American Made Advocacy: Five Years of Educating, Advocating, and Influencing Legislation and Policy
Defense Speak Interpreted: Understanding What the Department of Defense Is, and Isn’t
The Right Approach: The End of an Era—DoD Proposes MIL-PRF-31032 Cancellation
In the early 1990s, I was working for EMD Associates, an EMS company (now a Benchmark Electronics operation in Winona, Minnesota), when I became familiar with Dr. W. Edwards Deming. EMD was an early entrant into the SMT provider world, and we were very focused on quality. Deming was one of several experts that we used as a guide for our company.
As part of that journey, I was able to attend a live Deming seminar in Phoenix. There were several speakers as part of his team, but Deming, who was at least 90 years old at the time, spoke in very memorable fashion each day on key topics.
Deming’s demonstration of the Red Bead Experiment left the most profound impression on me. You can see a shortened video from the Deming Institute of his experiment online1. (In fact, I may even be in that audience.) This experiment involved several “willing workers” (audience volunteers) who were asked to use a dimpled paddle to randomly remove 50 beads from a box of hundreds of white (good) and red (defective) beads. Those “workers” who had fewer relative defects were rewarded and promoted. Those who had more defects were put on probation and dismissed. The company ultimately failed, allegedly because of the workers’ inability to produce enough good product. It was, of course, all random and had nothing to do with the skill of individual workers. The system was simply not capable of producing the output they required to succeed and the variation across workers was just a normal variation within that system. Deming’s colorful commentary was both humorous and sobering, but the impact is etched in my memory. Since then, I have always tried to understand the impact of normal variation on what I think we see.
Now, I recall a more humorous anecdote from that seminar when he reminded the audience that he clearly expected them to pay attention and listen. He had an audience of maybe 300 to 400 people, and at the end of the first session, they had an opportunity to ask him questions. There were microphones on stands in multiple aisles where long lines of people quickly formed behind each one.
To read this entire article, which appeared in the July 2023 issue of PCB007 Magazine, click here.
