Lean by Doing

Early along, as a student of the Toyota Production System (TPS), now referred to as lean, I struggled with some of the concepts and systems. For example, Shigeo Shingo’s claim that a four-hour machine setup could be reduced to less than 10 minutes made me a skeptic.

“Perhaps, when Mr. Shingo talked about single-minute exchange of dies [SMED], he was referring only to machinery in which the changeover involved die changes,” I thought. “CNC lathes were different; no chance for single-minute changeover there.” I kept these doubts to myself although I believe the machine operators had similar doubts. We’d made modest setup improvements, but nothing like Shingo was suggesting. Anything beyond low-hanging fruit took us too far out of our comfort zone.

In 1996, a consultant from TSSC, a division of Toyota, began working with us. Through her questioning it quickly became apparent to me that our doubts arose from “conceptual blind spots” developed through old setup practices. These were preventing us from seeing a multitude of improvement opportunities. With some priming from our consultant, we found waste in areas where, as Shingo would say, “it was thought not to exist.” Under the guidance of our TSSC consultant, our well-ingrained existing frameworks for analyzing problems were gradually unlearned as part of our lean learning.

Our consultant, who had no special technical expertise with CNC equipment, nevertheless brought a conceptual framework to the table that turned our skepticism into science. She used a method I refer to as “directed discovery” constantly challenging us with questions about our beliefs and expectations. For example, we were asked, “Why are four bolts needed for the tool holders?" After we got over the because-we’ve-always-done-it-that-way answers, one of the operators had a brainstorm: “Can we replace two of the bolts with guide pins?” he asked. We experimented with the idea and, notwithstanding the concerns of the CNC lathe manufacturer, the guide pin idea worked!

Over a four-month period, focusing on 66 parts that ran on one CNC lathe, setups were reduced first from an hour to 30 minutes, then 20, and finally to 8 minutes for any changeover between any two parts. At that point the operators began measuring the setups in seconds.

An abstract understanding that we had gained from reading Shingo’s book was now replaced by tacit learning that could only have come through guided discovery and continual practice: learning by doing, or in this case, lean by doing.

There are two morals to this story. First, book study or classroom activities alone cannot create deep learning, and may, in fact, cause a person to privately harbor doubts. Only with constant hands-on practice can we remove the conceptual blind spots that obscure the full benefit of lean tools. Second, the perspective gained through the mentoring from our consultant was necessary to help us see beyond our conceptual blind spots. Experienced eyes can guide the inexperienced.

GBMP has chosen “Lean by Doing: Accelerating Continuous Improvement” as the theme for its 12th Annual Northeast L.E.A.N. Conference, which runs Oct. 4–5, 2016, in Worcester, Massachusetts. We have invited some of the most hands-on lean practitioners and experts to share their tacit learning with every participant. More than half of our sessions will be hands-on learning. Hope to see you there.

Bruce Hamilton will be our guest on Quality Digest Live this Friday, Aug. 5, 2016, at 11 a.m. Pacific/2 p.m. Eastern.

About The Author