“Our challenge is to create metrics to judge and grade the systems we have in place, in order to eventually replace them with systems that are more efficient, flexible, and sustainable.” Ben Schafer, chair and associate professor of civil engineering

Hopkins, says Igusa, is uniquely equipped to integrate large and diverse variables into decision making through the development of MIND: the Meta-model for Infrastructure Needs and Decision-making. “MIND is a concept that we hope to develop here. It would be one part of a large collection of projects on infrastructure renewal, and serve to integrate most of these projects,” he says. For instance, one of the proposed projects would focus on developing new robotic sensors for electric transmission lines, another would develop strategies for deploying these sensors, and a third would work on predicting costs to society of service interruptions—all to determine the potential value of the sensing robots. The role of MIND would be to combine all this information to assess whether a municipality serves the public; researchers would employ both statistical and machine learning tools to create this assessment. “The output of MIND would go to the actual decision makers in the government as well as to the private infrastructure builders and operators,” explains Igusa. “Hopkins is well-positioned to develop MIND as well as many other related projects on infrastructure renewal because of our faculty expertise in these areas—things like robotics, utility markets, statistics, public health and policy—and because of our collaborative tradition, which is essential to this approach.”

One of the unique contributions the Whiting School can make to the national infrastructure renewal effort comes from this ability to create advanced models of uncertainty and apply them to fundamental issues like keeping the lights on. Recently, assistant professor of geography and environmental engineering Seth Guikema led a team that developed a computer model to predict power outages likely to occur from an approaching hurricane, indicating not only where the outages will occur but how many homes and businesses will be without electricity, and for how long. The model—described in an article published in the journal Risk Analysis—predicts the effects of future Gulf Coast hurricanes by analyzing data from five previous storms that ravaged the area: Dennis, Danny, Georges, Ivan, and Katrina. It is actually two different kinds of data sets combined to create one snapshot of the likely future outcome. A detailed accounting of the electricity infrastructure (including location of poles, transformers, sub-stations, and other physical assets) is married to variable information unique to each storm, including wind speed, soil saturation, total precipitation, and related measurements. By running the numbers through a complex set of algorithms, Guikema’s team is able to make real-time predictions of where power losses will occur, an invaluable tool in enabling utilities to cost-effectively marshal cleanup and restoration resources.

Guikema’s research has immediate real world applications—particularly if global warming scenarios predicting more frequent and more disruptive weather patterns hold true—and was funded in part by a Gulf Coast utility company to improve resource management. But the real challenge, says Guikema, is to integrate all interrelated infrastructure in one model. “What we want to be able to do is look at all interdependent systems during a disaster— power, cellular, water, cable, and landlines— to get an idea how really large-scale systems respond. Fundamentally this comes down to the question, How do we define and measure infrastructure? There is a lot of basic engineering research needed to figure out how these systems respond.”

The need for advanced and accurate predictive capabilities is becoming ever more acute as the infrastructure’s built environment ages and new system challenges arise from global climate change. Baltimore planning chief Tom Stosur worries about the prospect of a flooded downtown: “The big concern is what the rise in sea levels means for a coastal city like Baltimore. A rise of even a few inches makes a huge difference. If sea levels rise and storms pick up, suddenly the 100- year flood becomes the five-year flood. Storms are likely to have a huge impact but no one currently is doing control or planning for this. We are in a learning mode.” Climate change issues are trans-national,countries around the world, warns alumnus Ralph Gakenheimer ’57, a professor of urban studies and planning at MIT. So even as Whiting School research focuses primarily on the needs and challenges faced by infrastructure in the Mid-Atlantic region, the discoveries and lessons learned will have global implications.

“Here in the Mid-Atlantic region we have a great example of a completed, in-place, developed world infrastructure,” says Schafer, “which means it’s one of the most difficult in the world to fix. If we were starting from scratch you could certainly come up with something better. But the bottom line is that people get upset if they can’t flush their toilets for a week. So you can never bring the infrastructure offline.

“Our challenge is to create metrics to judge and grade the systems we have in place, in order to eventually replace them with ystems that are more efficient, flexible, and sustainable. The goal is to make the infrastructure less ad hoc.”

The coming changes and challenges to infrastructure provide the Whiting School with an opportunity to stake a leading role in discovering, designing, and implementing transformative technologies that can reshape the world. “A perfect storm has been brewing for a while, and it presents a compelling opportunity,” says Dean Jones.

“We have a tremendous advantage because of our cross-disciplinary approach. We are very well aligned with what is needed right now.”