Researchers are launching a project that could help predict when critical services will fail and provide utilities with strategies on how to best respond to unprecedented disaster.

Kalyan Piratla, center with arms outstretched, shows his team some of the research he has done. The others are, from left, Kumar Venayagamoorthy, Ilya Safro and Fred Switzer.

Kalyan Piratla, center with arms outstretched, shows his team some of the research he has done. The others are, from left, Kumar Venayagamoorthy, Ilya Safro and Fred Switzer.

The team will work with utilities to identify which part of the Charleston peninsula to study and expects that the computer models they develop could be applied elsewhere.

Kalyan Piratla, an assistant professor in the Glenn Department of Civil Engineering, is leading the two-year project with $500,000 in funding from the National Science Foundation.

“The most exciting part is the interplay between psychology, engineering and computing disciplines,” Piratla said. “We’re used to working in silos. I want to really understand how interactions can be enhanced for decision-making in the face of crisis.”

The idea behind the research is that the power lines, pipes and other infrastructure making modern conveniences possible are interconnected, even if the various pieces are run by different utilities.

Most troubling, the failures can begin to cascade once they start.

For example, the Southwest in 2011 lost a significant amount of its gas-production capacity at the same time cold weather drove up the demand for gas. The combination led to reduced pressures in gas-transmission pipelines.

About 50,000 homes and businesses were left without natural gas. And when some power plants couldn’t get enough natural gas, about 1.3 million electric customers lost power.

How quickly the power, natural gas and water start flowing again after disaster depends in large measure on how decisions are made behind the scenes in utility control rooms, Piratla said. In a disaster, operators need to make critical choices, such as where to send crews and which areas to prioritize.

Piratla and his team aim to develop computer simulation models that map out the myriad interconnections in infrastructure. The models will help researchers better understand how one failure can lead to another and begin to fall like dominoes, he said.

They also plan to build a computer model that enables control room operators make best decisions and communicate more effectively. The models could serve as a valuable tool for emergency preparedness training and policy-making, Piratla said.

The models could also pave the way for a more advanced program that would give utilities real-time suggestions on how to respond when unprecedented disasters strike, he said.

Researchers are starting with Charleston because it is vulnerable to several hazards, including hurricanes, storm surges and earthquakes.

“For this reason, it is a good case study for investigating the vulnerabilities arising from interdependent critical infrastructure,” Piratla said.

The project will be the first study of its kind to examine how utility control room operators make decisions and how to best inform them of unprecedented cascading failure risks in critical infrastructures.

Undergraduate students will be involved in the research through the university’s Creative Inquiry program. Graduate students will also be involved.

James R. Martin, chair of the Glenn Department of Civil Engineering, congratulated Piratla on the grant.

“Dr. Piratla’s work in intelligent infrastructure is helping keep Clemson on the cutting edge of an important, emerging topic,” Martin said. “He has assembled a team of researchers with impressive credentials. They are well-positioned to have global impact.”

Co-principal investigators on the grant are: Kumar Venayagamoorthy, the Duke Energy Distinguished Professor of Electrical and Computer Engineering and director of the Real-Time Power and Intelligent Systems; Fred Switzer, professor of psychology; and Ilya Safro, an assistant professor in the School of Computing.