Emergency Evacuation: Get a road of this

      Mix together a Category 5 hurricane, monumental grid lock, and a creative civil engineer — and what you get is a clever new way to find the path of least resistance through stop-and-slow traffic on jammed highways and major surface arteries during emergency civic evacuation.

   The new technique monitors Bluetooth signals from commuter cell phones to derive travel times, road speeds and vehicle movements. Normally, Bluetooth wireless communication technology connects and exchanges information for cell phone hands-free headsets, wireless keyboards, Internet access for personal digital assistants and wireless networks for laptops and personal computers. Darcy Bullock, a professor of civil engineering at Purdue, found a way to use wireless Bluetooth signals to provide low-cost tracking data on everything from the speed of the morning commute to the sluggishness of airport security lines, to aiding in the emergency evacuation of entire cites.

   "I was on the faculty at Louisiana State University in the 1990s, and later watched how hard my colleagues worked to dynamically plan and adjust routes during the Katrina evacuation," Bullock says. Planning evacuations on the fly, while the storm is raging, leaves ample room for improvement.

   When the wind died back, Bullock set out to see if he couldn't devise a method that uses pervasive Bluetooth signals from cell phones and other wireless gadgets to track and monitor how long it takes vehicles to travel from one point to another on freeways or major surface arteries.

   Such a capability would give law enforcement and emergency managers a layer of visibility they don't currently have to devise better routes and allocate resources. Although the system was designed to be used during mass evacuations by emergency managers and law enforcement, a major benefit of Bullock's system is it can be used every day to time traffic signals more intelligently, or manage traffic through construction work zones to reduce congestion.

Phone tree

   This is how it works. Every Bluetooth device has its own distinct digital identification signal, called a media access control, or MAC. By mounting a simple whip antenna on power poles adjacent to roadways or on freeway overpasses, the digital MAC address of devices can be easily read, logged and time-stamped. If these MAC addresses are logged at multiple locations, the unique MAC addresses can be matched, and the difference in time stamps can be used to calculate travel time.

   Law enforcement and emergency planners could use the data to make informed decisions about how to allocate resources or adjust evacuation routes during civic crises. If the information were made public, commuters or evacuees could access the travel-time information using the same portable electronic devices that make the system possible, or the data could be used to display travel times on message signs.

   "Information is a commodity people are aggressively seeking, and this method promises to cost-effectively provide information that has never been widely available to either commuters or emergency managers," Bullock says.

   Last year, Bullock's Bluetooth scheme was implemented at six interstate locations on I-465 and I-65, and two major surface arteries in the Indianapolis metropolitan area the week of the 2008 NASCAR Brickyard 400 auto race. On race day, nearly 19,000 unique MAC addresses were captured at the eight locations.

   Using that information, travel time estimates were provided to Indiana Department of Transportation (INDOT) mobile data terminals. After the race, INDOT processed the data to obtain travel time plots and origin-destination matrixes to provide a quantitative evaluation of race day traffic management operations.

   Bullock believes his Bluetooth technique improves on existing traffic monitoring methods, most of which rely on roadside sensors or closed-circuit television cameras.

   "Collecting travel time is very difficult to do with fixed infrastructure, such as loop detectors that only measure spot speeds and extrapolate segment travel time," Bullock says. "Our Bluetooth technique provides direct measurement of a sample of the population that we believe provides a much more robust indicator of travel time."

   While this may resemble another devious way Big Brother has found to invade personal privacy, by using our cell phones to follow where we go, in actuality only the anonymous MAC address is transmitted. Since MAC addresses are not associated with individuals, personal information is unavailable. However, unlike video surveillance, bar codes and radio frequency identification tags, skeptics have ultimate control over this tracking system; Bluetooth transmission is disabled when cell phones are turned off.

Eye in the sky

   Another cell phone-based traffic management idea has emerged in Japan.

   Researchers at Kyoto University are working on an emergency evacuation system based on GPS-enabled mobile phones. The Kyoto system shifts the emergency management paradigm from one information push to all users to information individualized to each user.

   "Current evacuation or navigation services simply broadcast the same instructions over a large area; but what is needed is a system that can provide individualized instructions to each person," explains Toru Ishida, a professor in the Kyoto Department of Social Informatics. Ishida says the use of GPS-capable cellular phones enables the provision of personalized navigation instructions to suit each user circumstance.

   "In this system, each user receives updated maps and instructions from a central control directly on their mobile devices," Ishida says.

   Under the Japanese system, evacuees can check their current position, indicated by red dots, on maps transmitted to them on demand. Evacuation sites are indicated by blue rectangles on the map. The movement of other evacuees is shown by green arrows, making it easier to avoid crowded areas. Other areas to avoid, such as collapsed buildings or structure fires, are also depicted on the maps.

   Neither Bullock's Bluetooth tracking or Ishida's GPS scheme are panaceas for law enforcement or emergency managers charged with developing complex mass evacuation strategies, but each may play a role of some kind in future efforts.

No wheelchair left behind

   Developing effective mass evacuation will require input from many desks, including coordination across all levels of government agencies, as well as non-government organizations.

   "Disasters do not recognize jurisdictional boundaries or socioeconomic status," says John Renne, a professor of transportation studies at the University of New Orleans. In the days following Hurricane Katrina, Renne launched the Transportation Equity and Evacuation Planning Program at the UNO Transportation Center, the goal of which is to provide research and outreach to improve evacuation planning and practice for all members of society.

   Renne is especially concerned about evacuating the low-mobility population — people in low income areas, and people with physical and mental disabilities.

   "The poor and infirm are usually more impacted because they have a higher level of dependence on government for evacuating," he says. Renne recently authored a 110-page study for the Federal Transit Administration, titled "National Study on Carless and Special Needs Evacuation Planning: A Literature Review."

   According to the 2000 U.S. Census, seven cities had carless populations even higher than the 27 percent in New Orleans, including Washington, D.C., Baltimore, Philadelphia, Boston, Chicago and San Francisco — all with carless populations between 29 and 37 percent. Nearly 60 percent of the population of New York City has no car.

   Nationally, approximately 10 percent of the population is disabled and many of these individuals don't drive, even if a family car is available. As the population ages, more and more people will become mobility-restricted.

   "Even the elderly who have cars may be reluctant to drive them during a mandated long-distance evacuation," Renne says.

   Disabled and low-income groups face disproportionate risk. During Katrina, for instance, more than 1,300 (71 percent) of the 1,836 deaths were people over age 60, and more than 860 (47 percent) were over age 75.

   "Until we begin to plan and coordinate across municipal and county government to coordinate with transit agencies and nonprofit service providers for vulnerable population groups, we will fail to have effective mass evacuations," Renne says.

   Renne says the solution is not complicated, but is rarely put in play because it requires a combination of strong leadership, adequate funding and resolute commitment.

   "Currently, there are no standards even specifying which government agencies should have the mandate for planning and funding regional evacuations, especially with respect to accommodating the needs of the carless and other vulnerable population groups," Renne says.

Escape clause

   Renne says his research has found that cities that have had experience with major evacuations, such as Miami, New Orleans and Houston are much better prepared than cities that have not had the same experiences. For example, the evacuation of New Orleans during Hurricane Gustav in 2008 was much improved over Hurricane Katrina in 2005, and the evacuation of Houston-Galveston during Hurricane Ike in 2008 was much more efficient as compared to Hurricane Rita in 2005, he says.

   "Cities like New Orleans and Houston were embarrassed during Katrina and Rita, so a lot of effort across local, state and federal levels was invested to ensure mistakes would not be repeated," Renne says.

   However, cities in other parts of the country that have not had mass evacuation incidents are not as well prepared because evacuation planning has not been as much of a focus for those leaders.

   Brian Wolshon, a civil and environmental engineering professor at LSU, recently completed a study for the Transportation Research Board titled "Transportation's Role in Emergency Evacuation and Reentry."

   He says in some cases, formal detailed evacuation plans do not exist, even when there are fairly regular evacuation requirements. In Southern California, for example, conditions associated with wildfires are too variable to establish firm evacuation routes. The number of people threatened, the size of the region, amount of warning time, available routes and shelter destinations are at whim of changing winds and weather conditions.

   "As a result, emergency preparedness and response agencies find it more effective to work from a general framework that permits flexibility to respond to rapidly changing conditions," Wolshon says. Traffic modeling systems can be useful here.

   Renne and Wolshon have used a traffic modeling program known as TRANSIMS to help model evacuation strategies for New Orleans. But ultimately, Renne believes traffic management itself is not the biggest issue.

   "The major problem is that there is no policy on which agency takes responsibility for planning and implementing regional-scale evacuations," he says. "Unless there is a mandate for local governments to collaborate, we will continue to fail to prepare our regions for mass evacuations."

   Douglas Page writes about science, technology and medicine from Pine Mountain, Calif. He can be reached at douglaspage@earthlink.net.