Highlights from a Keynote - Strategies for Successful DMAIC, Part VI

Highlights from a Keynote - Strategies for Successful DMAIC, Part VI

Posted by Scott Sterbenz on May 29, 2019 7:54 am

Hello, and thanks for tuning in for this segment of Highlights from a Keynote—Strategies for Successful DMAIC Problem Solving.

Today, I’ll be telling you a story that enforces the fifth concept of good DMAIC problem solving—that it’s a good idea to focus on the physics and logical thinking, while leveraging the math.  Unfortunately, a lot of DMAIC training focuses so heavily on the math, the instructors forget to teach the students how to think like a problem solver. In all the years I have been practicing DMAIC, never has it been the math that has solved the problem. Instead, the math has helped me prove what physics or logical thinking has indicated what the problem is.

To illustrate this, many years ago I was working on an issue with our SecuriCode keyless entry system. For this problem, Ford decided to move this extremely popular feature from a physical keypad on the door sheet metal to a touch keypad integrated into the B-pillar trim. During development, we were having a problem with the integrated circuit board cracking, causing one or more of the numbers not to respond when touched.  When something is breaking, there is a fundamental stress-strain relationship; that is, the stress on the part is greater than the strength of the part.  For the SecuriCode system, there were several parts sandwiched together, with the integrated circuit board in the middle. Clearly, the only way for the circuit board to break was if the shapes of the sandwiched parts didn’t match, causing the weakest part to conform and potentially crack.

How would I be able to prove this? It turns out that when I scanned the parts, the curvature followed a perfect quadratic polynomial. I remembered from my 8th grade math class that the a-coefficient characterizes the sharpness of the curve. All of the sandwiched parts had the same a-coefficient except for one. Through an improvement action, we were able to match that a-coefficient to the others, and the issue disappeared. It wasn’t the math that solved this problem; it was understanding the existence of a stress-strain relationship and using the math to prove it.

The next installment of this series will be the last one, so stop back in a few weeks and find out what is the sixth and final characteristic of successful DMAIC problem solving.
Best, Scott C. Sterbenz, P.E. ASQ Six Sigma Forum