Jump to: site navigation, content.

Thursday, February 25, 2010

UT Dallas professors suggest efficiency techniques for airport terrorism defense systems


Local researchers say passenger profiling, when coupled with airline screening systems, could strengthen airport security.

— An attempt to set a trans-Atlantic jetliner aflame above Detroit on Christmas Day has reignited a simmering debate about improving airport security on the ground.

Huseyin Cavusoglu

Huseyin Cavusoglu

As the U.S. redoubles its efforts to tackle terrorism in the air nearly a decade after 9/11, researchers at the School of Management at UT Dallas say that passenger profiling, when coupled with airline screening systems, could strengthen airport security while addressing concerns about passenger inconvenience and civil rights.

Dr. Huseyin Cavusoglu and Dr. Srinivasan Raghunathan, information systems professors at the School of Management, along with grad student Byungwan Koh, recently completed a study on the value of passenger profiling in the transportation industry. The paper, “An Analysis of the Impact of Passenger Profiling for Transportation Security,” which offers one potential way to handle the nation’s imperfect airline screening systems, will be published in the academic journal Operations Research some time in 2010.

“The biggest challenge is that we are trying to find a needle in a haystack because the fraction of criminals in the population is very small,” Cavusoglu said. “What we say is that it’s not enough to have a screening system, but if the screening system and profiler complement one another, we have a better shot at finding the needle.”

Although previous studies explore the effectiveness of profiling, Raghunathan says this is the first major study that examines how profiling can work with screening devices to create a system that balances the needs of airlines with that of customers. The UT Dallas professors began looking for a security system that provides the benefits of catching attackers while effectively outweighing the costs – in money, time, inconvenience, privacy, and liberties.

In the study, the researchers first considered a baseline scenario in which all passengers classified as likely attackers by a profiling system were manually inspected while all others were sent through a screening system, the approach the Transportation Security Administration (TSA), the government agency charged with transportation security, took when it began using a profiling system after 9/11.

In a second scenario the researchers proposed, all passengers classified as potential attackers by a profiler were sent through one screening system while the remaining passengers were sent through a different screening system. So whether or not a profiler identified passengers as likely attackers, all passengers went through a security system rather than being singled out, and all underwent a manual inspection.

Srinivasan Raghunathan

Srinivasan Raghunathan

The researchers found that although the TSA’s previous approach targeted inspection of potential attackers, such inspections did not necessarily improve the potential for catching attackers; nor did this approach reduce the inconvenience to innocent passengers. On the other hand, the School of Management research trio found, if the TSA uses a two screening-device architecture along with a profiling system -- configuring each screening device optimally for each of the two groups of passengers -- profiling would minimize passengers’ inconvenience while maximizing the chances of catching attackers.

Using this type of system “may not only blunt the criticism that profiling is discriminatory -- since every passenger will have to pass through a screening device, but also benefit normal passengers and the overall society economically,” the paper said.

“In this kind of setup, the two groups of passengers (those who have been identified as possible attackers and those who have not) would be sent through screening systems, but two different screening systems. One could be very sensitive and one less sensitive. Then you’re treating them somewhat in the same fashion in the sense that both groups will have to go through a screening system. That might take away some of the criticism that the opponents of profiling have,” Raghunathan said.

Cavusoglu and Raghunathan, who have studied security issues in many other areas, began looking at what the TSA was doing to keep the skies safe following 2001’s terrorist attacks.

“After 9/11, there was a lot of focus on security, and we, along with many others, began to look at the TSA’s strategies, whether they were successful in identifying attackers and whether they benefitted passengers. We used our prior work on information security and began looking at whether profiling systems, which the TSA had tried implementing in the past, really helped achieve security,” said Raghunathan.

Although the TSA over the last few years has introduced a number of screening devices -- ranging from simple metal detectors to sophisticated, three-dimensional X-ray machines -- to detect would-be criminals, these systems often create false alarms. The danger, say researchers, is false alarms not only frustrate innocent passengers but also can cause security personnel to ignore the signals, defeating the devices’ intent.

Screening and profiling systems are similar because both are designed to detect security threats; however, their functions are different, says Raghunathan. Screening systems, such as scanners and pat-downs, help reveal prohibited items that could be used to carry out an attack. Profiling, on the other hand, is “not a legitimate substitute for real evidence,” according to the paper. Rather, it is a system designed to identify attackers who may carry forbidden items.

After 2001, the TSA introduced several versions of a profiling system that classified passengers into “high-risk” and “low-risk” categories based on information such as gender, race, and age, as well as behavioral screening information, such as looking at flight information, frequent-flier status and mode of ticket payment.

But on Christmas Day, 2009, U.S. intelligence agencies and airline security personnel missed many of these kinds of behavioral clues when Umar Farouk Abdulmutallab was allowed to board a Detroit-bound airplane. The 23-year-old, accused of the failed bombing attack, paid cash for an expensive ticket, was traveling without any checked luggage over a two-week trip during the busy holiday season and had raised security concerns among some agencies and for his father.

“Although screening devices are considered to be highly efficient and effective in detecting explosive items, they are not foolproof, as we have seen in a recent incident in which Umar Farouk Abdulmutallab tried to blow up a jetliner,” said Cavusoglu. “If there had been a profiler that relied on the intelligence that was available at that time -- such as a specific warning from his father and being on a watch list -- this person could have been identified by the profiler.”


Source: University of Texas at Dallas



Share: 
del.icio.us Digg DZone Facebook Fark Google Google Reader Reddit Slashdot StumbleUpon Technorati Twitter YahooBuzz YahooMyWeb YCombinator


Sponsored Links

What do you think?

:

:

 Find out how to share this comment with Facebook

See more stories in:


Latest comments...

Peter Max

Haha, unlisted. It has been corrected.


Pop icon Peter Max exhibits paintings at the Crescent Hotel this summer

"humbleness"??????

Um, Mr. Means (reporter), your fourth-grade English teacher is going to smack yo


Peter Max

Taylor Swift looks an awful lot like the Texas flag.

Must be that modern art stuff. Huh?


Stay connected