INDUSTRY

By Karen Green, NCSA -- In the summer of 2000, a new organization called the Multi-Sector Crisis Management Consortium (MSCMC) formed in the Washington D.C area and held its first meeting at the Alliance Center for Collaboration, Education, Science, and Software (ACCESS) in Arlington VA. The MSCMC set out to examine how cutting-edge information technologies can be used by public and private sector organizations to prepare for, respond to, and mitigate local, state and national disasters, including natural disasters and terrorist attacks. ACCESS hosted the monthly meetings as well as several workshops and NCSA and Alliance researchers provided information about new technologies--including data mining and knowledge management, computer visualization, and intelligent sensors. Everyone involved knew the consortium's work was important, but none foresaw the sense of urgency that would pervade its first anniversary meeting on Sept. 19, 2001. No one had predicted the events of Sept. 11, 2001, the day when the unthinkable happened. In light of recent events, what had been planned as an informational meeting and one-year anniversary reception became a serious discussion of questions now on the minds of millions of people. How do we prepare for and manage national-scale crises? How do we build a more effective emergency response system that fully utilizes cutting-edge communication technologies? How can information technology be put to use to decrease the threat of terrorist attacks? John R. Harrald, a professor of engineering management and systems engineering at George Washington University and director of the university's Institute for Crisis, Disaster, and Risk Management, led a discussion about these questions. Participants included a full house at the ACCESS center and attendees from several remote locations who participated via the Access Grid. According to Harrald, the key to a better emergency response system is to bring together the people who understand the needs and problems associated with disasters (the emergency response teams) and the people who can help solve those problems (the developers of innovative technologies in industry and academia). "First responders need to be aware of the technology that is available and its benefits, but education needs to go both ways," he said. Those who develop new technologies, structural engineers, and data analysis specialists need to go to the scenes of natural and man-made disasters, according to Harrald. This first-hand experience will help them understand the problems routinely faced by emergency workers. Better communication among the various agencies that respond to disasters--from local police and firefighters to the state National Guard to the FBI--would also improve disaster response, since most groups have their own communication systems and information resources and do not often share that information with other groups. The consortium leadership challenged participants to work together to rapidly deploy advanced tools and technologies into agencies that deal with crisis management and emergency response. In addition, data mining techniques and computer simulations could be used to predict risks and decrease the loss of life and property caused by disasters. Michael Welge, director of NCSA's Automated Learning Group, presented information on data mining techniques for turning data into useful information. Effectively responding to an emergency requires more information and organization than handling day-to-day activities, he said. The ability to analyze vast amounts of data quickly and use it to determine emergency response can save lives, prevent panic, and help determine who or what is responsible for the crisis. Data to knowledge (D2K) techniques analyze data to predict outcomes or suggest solutions to problems. Such data analysis can provide a wealth of information, including the best escape routes to use in an evacuation, the probable behavior of a building hit by a bomb, or the most effective way for multiple emergency response teams to divide management responsibilities in a large-scale crisis. "In New York, the phones and cell phones all crashed," noted an audience member at ACCESS. "We need to start documenting these things. We need to have the data to examine so we can determine what happened." In addition, remote sensors can provide much data useful for preventing and managing disasters. Sophisticated sensors at airports, for example, could gather a wide range of information about airline passengers. This information could be used to weed out potential terrorists, or at the very least identify terrorists after a crime has been committed. But even with all the technological tools available, Americans and their allies need to realize that terrorism is as unpredictable as natural disasters, said Harrald. "Our tendency is to fix the latest problem, such as airline safety," he said. "But the reality is you can stop a threat or close a door, but our society is still going to have to deal with the fact that we are vulnerable." Advanced technology centers like ACCESS, which offer high-speed connectivity and the Access Grid distributed collaboration technology at a neutral location, could play a role in the future of crisis management. Last year, ACCESS was the master control center for a simulation that involved three terrorism scenarios in Portsmith, NH, Denver, and Washington, D.C. ACCESS was chosen partly because of its location in metropolitan Washington D.C. and partly because it already had the information infrastructure needed to serve as a command center for a national emergency. The crisis management community also needs to develop a more dynamic approach to preparing for and responding to disasters, said Tom Prudhomme, NCSA's director of External Programs. That approach should include real-time collaboration using a range of communications techniques as well as education and training. "We need a quicker time cycle for adopting new plans and updating plans," said Prudhomme. "We need to use data mining techniques to analyze sensor data for early detection of risks and be able to analyze data in real time. That will help us detect possible events and understand our vulnerabilities." -- This story originally appeared in NCSA's Access Online