CISAC scholar computes counter-terror problems

CISAC faculty member Lawrence M. Wein works at solving mathematical problems posed by potential terrorist attacks.

It's been said that World War I was the chemists' war and World War II the physicists' war, but that World War III is destined to be the mathematicians' war.

With the gravest threat to U.S. security now posed by rogue terrorists who simultaneously hold a grudge and have access to weapons of mass destruction, some of this country's prime human calculators are on the case.

Combatting terrorism not with mortars and missiles but with mathematical models, they intend to prove this theorem: There literally is safety in numbers.

In their vanguard is an amiable Stanford University professor who, by devoting himself to the application of math principles to doomsday scenarios, is beginning to acquire the nickname "Dr. Doom." The attacks of Sept. 11 inspired mathematician Lawrence Wein to channel his expertise into some of the most compelling questions of our time.

He has ciphered the risks of our "woefully inadequate" inspection of container ships, assessed the effectiveness of border-control fingerprint checks to spot terrorists, and performed what may well be the first math analyses of hypothetical botulism, anthrax and smallpox contaminations.

Recall the agitation last year over warnings that the nation's milk supply was vulnerable, based on the calculation that one terrorist with a few grams of botulinum could contaminate a tanker and potentially poison 100,000 gallons of milk? That's just one of many tidings of comfort and joy brought to you by Wein.

Another such upper is that just about 6 percent of containers shipped into U.S. ports will be categorized as suspicious and subjected to tests for a nuclear device, based on a system that relies largely on reporting by the shipper. The rest, Wein noted, "just waltz right into the country without an inspector laying an eye on them."

At Stanford's Graduate School of Business, he teaches a core course in operations, and he says the parallels are strikingly similar: Just as McDonald's needs a well-designed distribution system to get its hamburgers out quickly, so the government needs a well-designed distribution system to get vaccines and antibiotics to citizens who might sicken or die from a bioterror attack. In math lingo, some of these computations rest on "queueing theory" -- the notion that a lot of needy people must line up behind a limited number of distributors.

And what do we do after crunching millions of numbers to arrive at specific prescriptions? Wein has taken on the role of necessary nag to a torpid federal bureaucracy and recalcitrant industries.

In fact, the feds scrambled to try to suppress publication of his damning milk study as a threat to national security -- an argument the journal Proceedings of the National Academy of Sciences rejected before publishing the work.

These days, Wein regularly testifies at congressional hearings, addresses scientific forums and pens op-ed pieces in national newspapers about precisely what the government should be doing -- and where it is falling short.

"I believe my work demonstrates that numbers really matter, and we need to pay attention to what they tell us," said Wein, a critic of the federal government's homeland security failures in key areas, particularly at the ports. "I really don't bring my own personal political views into this -- I would be just as critical of the federal bureaucracy if a Democrat were in the White House.

"Bureaucracy just isn't designed to respond nimbly."

On the other hand, politicians are eager to do something, anything, that might thwart a terror attack or save lives. But in an information vacuum, that makes them easy prey for entrepreneurs hawking all kinds of safety gadgets.

So, much of Wein's math analysis goes beyond documenting threats to assess the most effective, cost-efficient remedies.

Take the milk scare, for example. Wein urged the government to mandate that milk tanks and trucks be locked, that two people be present when it is transferred along each part of the supply chain, and that milk truck drivers use a 15 minute test to detect any toxins. The estimated cost of all this: 2 cents per gallon.

Some other experts disputed Wein's assumptions, and dairy industry groups insisted they already had taken extra steps such as raising pasteurization temperatures to secure the milk supply. They dismissed the scenarios Wein described as "highly unlikely or impossible."

To avoid being victimized by a radioactive device aboard a tanker -- otherwise known as a "poor man's missile" -- he and colleague Stephen Flynn of the Council on Foreign Relations advocate using techniques such as gamma-ray imaging to screen 100 percent of containers. They calculate the cost at $7 per container.

Some of Wein's calculations have translated into real reform.

In 2004, he presented to the White House his findings that the system of collecting only two fingerprints from incoming foreign visitors ran a high risk of missing terrorists because so many prints turned out to be of poor quality. His solution was simple: Take prints from all 10 digits. The government adopted it.

His 2003 research into the most effective response to an anthrax attack prompted pilot programs that are now testing his finding that the best way to widely distribute antibiotics would be by mail carriers, who already go door to door.

Albert Einstein once observed: "As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality."

And sure enough, one of the impediments to fighting terrorists with mathematics is the paucity of hard data.

An anthrax scenario illustrates the point. Estimates of fatalities per kilogram of poisonous agent released vary from a low of zero to a high of 660,000. That's because of variables: the dose required to cause infection, the percentage of people who survive infection, the degree of aerosol dispersion, the density of the population, the environmental stability of the agent and the effectiveness of a public health system response.

As Wein acknowledges, "It's not like we can do massive clinical trials on this."

The trail of what-ifs is so convoluted that in 2002 the National Academy of Sciences cited "an irreducible uncertainly of several orders of magnitude in the number of people who will be infected in an open-air release."

This makes some experts suspicious of anybody's calculations, even from a researcher with a pedigree like Wein's.

"With so much uncertainly surrounding the outcome of a bioweapons attack, it does not make sense to plan extensive biodefense programs when more-certain public threats, particularly those involving nuclear weapons, require attention," argued Allison Macfarlane, research associate in the Science, Technology and Global Security Working Group at MIT.

But Wein is undeterred, and is busy on new assessments. Yet he insists his work has not impaired his ability to sleep soundly.

"People really are much more likely to suffer from cancer than a terrorist attack, but that doesn't merit the same attention," he said. "We should do what makes sense to protect ourselves from both."