Silver Circle Profiles
A Look at Doug Samuelson
posted: August 2006How often do you have the chance to earn $2 billion for your employer?
Doug Samuelson had that opportunity while working for the US Department of Energy in 1980. The government suspected that a major oil company had broken an energy pricing regulation in the mid-1970s. At the time, the government had price controls on oil pumped from existing fields, but not from new ones.
Using SAS and the Proc PLOT text-based plotting function, Samuelson and a team at the Department of Energy proved when the company had engineered a large field, pushing all of the oil from a series of existing wells near the edge of the oil field toward another series of wells near the middle. It was the same oil and the same field – just a shift in where it was pumped out. Under federal regulation, when a field was "unitized" in this way, it had to be reported as a unit as soon as "a significant change in producing patterns" took place; this is because the way it was reported greatly affected whether the oil was subject to a price ceiling. To prove when this had happened, the government team needed to process and summarize month-by-month production reports for more than a thousand properties that were part of the unit.
SAS® makes quick work of analyzing data
Samuelson was amazed at how quickly SAS helped him analyze the data – a few days to run the plots, as he recalled. Remember, these were the days of bulky mainframes and programs that took all night to run, and other people's attempts to analyze the data had consumed weeks without success. The Department of Energy team plotted the results onto an aerial map of the oil field, and it became a key piece of evidence. "We had a ring of red dots near the periphery of the field. About a third of the way from the red ring to the center was a nearly concentric ring of yellow dots," Samuelson recounted. The trial court imposed a $2 billion judgment against the oil company, and the US Supreme Court upheld it.
Samuelson believes the opportunity to plot the results visually had a dramatic impact on the case. "We needed a simple picture that anyone could understand," said Samuelson. "We created the most valuable picture in the history of the world. With all due respect to da Vinci, Rembrandt and van Gogh, what other picture, anywhere, anytime, ever fetched $2 billion?"
Samuelson's use of SAS® covers multiple arenas
The oil case was one of the first times Samuelson used SAS, and he has rarely been without the software since. "It boosted my career. I knew I needed to keep using it," he said. Since then, the operations research analyst has used SAS products in jobs with numerous government agencies, private-sector companies and his own firm. Operations research, his specialty, involves the analysis of outcomes and determinants of outcomes in complex systems using predictive modeling, statistical pattern recognition and forecasting techniques.
An adjunct faculty member at George Washington University and the University of Pennsylvania, Samuelson is a longtime contributor to OR/MS Today. "He is the academic equivalent of a syndicated newspaper columnist. He's got quite a following," said John Sneed, President of TESLA Inc., an electric systems load forecasting provider, who has been Samuelson's colleague for many years. Samuelson is also an inventor and entrepreneur. In 1989, he patented predictive dialing for outbound telephone call centers, for which he was a finalist for the 1998 Wagner Prize for OR practice.
SAS helped Samuelson and his employers root out Medicare fraud and look for possible patterns of fraud and error in vote counting in Ohio following the 2004 election. (Note to conspiracy theorists: He and his colleagues didn't find evidence of enough wrongdoing to change the outcome of the election.) He's looked at the health effects of hazardous materials. He's examined the costs to industry of tighter regulations – a project he undertook for the Federal Aviation Administration when the government wanted to know how controlling noise around airports would affect the airlines.
He is currently working for the Homeland Security Institute, a privately owned, federally funded research and development center in Arlington, Virginia, where he uses JMP®
One of Samuelson's recent favorites was a Medicare project in 2000. It takes a few months for the Social Security Administration to process a death certificate. One of Samuelson's colleagues noticed that hospitals make discharge data, including condition at discharge, readily available well before other government records are updated. By matching the data from Medicare's databases – and analyzing it using SAS and Dataprobe – the government could quickly determine which durable medical goods providers were billing for services well beyond the beneficiary's date of death. As a result, the government recovered more than $50 million. The power and ease of use offered by SAS have been particularly fruitful for Samuelson's consulting work. "Most operations research is done within the confines of large companies or large consulting firms. To succeed as a one-person shop is quite an achievement," notes colleague Sneed. "He's done innovative work."