Oil Spills, Peanut Butter, and System Safety

John Helferich, SDM ‘10

Former Mars SVP of R&D is now an SDM student

By David Rosenbaum
May 25, 2010

Who will be blamed for the Gulf oil spill? The U.S. government? BP? “They’ll blame the oil rig operators,” says John D. Helferich, System Design and Management (SDM) 2010, “This is unfortunate, but predictable, because the operators are all dead.”

“However,” Helferich continues, “all accidents are caused by system failures, usually due to production pressure that leads to safety rules being broken. That’s pretty much what happened with the 2009 Peanut Corp. of America recall. Somebody said we have a problem; someone else said, ‘Keep the line moving.'”

Helferich has been creating systems for a long time. In this, he is a bit different than most SDM students.

Actually, says Helferich, “I’m different than all of them.”

An MIT alum (S.B. in Chemical Engineering, 1979), Helferich worked at Mars (the sixth largest privately-held company in the U.S.) for over 20 years, the last 11 as SVP of R&D, responsible for over 350 employees. He championed Mars’ research into the anti-oxidant properties of chocolate, retired at 50 (“because I could”) and began teaching as an adjunct lecturer at Northeastern University’s MBA program.

“I did okay as a teacher,” says Helferich, 50. “I mean, I spent 20 years at the top. But I started to wonder if I should know more.”

Helferich began auditing ESD courses and was struck by the research of Professor Nancy Leveson, who began working on computer software safety analysis and moved on to aviation, aeronautics, and refineries, using the System-Theoretic Model of Accidents (STAMP).

“Her class was so cool,” says Helferich. “We’d step through a disaster per week, which was a great learning experience.”

Using the STAMP system, Helferich is investigating the vast 2009 peanut butter recall. “Food safety systems,” Helferich explains, “currently are based on a model developed to feed the Mercury astronauts. NASA went to Pillsbury and said, ‘We need a food that will neither crumble nor make them sick.’ Pillsbury realized it couldn’t test all the food, so it used the Hazard Analysis and Critical Control Points methodology used for missiles. That’s a good method, but it can’t take into account system pressures beyond the production line.”

Today, says Helferich, there are new pressures on the food supply chain–more imported food; the trend toward more locally grown produce may cause it to drift from a safe state to an unsafe one.

“If everyone ate nothing but canned food, we’d all be safe,” says Helferich. “But who wants that?” In this interconnected world where fruit picked in China on Monday shows up on Cambridge tables on Wednesday, new questions about safety must be addressed–and addressing them is why Helferich came to SDM after a long, successful career.

“An MBA would have been ridiculous for me,” Helferich says. “I have all that experience. I don’t need to get ahead. There are a lot of MBAs. There are only 50 SDMers. SDM returns me to my roots as an engineer. And being around people with all different kinds of experience is fun.

“For me, it’s about service to my industry,” he concludes. “I’m done working, but I’m not done advocating.”

John Helferich, SDM 10
Photo by L. Barry Hetherington