Feds may unleash Six Sigma on terrorism - Management tool
could save lives
By Del Jones
USA TODAY
At a time when fighting the war on terrorism has become arguably the most
important issue facing the USA, authorities are looking into an unlikely weapon
to aid their fight: Six Sigma.
Six Sigma is nothing like a laser-guided smart bomb but rather a
statistics-heavy regimen of analyzing problems that has saved corporations
billions.
Can an arcane management process save lives by helping prevent terrorist
attacks? Mikel Harry, often called the father of Six Sigma, says it can, in a
major way. He estimates the USA would be safer from terrorist attacks by a
factor of hundreds or thousands.
Companies like to brag about improving efficiency, but as a matter of policy,
the secretive Central Intelligence Agency won't say whether it is investigating
Six Sigma as a way to strengthen its terrorism-fighting arsenal. However, Six
Sigma experts say they have been called in to consult with various homeland
security agencies.
Whether corporate success can be repeated in a federal bureaucracy is an open
question. Six Sigma is ''powerful stuff'' that could work even in the sprawl of
the U.S. government, Dell Computer CEO Michael Dell says.
Finding more uses
First used by Harry and the late Bill Smith at Motorola in the mid-1980s,
Six Sigma symbolizes 3.4 mistakes per million opportunities. The process was
originally used to eliminate assembly-line defects but has expanded into almost
every corporate operation. It uses a complicated approach to problem solving
called DMAIC (define, measure, analyze, improve and control).
Experts say it could be used in thousands of homeland security projects.
Consider the mountain of information that floods into the CIA, such as
intercepted phone calls, applications to pilot schools, etc. Suppose an e-mail
is intercepted that includes a disguised threat on the Golden Gate Bridge. A
quick decision must be made to discard the e-mail or take it seriously.
Discarding bad information is crucial because useless data can paralyze decision
makers further up the line.
There may be 50 points where such pass-fail decisions must be made about the
usefulness of a piece of information. In Six Sigma talk, these points are called
''decision nodes.'' If each of those 50 nodes passes judgment on 60 pieces of
information each day, there are 300 opportunities for a decision error each day
as intelligence moves up the chain to Security Chief Tom Ridge and President
Bush.
If decision nodes average 99.38% accuracy, they are at Four Sigma, which is
about the accuracy of services such as prescription writing by doctors and
airline baggage handling.
If improved to Six Sigma, accuracy is 99.99966%. That means only one of about
every 294,000 pieces of vital information would be erroneously discarded.
At Six Sigma, there is a 99.9% chance that all 300 decisions are accurate on
a given day. There is a 97% chance all decisions in a month will be right. Where
there is only a 15% chance that all decisions are right on a given day at Four
Sigma, there is a 15% chance that all decisions will be right over a five-year
period at Six Sigma.
Such efficiency would be invaluable when lives are at risk. ''A quantum
difference indeed,'' Harry says. That's how attaining Six Sigma in the war on
terrorism could make the USA 1,800 times safer, Harry estimates.
Skepticism remains
Not everyone is so optimistic. Even fans like Michael Dell warn it could
take years for U.S. intelligence agencies to fully implement Six Sigma but adds,
''It's possible.''
Six Sigma is lousy at fixing rare and random problems, says Elizabeth Keim,
president of the American Society for Quality and a Six Sigma consultant. And
terrorism, at first glance, seems to be the very definition of rare and random.
But Keim and others schooled in Six Sigma see events like the Sept. 11
attacks as the catastrophic result of a breakdown in the millions of frequent
and mundane preventive steps that must be taken with vastly greater efficiency.
These steps are drudgery: identifying the millions of people entering and
leaving the USA, mapping the entrances to water plants or detecting E. coli in
farm products.
''The particulars of Six Sigma are not exciting'' when used in business or in
war, says Dan Burnham, CEO of Raytheon, a defense electronics company steeped in
Six Sigma and which makes such anti-terrorism technologies as facial
recognition. It's the results that are striking, he says.
Deploying Six Sigma against terrorism would be little different than when it
was used to determine that most steps in a Japanese patent system's application
process were wasteful. The cost of each filing was slashed to $1,200 from
$48,000. Communication satellites are rented out by the second and are not
always used efficiently. General Electric used Six Sigma to make sure its
satellites were being used 97% from 63%, adding $1.3 million a year in revenue.
Former CEO Jack Welch, who drove Six Sigma deep into GE's culture before his
retirement, counts himself among the cautious optimists that Six Sigma could
work against terrorism.
Fighting terrorism isn't really much different than marketing, Harry says.
Marketing executives, like intelligence experts, must digest mountains of mostly
useless data, analyze the fraction that is important and persuade decision
makers to get the right product on the shelf just as consumer tastes are
changing. At its best, marketing influences consumer tastes, which like
terrorists, are a moving target.
Some abandon it
Even after hearing about the billions saved at companies like GE, others
have abandoned the Six Sigma effort in frustration. Fewer than 15% of the Fortune
1,000 are using it in a significant way.
Burnham says it won't work without an ''obsessive, compulsive'' CEO. Welch
says he was a ''raving lunatic'' about it. Some GE employees privately say that
the most common-sense solutions can no longer be made easily but must pass a
rigor of charts and statistics.
Companies that don't stick it out for at least five years soon revert and
lose all progress. That doesn't bode well for the federal government, where
attention spans often don't survive election cycles. ''The federal government
makes GE look like an ant on an elephant's back,'' Harry says, and there will be
a ''a gantlet of pain'' to get it implemented.
Former House speaker Newt Gingrich, who speaks at Six Sigma conferences, says
he would be surprised if the CIA is seriously considering Six Sigma but says it
eventually will have to because of the nation's inefficient counterterrorism.
There is no better chance than during a time of emergency, experts say. And
the government isn't starting from scratch. Military contractors are among the
early implementers of Six Sigma, and generals and admirals are among the
converts. Air Force Secretary James Roche came from Northrop Grumman, a defense
contractor steeped in Six Sigma.
At least one local government is using it. A Six Sigma project helped get 98%
of the potholes in Fort Wayne, Ind., filled within 24 hours. It also will help
the city figure out how to get twice as many contaminated people through the
showers of a moving van converted for emergency use, Mayor Graham Richard says.
A Six Sigma onslaught against terrorism would produce many byproducts for the
government, including cost savings, Keim says. She predicts that it will improve
the fight against all crime. Statistics will lead to profiling, but it will be
profiling based on science, not prejudices, she says.
It may be coincidence, but CIA Director George Tenet is sounding like a
convert. In his Oct. 17 testimony to the Senate Select Committee on Intelligence
he used some Six Sigma vocabulary in calling the intelligence breakdown relating
to Sept. 11 an ''error'' that exposed weaknesses in the process. He used the
word ''customer'' for the recipients of intelligence. Six Sigma at corporations
is built around customer needs.
''In the counter terrorism business, there is no such thing as 100%,'' Tenet
said.
Neither is perfection achievable in business. But in the quest for the near
perfection of Six Sigma, companies say they have achieved improvements that did
not seem possible. ''You can't tweak the existing system and get there,'' Harry
says.
|
