In a recent study in the Journal of Transportation Security, U. of I. computer science professor Sheldon Jacobson tackled the problem of how best to screen the baggage of airline passengers to prevent terror attacks. He was interviewed by News Bureau life sciences editor Diana Yates.
The Transportation Safety Administration has been screening passengers and their baggage for decades now. Why is the science of airport screening still so uncertain?
The events of September 11th dramatically changed airport security operations. The piecemeal and reactive nature of many of these changes has resulted in large increases in costs, with only marginal increases in security, making many of these operational changes both fragile and unsustainable. Given that almost seven years have now passed since these events, there is value in reflecting back over what has transpired, but most importantly, what should have been done. In an ideal world, we can ask how the resulting experience and knowledge generated could be optimally used. Indeed, if aviation security system operations could be redesigned from scratch, using all available information, what would such systems look like? Our nation continues to face an adversary that is enamored with the drama of disrupting and inflicting damage on our commercial aviation system. Given the agility of these individuals, aviation security systems are often ill-prepared to move in step with such an ever changing threat. As a result, we are faced with cumbersome policies and procedures that often cause significant inconvenience, and in some circumstances, greater insecurity.
You and your colleagues analyzed different screening approaches for passengers deemed higher or lower security risks. What did you learn from this analysis?
The Transportation Safety Administration has made enormous investments in technology to enhance security, and more modest investments in passenger prescreening systems. Our research team (with Laura McLay of Virginia Commonwealth University and John Kobza of Texas Tech University) developed a modeling framework to measure the tradeoff between aviation security baggage screening technology enhancements and passenger prescreening accuracy. We found that unless passenger prescreening can be done well, the prohibitive costs of using such technologies may not be warranted. In essence, our analysis shows that the accuracy of the prescreening system in assessing passenger risk is more critical for reducing the number of successful attacks than the effectiveness of the checked baggage screening devices at detecting threats when few passengers are classified as higher-risk. Therefore, by focusing more attention on identifying the sources of the threats, there will be fewer threats in the systems, and those that remain will be more likely to be detected.
What simple recommendations would you make as a result of this study?
Systems like the Registered Traveler program are a step in the right direction. (Editor's note: Registered Traveler programs, implemented in about a dozen airports nationwide, allow frequent flyers to move more quickly through airport security for a membership fee.) Finding real threats is like finding a needle in a haystack. If the haystack can be partitioned, then appropriate security resources can be targeted, and hence, are more likely to find the needle. The key driver in accomplishing this is each passenger's willingness to voluntarily share information, such as with the Registered Traveler program. With better prescreening systems, technology solutions operate more effectively and can be used more appropriately. Aviation security is not a solution that "more money" will solve. We have found that doing less screening, but doing it more intelligently, can make the entire system more secure for everyone.