LEARNING FROM FAILURE

It’s graduation time at SMU, and Dr. Paul Packman, chairman of the university’s civil and mechanical engineering department, feels better than he did last year. This year, for the first time, graduating seniors were offered a chance to learn about what Packman calls one of the most important aspects of engineering: failure.

Packman teaches a course in Failure Analysis, which involves studying wreckage – from airplanes, automobiles, trains, motorcycles, you name it – and figuring out what went wrong. Packman, who has taught at SMU for five years, is considered one of the leading experts in failure analysis; he says he is one of 100 qualified people in the field in the United States.

In November 1980, the MGM Grand Hotel in Las Vegas burned, and 84 people were killed. In July 1981, two walkways at the Hyatt Regency Hotel in Kansas City collapsed, killing more than 100 people. In January 1982, an Air Florida jetliner crashed into the Potomac River in Washington, D.C.; and most recently, a Pan Am jet crashed into a residential neighborhood near New Orleans International Airport. In each case, Packman was hired to find out what had failed. He decided to pass this information on to his students.

“What happens in failure analysis bothers me,” he says. “Nobody ever teaches it. We teach [students] how to build a bridge, but it’s all too academic. Where do you learn from failures? We spend three and a half years teaching these kids how to do things right; we should spend at least one semester teaching [them] how things can go wrong.”

The course was well-received by SMU seniors and graduate students as well as employees from Texas Instruments, E-Systems and Vought Corp. Packman says that the process of putting things back together is foreign to engineering students because they have been taught from the beginning to put things together logically. When students are suddenly faced with what Packman calls a “3-D jigsaw puzzle made of junk,” they become frustrated. Packman teaches them how to painstakingly find the problem, then the cause.