Stanford Business

AUGUST 2006


Calculating Ways to Stop Terrorists


Photo-illustration by Kevin Irby

Mathematical models are changing the way we screen cargo, view terrorist cells, and fight biological weapons.

by John B. Stafford

Throughout the 20th century, mathematicians used equations to combat many national security threats, at their apex breaking German codes and hastening the end of World War II. During the Cold War, game theory helped formulate military strategies. The Sept. 11, 2001, terrorist attacks gave birth to a “war on terrorism,” and mathematicians are once more employing their knowledge to fight this new kind of war.

At the Graduate School of Business, professors Lawrence Wein, Hau Lee, and Seungjin Whang have developed mathematical models designed to improve security at U.S. ports. Professor Glenn Carroll, along with alumnus J. Richard Harrison, PhD ’86, has used modeling to interpret how terrorist cells retain their organizational coherency. Their models show, nearly five years after 9/11, the nation’s ports still lack adequate security and that a strategy of disrupting terrorist cells by removing its members may be largely ineffective. On the other hand, modeling has led to some security improvements such as better fingerprint checking at U.S. borders.

“As an applied mathematician, I turn issues and problems into the world of mathematics,” explains Wein, the Paul E. Holden Professor of Management Science, who teaches operations classes about business processes used to deliver goods and services. Since 2001, his research has focused on terrorism issues. He performed the first mathematical analyses of a hypothetical botulism poisoning, anthrax outbreaks, and smallpox infections.

“One overriding theme of my work is that all these homeland security problems are operational problems,” Wein says. “So just as McDonald’s needs to get hamburgers out in a rapid and defect-free manner, so too does the government have to get vaccines and antibiotics out and test the borders for nuclear weapons or terrorists in a rapid and defect-free manner.”

PORT SECURITY MODEL PROPOSED

Along with Stephen Flynn of the nonpartisan Council on Foreign Relations, Wein developed a model to harden security at U.S. ports and prevent terrorists from bringing a nuclear weapon into the country inside a shipping container. It draws on aspects of nuclear physics, queuing theory, and game theory. The model, recently accepted for publication in the journal Risk Analysis, has received the attention of the California State Assembly. In April, Wein testified in Long Beach, Calif., before the Assembly Select Committee on Ports.

“The detonation of a nuclear weapon on U.S. soil is the most feared type of terrorist attack,” Wein said. “With aggressive post-9/11 engagement strategies in place over U.S. airspace, and the difficulty of traversing the Atlantic or Pacific oceans in a pleasure or fishing vessel in an undetected fashion, the most obvious vehicle for smuggling a nuclear weapon into the U.S. is the maritime shipping container industry.”

As evidence of ineffective port security, he pointed to two successful attempts in 2002 by an ABC News team to smuggle depleted uranium, which he said should be easier to detect than the highly enriched uranium used in nuclear weapons, into the country inside two 40-foot containers.

U.S. Customs agents actively inspect a minority of shipping containers based on information produced by the specialized Automated Tracking System (ATS), which analyzes shipping manifests and customs documents to identify “suspicious” containers for inspection.
“The ATS flags approximately 6 percent of containers, and most ports barely have the inspection resources to test these containers, giving nearly all of the remaining 94 percent of containers a free ride through the inspection process after clearing the ATS,” Wein testified.
He and Flynn suggest the government instead use a multi-layer, passive screening system on every container coming into the country. Customs would photograph a shipping container’s exterior, pass each container through a radiation monitor, and take gamma-ray images of the container’s contents. If terrorists shielded a nuclear bomb with a heavy metal such as lead to conceal it from the radiation monitor, the gamma-ray imaging would allow Customs agents to see the shielding and mark the container for inspection. According to Wein, the cost of inspection in this manner ranges from $6 to $25 per container, which he and Flynn propose be charged to the shipper. He urged the state legislators to take the initiative because “the federal government is dragging its feet.”

After the Bush administration cleared a Dubai company to run several U.S. ports, the American public became alarmed by port security, however, and in May, the U.S. House of Representatives passed a bill that would provide $7.4 billion to improve port security. Authored by Reps. Dan Lungren and Jane Harman, both of California, the bill would provide funds to hire 1,200 more port inspectors and require the Department of Homeland Security to develop a plan for the eventual radiation screening of containers. Wein characterized the bill as a “step in the right direction,” but noted it does not appear to contain all of his recommendations.

SECURITY THROUGH SUPPLY CHAIN MANAGEMENT

In addition to Wein’s recommendations, Hau Lee, the Thoma Professor of Operations, Information, and Technology, has argued there is an opportunity to bolster port security before the containers even reach the United States. Using lessons from quality assurance programs, also known as “total quality management” principles, Lee, whose research focuses on supply chain management, and his collaborator Seungjin Whang, the Jagdeep and Roshni Singh Professor of Operations, Information, and Technology, have developed a quantitative model that suggests greater emphasis on security early in the supply chain—at the factories and distribution centers where goods are placed into containers and at the ports where those containers are loaded onto ships—may be a cost-effective, efficacious way to improve homeland security.

“I have tried to demonstrate how economic value can be created throughout the supply chain from better monitoring to ensure security,” Lee says.

Lee’s model, originally published in the International Journal of Production Economics, argued shippers could save money and increase the security of their supply chain by designing and applying processes to prevent terrorists, not to mention smugglers and thieves, from tampering with containers before and during transit. Lee said tampering could be reduced by using “smart containers” equipped with sensors capable of detecting changes within the container, such as temperature or air pressure fluctuations, and fitted with radio frequency identification (RFID) chips containing standardized shipping information that could be scanned automatically at the port of arrival.

To produce its conclusions, the model compared the cost to an unidentified major Silicon Valley printer manufacturer using smart containers and improved supply chain management to the cost of the government doubling the number of active inspections. With Seattle as the port of entry and the goods originating in Malaysia, the model showed a savings of approximately $400 to $1,000 per container to the shipper. Lee says the company ships an average of 5,000 containers per year, so its aggregate savings could be as much as $5 million. The amount of savings, he says, varies based on the value of the goods.

“The success of this model is that it quantitatively demonstrated this kind of technology works,” Lee says. But the key to success in the real world is industry-wide adoption. “If more companies would participate, a standard would emerge, and everybody would be better off.”

One encouraging sign, Lee says, is the recent decision by Hutchison Whampoa Ltd., one of the world’s largest port operators, to start implementing his recommendations. The firm operates the Port of Hong Kong, among others.

Lee does not view his model as a replacement for Wein’s proposed screening system; rather, he believes the two systems should be implemented in tandem. “We must use multiple methods to help us screen better,” he says. “It is like a doctor using several tests to screen someone for ailments.”

DISRUPTING TERRORIST ORGANIZATIONS

Another method of preventing a terrorist attack is to disrupt organized cells before they are able to act. In their book Culture and Demography in Organizations, Glenn Carroll, PhD ’82, and co-author J. Richard Harrison, who met while they were both doctoral students at Stanford, describe nearly 15 years of research studying the maintenance and transmission of culture in organizations. At the core of their research is a mathematical model formally describing how cultural transmission works. It yielded some surprising conclusions about how to fight terrorist organizations.

“One thing that analyses have noted for decades, if not centuries, is that the cultural character of organizations tends to be fairly inertial,” says Carroll, the Laurence W. Lane Professor of Organizational Behavior and Change. “If you leave and return to an organization in 20 years, it’s not going to be a radically different place. How is it that you can have a fairly persistent culture in the midst of rapid change? Our model was designed to address that problem.”

Carroll and Harrison first used their model in 2002 to analyze how terrorist cells maintain and transmit their “culture.” Harrison, an associate professor at the University of Texas at Dallas, was invited by a contractor for the Department of Defense to adapt and present their model of cultural transmission at a series of meetings in the Washington, D.C., area.

“The emphasis was on formal mathematical models examined using computational methods, which tended to address the network nature of terrorist organizations,” Harrison wrote in an e-mail. The model provides a method for understanding how terrorist organizations recruit new members, train those members, and then keep them committed to the organization over time.

“If you think about the indoctrination part of the terrorist organization, members were indoctrinated and then went out living normal lives and yet they remained committed to the organization for years,” Carroll says.

As part of their study, Carroll and Harrison tried to deduce the most effective ways to disrupt terrorist organizations. Their findings appear to contradict conventional wisdom: Removing an entire cell has little effect on the overall organization. They also found that removing a terror cell’s leadership is unlikely to disrupt the cell’s functioning.

“Several government officials at the meetings believed that if we could just take out Osama [bin Laden], al-Qaeda would go away,” Harrison wrote. “The model that Glenn and I used, however, showed no significant effect of removing a leader, and other models supported our conclusion.”

Thus, if Osama bin Laden were captured, Carroll and Harrison’s model suggests another leader might simply step in to replace him, just as successors emerge in corporate settings.

Instead, Carroll and Harrison suggest terrorist organizations are most vulnerable to the incremental removal of cell members over a prolonged period of time. In concert with removal, the model suggests delaying the admission of new members also may be an effective way to disrupt the organization.

Despite these strategies, the model suggests terrorist networks will generally retain significant capabilities to conduct their operations. “One thing we find is that these terrorist networks are quite robust,” Carroll says. “They are not easy to destroy.”

Thus, according to Harrison, the most effective tool in fighting terrorism may actually be to eliminate the motivation for terrorism itself.

“The general consensus is that the only way you can really be effective in countering terrorist activity is to remove the root cause of the organization to start with,” he said. “If you do anything else, you are just going to be chipping away at the edges.”

While these researchers do not always see their models implemented, their work often influences public policy: Wein and collaborator Edward Kaplan of Yale University proposed that the government use U.S. postal workers to rapidly distribute antibiotics in the event of an anthrax attack. Their recommendation has been implemented in Washington, D.C., and Seattle is testing a similar program. And Wein’s congressional testimony on the fingerprint identification system used at the U.S.–Mexico border has led to a switch from a 2-finger to a 10-finger testing system, greatly improving the probability of identifying a terrorist. These are reminders that equations, too, can save lives.

John B. Stafford is a writer living in the San Francisco Bay Area.

Back to Top