Researchers Get $12.5 Million To Study Network Complexity

• By Dian Schaffhauser
• 09/26/13

The United States Army has decided that investing millions of dollars into research that examines why an outage in the power grid might cause failures in gas distribution several states over or how a malicious tweet could bring down an entire stock market system is worthwhile. The focus: to better understand behaviors of collective phenomena in complex systems — networks — and figure out how they can be controlled.

Researchers at the University of California, Davis Complexity Sciences Center recently received funding for two separate projects. Each will get $6.25 million over five years from the U.S. Department of Defense's Multidisciplinary University Research Initiative.

One project, led by Raissa D'Souza, professor of mechanical and aerospace engineering and computer science, will research how collections of different systems interact. Ultimately, she and her team hope to find ways to intervene in networks in order to better control them or prevent catastrophic breakdowns.

The other project is being directed by Jim Crutchfield, professor of physics and director of the Complexity Sciences Center. His team will examine how networks can manipulate both information and energy to develop what Crutchfield is calling a new "physics of information." Future applications, he said, could range from the creation of "microscopic computers" to controllable "molecule-sized machines."

"Complex systems and complexity, traditionally, [is viewed as] collective phenomena — complexity that arises from a large number of simple elements, people, atoms, molecules, genes or proteins," said D'Souza in a video about her project. "When they come together, you can observe behaviors that you wouldn't have anticipated from the individual agent."

The proof of that often comes to the surface during emergency situations, such as earthquakes or hurricanes, she noted. These events, she said, "unearth couplings [of] things we didn't understand before."

For example, a local power grid relies on communication networks, is influenced by networks of traders in energy markets, is affected by natural disasters in other areas, and must answer to policy and public sentiment. A failure in one place could start a domino effect that impairs more than just the grid's power customers.

The work being led by D'Souza will draw on theoretical research as well as specific examples, such as disaster recovery in critical infrastructure. "Engineers have a lot of practical experience with how networks behave, and we want to translate that real world experience into theory," she said.

Other aspects of these multi-disciplinary initiatives include studying monkey colonies at the University of Wisconsin's Center for Complexity & Collective Computation, networks of nano-electrical mechanical devices in the laboratory at the California Institute of Technology, and trade and political alliances between nation-states from UC Davis.

The U.S. military too depends on complex networks for communications, supplies, and transportation, all of which are affected by shifts in social and political networks, she added. This new research could help understand how to prevent a failure in one area from generating new failures in other areas.