraduate here at UC Berkeley, I was working on a diagnostic system for Parkinson’s disease. Normally, the diagnosis of Parkinson’s is very subjective. One of the tell-tale signs is a person’s gait. A doctor who has seen a lot of cases will watch a person walk and make a diagnosis. I was putting thumb-size wireless sensors that measure motion on people’s feet and taking measurements.
Question:
Did it work?
Answer:
Actually, yes. There was between 93 and 97 percent accuracy in being able to distinguish between the two. It was exciting and I got to work with the director of the Parkinson’s center at UC San Francisco. I hadn’t planned on going to grad school but that project got me interested, so I applied to the Ph.D. program in civil engineering. I liked the fact that there was something really solid and concrete about civil engineering.
Question:
Literally concrete. How do we address the problem of our country's aging and increasingly unreliable infrastructure?
Answer:
I use one particular technique: Bayesian networks. These are mathematical models that help us imagine a whole system. In a water system, for example, you’ve got a whole series of interconnected pipes. You want to get water from point A to point B, but it takes a network made up of nodes, or points, and these nodes are connected by lines or links. Using these models, we can see how the individual components are related.
Question:
That sounds fairly intuitive. Has anyone looked at infrastructure this way before?
Answer:
Not to this degree of complexity. What I’m working on right now is an algorithm that will allow us to model large, complex systems.
Question:
What can this modeling do to help us avoid the disasters we've seen over the last decade?
Answer:
I look at different components of a system, identifying which might be more susceptible to failure. You want to inspect those more frequently to prevent failures. The other part of my research is geared to crisis. In the event of a system failure, or an earthquake event, I try to see which parts are the most susceptible to failure. In a big sense, what I study is the reliability of infrastructure: water systems, power systems.
Question:
So the definition of reliability includes a system's resilience to emergencies.
Answer:
Safety is a big concern. I had an interesting meeting in Sacramento with Jerry Hill’s aides. He represents San Bruno, where a natural gas pipeline burst and was really a catastrophe. His staff did a big investigation, looking into how these things could be avoided. It was really interesting to get their perspective and tell them about my research.
Question:
Did you choose your research project because of the problems with getting infrastructure projects funded?
Answer:
This was one of the things I talked about with legislators in Sacramento. Ideally, you’d be able to replace everything, but clearly that’s not possible. Given limited resources, my research is geared to where best to spend those resources. We have to ask: Which components make the system the most reliable?
Question:
Bayesian models strike most people as abstract but we get the sense that the stakes feel very real to you.
Answer:
When Hurricane Katrina hit, students from New Orleans came in to live in our dorms and go to classes with us. We had events to help them transition to their time, and help them feel at home. So I remember that.
Question:
Are you old enough to remember the Loma Prieta earthquake in 1989?
Answer:
I was young but I remember it. Growing up in the Bay Area definitely influenced me. After the earthquake in 1989, when cars couldn’t get anywhere, I realized how many people depend on infrastructure, and how important it is to keep everything working.
Question:
What brought you back after going to prep school in the East?
Answer:
I got into MIT and Berkeley. A lot of people from my high school went to East Coast colleges, but I’m glad I chose Berkeley. The environment of a big research university is really unique: the energy on campus, so many top notch faculty, so many resources. Now that I’m in my fourth year of the Ph.D., I’ve finished my requirements, so I’ve been able to take two courses at the School of Journalism. I want to take my engineering knowledge into the public debate.