Improvement Insights Blog
Deming on Spec Limits vs. Variation
Deming contrasted US and Japanese focus: spec limits vs variation. Here’s the essence:
“Hi, I’m Jay Arthur, author of “Lean Six Sigma Demystified” and QI Macros [software].
“I don’t know about you, but every once in a while I go back and read through books that I’ve gotten in Quality over the years. I went back into Deming’s “Out of the Crisis.” It was very interesting. There’s a thought he put in here and I’ll read it to you: “We in America have worried about specifications: meet the specification. In contrast, the Japanese are worried about uniformity, working for less and less variation around the nominal value…”
“What does he mean? Well, if you think about the whole idea of specification limits, the whole idea is if your production fits inside of there, guess what? You’ve met the specifications.
“But what do the Japanese think about? They think about “How do I minimize the variation around some target value?” Right? So we have a target value, but maybe it’s off just a little bit but it’s still better… off just a little bit.
“I remember a story they told us way back when I was getting trained: Ford was manufacturing transmissions in Japan and America. The Japanese transmissions had far less warranty problems than the American transmissions, so they decided to have some fun. They got five transmissions from each, pulled them apart, put the gauges on top of them and they were freaked out because there was no variation in the Japanese transmissions. They were all spot on. The American transmissions had some slop in it, right? This gets back to Taguchi’s whole idea: the Taguchi loss function. If this is your target value, as you move away from that value, there’s an incurred cost, right? There’s an incurred cost.
“So that’s what I remembered from re-reading Deming. Maybe it’s time to dig one of those old books out and look back through it.
“I’m Jay Arthur; that’s my Improvement Insight for this week. Let’s go out and improve something.”