In one of my earliest roles as a QA professional, I was entrusted with overseeing the quality of consumer products from over 20 injection-molding machines and assembly lines. Next to each machine was a scrap barrel.
As I walked down the center isle of the factory, I became conditioned (subconsciously) to expect certain levels of scrap to be present in the 20+ barrels at certain times throughout each work shift. The amounts of scrap rarely varied much from what I expected. It was highly subjective. No obvious measurements were being made. I wasn’t keeping a log.
Because I was able to recognize these patterns, and associate them with underlying production quality factors, it was not unusual for me to walk directly over to a machine or assembly line, for which less than the expected amount of scrap had been generated and identify at least one commonly recurring type of product defect being overlooked by the operators. For a process with greater than the expected amount of scrap, I would immediately notice that an unauthorized machine setting change had been made or a piece of equipment needed repair.
My coworkers accused me of having ESP. I just called it ‘The Quality Touch’. It took me a while to realize I was just subconsciously processing patterns I had come to recognize.
Today, I believe most organizations can be less concerned with reducing scrap rates. Better product/process designs, such as those employing automation, machine learning, and artificial intelligence (which also use pattern recognition)1, have empowered manufacturers to drive quality-related scrap to levels below what traditional lot acceptance sampling plans can detect.
There are always new ways and reasons to apply pattern recognition to quality improvement. Better ensuring patient outcomes in health care facilities and improving accuracy for medical diagnoses are two such frontiers. Management of our global environment and oversight of public health crises, such as a viral pandemic, are also emerging uses for pattern recognition.2
Pattern recognition superpowers may be subconsciously conditioned for humans, and subjective, but it should come as no surprise that this soft skill remains one of the most valuable contributions a quality professional can provide today. This is essential to mitigate the risk of what is referred to as “The paradox of automation” 1, whereby automated systems will also produce undesired results, faster than humans possibly could, indefinitely, until a human recognizes the problem and intervenes.
Want more information about applying pattern recognition for quality improvement to your products or processes? EMMA International can help! Contact us via our website, by phone at 248-987-4497, or by sending an email to: firstname.lastname@example.org.
1 “Automation”, “Machine Learning”, and “Artificial Intelligence”, last updated, respectively, September 04, 14 and 15, 2022, by Wikipedia® contributors; text is available under the Creative Commons Attribution-ShareAlike License 3.0; accessed 09/15/2022 via https://en.wikipedia.org/wiki/Automation, https://en.wikipedia.org/wiki/Machine_learning, and https://en.wikipedia.org/wiki/Artificial_intelligence
2 “Pattern Recognition & Prediction Made Simple”, February 18, 2022, by Steve Holcomb; ©2022 LinkedIn Corporation; accessed 09/15/2022 from: https://www.linkedin.com/pulse/pattern-recognition-prediction-made-simple-steve-holcomb-1e/