UB researchers investigate how ergonomics in airport screening can affect the saftey of our nation's air travel
Story by Blair Boone, Ph.D. '84
In the first six months of 2004, U.S. airlines carried nearly 308 million domestic passengers on almost five million flights, reflecting an eight percent increase in passengers over the same period in 2003. For a travel industry still recovering from the shock of 9/11, that’s an encouraging number. However, for the Transportation Security Administration (TSA), the federal agency created in the wake of the terrorist attacks to improve screening of air passengers and their luggage, it’s a mind-boggling challenge.
Photo by Paul Francis
That’s why the agency selected Colin Drury, UB Distinguished Professor and chair of the department of industrial engineering in the School of Engineering and Applied Sciences, to receive a $538,000 grant to establish the Research Institute for Safety and Security in Transportation (RISST). Drury is an internationally recognized expert on how human factors affect the accuracy and efficiency of aviation inspection. He has more than 30 years’ experience in human factors research, including 15 years of specialized research on aircraft inspection and safety, and more than 200 publications on industrial process and quality control.
The institute is building on work Drury has done for the Federal Aviation Administration (FAA), applying and extending human factors principles, or ergonomics—how human beings interact with machines and technology—to meet the new security requirements of the TSA. Research at RISST will examine how security personnel can more efficiently use, and with a lower error rate, screening equipment, including everything from metal detectors, X-ray machines and CT (computerized tomography) scanners to trace equipment that tests for explosives residue on passengers and baggage.
“We study intermediate-term improvements to the security system,” Drury explains. “We do it because there’s a human being in every security system, and unless you treat the human being with the same care with which you treat the machines, your system isn’t going to be as effective as you thought.”
According to Drury, while technology such as metal detectors or CT scanners is specific to each task, what screeners are doing in each instance is essentially the same: actively searching for something, and then using different decision-making techniques to decide if they’ve found something dangerous.
“These functions of ‘access-search- decision’ are the same things you’d be doing if you were looking at a screen for checked bags or carry-on bags, or even if you were looking at multiple screens to keep an airfield perimeter secure,” Drury says. “If we find out, ‘Here’s a really good training system,’ or, ‘Here are some good principles of display design,’ then it should apply to all of them.”
The TSA has already been doing similar research but with short-term goals that are related to selecting the equipment—hardware and software—and developing training and user techniques. In fact, prior to 9/11, the FAA was developing many of the security procedures that are now being used. These FAA personnel and techniques were transferred to the TSA when the new agency was created on November 19, 2001.
The UB researchers form the only two groups focused specifically on human factors in inspection tasks. Drury cites a program at Clemson University—headed by one of his former graduate students—as the only other center with a similar focus.
“Being human factors–oriented, we’re concerned with how you can change the hardware, software and organization to fit the person, not the other way about,” he says. “If you don’t have all that right, it’s going to be extremely difficult to fit a human operator into it.”
While the institute will have the capability to install in the laboratory actual equipment that is used at airports around the country, Drury says that may not be necessary. “Much of the technology base lets the operator interact with the technology through a computer screen. So we can do a lot of the research with simulators and emulators where the operators see what they normally see,” he explains.
As anyone who has waited in line at an airport knows, the TSA screeners face enormous pressures to be both accurate and fast. However, as Drury points out, there are problems. While there are huge consequences to missing an actual threat, there’s a penalty for false alarms. Furthermore, clearing an airport terminal for a nonexistent threat, for example, would cause a significant disruption in the transportation system and try the public’s patience, possibly leading to less tolerance of security procedures. And the pressure to keep flights on time is always there, even as the amount of air traffic is rebounding.
“The sheer volume of travel precludes a 100 percent foolproof system,” he says. “The consequences of a mistake are so high, and the probability of an attack per passenger is so low, it makes threats quite difficult to guard against.”
And for those who have a fear of flying, Drury adds, “Statistically, the scariest thing about flying is still driving to the airport.” To keep things safer and keep the lines moving, he advises travelers to follow the advisories on the TSA Web site (www.tsa.gov), such as removing your jacket, watch, keys and shoes before getting to security. He also advises travelers to be alert to odd activities, citing incidents such as the transatlantic flight with the man who became known as the “shoe bomber” and who was thwarted in setting off the bomb in his shoe by alert passengers.
And not all of the TSA’s screening techniques rely on technology for their effectiveness. Random screening isn’t high-tech; however, it works precisely because it is random. “People get really annoyed when they’re selected for extra screening,” says Drury, “but one of the ways to use that technique effectively is to have it entirely random, so you can’t game the system. I get picked out from time to time when I fly. If I get picked out, great. That means it’s working.”
The technology screeners do employ is impressive. A CT scanner typically screens checked baggage, creating a three- dimensional image of the contents of each bag. Checked baggage scanners are set to detect things such as explosives, rather than items such as knives, which can be legally carried therein, but are banned from carry-on luggage. Carry-on bags, overcoats and jackets are X-rayed at the security checkpoint by new, more sophisticated equipment. And, of course, each passenger must pass through a metal detector. Selected travelers may also be screened by new EDS (Explosive Detection System) devices that detect minute amounts of explosives.
While metal detectors have been used in security applications for years, CT scanners are relatively new for that task. Also, the improvements in hardware and software necessary to achieve the tremendously fast scanning rates required for baggage screening have given rise to the potential for new problems for humans operating the system.
As an example of the kind of problem the institute will examine, Drury cites the following: “If the system alarms, is the information that the human gets on the computer screen the information they really need to make a decision? How should we display it? Are there various color codes we could use, or will we do just as well with black and white?”
He continues, “If you want to manipulate the bag for a better look, how should the human-machine interface help you turn that bag around? Should you have up-and-down and sideways arrows, a joystick, or do you grab and turn it as though you were handling the case itself?” Other potential research questions include helping determine what equipment operators should train on, and how often they should be retrained.
Another critical issue is keeping operators alert during long shifts as they watch a computer screen. Along with fatigue, Drury describes the problem as, “If you give a person a job to look for X and X never turns up, they’ll get worse and worse at that job.” In contrast to millions of prohibited items being successfully confiscated by screeners at the passenger security checkpoints, alarms for items in checked baggage tend to be much less frequent. The TSA uses a system called Threat Image Projection System, or TIPS. “A quite nice piece of human factors engineering,” Drury remarks. As the operator checks the image of the luggage on the computer screen, TIPS can project a prohibited item to make it appear as if it were in the checked bag. This way, it tests operators for proficiency and helps keep operators alert. In addition, it rewards them for a job well done.
While the RISST will be a primary focus of his upcoming research, Drury’s human factors work has had many less high- profile, but equally important, applications. As the founder and former executive director of UB’s Center for Industrial Effectiveness, he has worked with industries in the Buffalo region to improve competitiveness, helping create and retain thousands of jobs. He has also served as a private consultant on human factors to a wide range of private and public organizations, including the U.S. Postal Service, and has done pro bono work for the National Research Council and the National Transportation Safety Board. “The work we’re doing now will hopefully make it easier for people to do a more effective job of security,” says Drury. “I hope these levels of security won’t always be necessary, but I don’t want my children and grandchildren to have a less secure world than I have.”
Blair Boone, Ph.D. ’84, is a Buffalo-based freelance writer.