COE/Tech School Advance Statistics Education to Benefit Industry
The total quality movement that has been sweeping industry for the past 20 years has increased the demand for manufacturing professionals--whether they be engineers, technicians, managers or machinists--who understand statistics. "There is a major push in industry for using statistical methods for quality control," says Søren Bisgaard-Frantzen, an associate professor of Industrial and Systems Engineering and director of the Center for Quality and Productivity Improvement. "If we could elevate the level of knowledge in industry of production workers and foremen, in terms of using statistics, the potential is just enormous."
Quality improvement programs rely on these statistical techniques for close tracking of variations in manufacturing processes. But often the people who most need to understand variability have little training in statistical methods. That's because statistics are not taught soon enough, if at all, and when they are taught, are not grounded in practical applications, Bisgaard says. To find ways to improve the way statistics are taught, he is working with Robert Brown, an instructor at Madison Area Technical College, to incorporate statistical techniques into the course work for MATC's machine tool program.
The trick, Bisgaard says, is to focus on real-world examples. "I had this idea that we shouldn't even tell them that we were teaching statistics. We should just do it," he explains. So Bisgaard, Brown and graduate students Visanu Vivatanaprasert, Chen Liao and Richard Lu designed two educational modules that demonstrate fundamental statistical methods by tracking subtle size variations in manufactured items. The graduate students then introduced the material as part of a tool-making degree program at MATC.
"The real problem was to simulate a sort of a manufacturing process," says Bisgaard, who although now a statistician and industrial engineer, started his career as a tool maker in his native Denmark. In an actual manufacturing plant, samples would be taken periodically to check for consistency. But although MATC has a large and well-equipped machine shop, it would have taken too long to make and measure enough parts to establish the needed statistical trends. So instead the researchers bought boxes of screws from a local hardware store.
After all of the screws were mixed in a large container, the technical college students simulated taking samples from a manufacturing process by randomly plucking groups of five screws, which they placed in envelopes labeled to represent hour intervals. The next steps were to measure the screws using micrometers and shout out the results to the UW-Madison graduate students, who charted the range and the variability of differences in the screw sizes. "Then we showed them how to compute these limits, how to interpret the charts, which got all of the theory in by example," Bisgaard says.
This exercise, Brown says, shows his students how to recognize the impact of variations in the products they manufacture. "It helps them identify and troubleshoot specific problems in the shop," he says. "They'll say that they've been fighting this problem all along, but didn't realize it was a problem." Bisgaard's students also taught a second module in which MATC students calibrate measuring instruments to help distinguish between true variability in manufactured items and the appearance of variability that results from instruments that aren't calibrated to the same standard.
The collaboration started about three years ago when Brown approached Bisgaard about creating a wide-ranging program that would help to fine tune the way MATC teaches metrology--the science of measurement, create a summer metrology institute and lead to the construction of a multi-million dollar mobile metrology laboratory, which will bring a high-tech classroom to schools and universities that couldn't otherwise afford such a facility. "We wanted to--we had to--find a way to get math and science teachers to incorporate measurement into their teaching so that students can have an appreciation for measurement early in their education," Brown says. In addition to giving UW-Madison graduate students a chance to develop innovative teaching methods, the collaboration provides internships for MATC students at the university's Center for Plasma-Aided Manufacturing.
From July 7 through 12, Bisgaard, Brown and representatives of firms such as the Ford Motor Company, Brown and Sharpe and Giddings and Lewis will attend a metrology institute at MATC during which a metrology curriculum will be developed and the mobile lab will be planned. Then, Brown says, the work of Bisgaard's students will start to pay even greater dividends. "When that mobile lab becomes a reality, when those trucks are really parked out there, then these modules are going to be part of that road show."