July 28, 2016

When quality came to Japan


Sarasohn (top) and Deming.
Edwards Deming is revered as the father of Japan's quality revolution. The revolution began in August 1950 when Deming, then working on the Japanese census, delivered a speech on "Statistical Product Quality Administration."

While Deming would long be a prophet without honor in his own land, the Japanese took his advice to heart, applying it to their assembly lines and rewarding those who met its exacting standards with the "Deming Prize."

Less well known is that Deming was building on the substantial work already done by Homer Sarasohn, who'd been recruited by General MacArthur to rebuild Japan's electronics industry following the war.

When his stay in Japan came to a close, Sarasohn, in turn, recruited Deming.

Robert Cringely endeavors to correct the record in this compelling essay from his PBS column back in 2000: "How Homer Sarasohn Brought Industrial Quality to Japan and Why It Took Japan So Long to Learn."

(And note Sarasohn's quip about Donald Trump sixteen years ago).

Sarasohn's recollections of what he discovered upon inspecting the state of Japanese manufacturing in 1946 certainly come across as wildly incongruous now.

With the exception of the Zero fighter and some aircraft engines, their designs were bad and their manufactured goods were shoddy. Having come from the Rad Lab, I was particularly appalled to see the primitive nature of Japanese naval radar. Their vacuum tubes were bad and the radios were even worse, since each was hand-wired by untrained, often unsupervised, workers. They produced goods in mass quantities, ignoring quality.

Despite the Zero's reputation, Japan's war machine produced nothing like the deadly and reliable F6F Hellcat. Grumman designed the fighter to be simple to build and maintain, and manufactured 12,200 Hellcats in two years, continually improving the frame and powerplant.

As a result, the Hellcat racked up a 13:1 kill ratio over the most widely produced Model 52 Zero. The Model 64 Zero might have begun to match the much improved flight characteristics of the Hellcat, but never made it past the prototype stage.

And by then, the successor to the Hellcat, the Bearcat (which also didn't see action in WWII), had leapt far past the Hellcat and the Model 64, setting performance records that would be eclipsed only by jet fighters.

Essentially, Mitsubishi made Zeros the same way an artisan makes a fine watch. As Hayao Miyazaki observes, "Structurally, the Zero was not designed for mass production." Each Zero was a one-off. It was amazing that Mitsubishi managed to build 10,000 of them.

Meanwhile, the U.S. would deploy four air-superiority fighters into the Pacific Theater: the F6F Hellcat, the P-38 Lightning, the F4U Corsair, and by the end of the war, the P-51 Mustang.

Mass production in Japan before the war emphasized the "mass" part of production, betting on the numerical odds to produce a usable number of quality components. The result was vacuum tube yields of 10 percent. Sylvania, by comparison, had pushed yields to 85 percent.

Jonathan Parshall and Anthony Tully point out in Shattered Sword: The Untold Story of the Battle of Midway that Zero pilots had so little faith in their radios that they often removed them to save weight.

The aircraft radios carried on the Zero fighter were of inferior quality and of limited range and power and were difficult to use. As a result, while all carrier Zeros had radios, pilots rarely relied on them.

One of Homer Sarasohn's students was Akio Morita, cofounder of Sony Corporation, whose breakthrough product was the transistor radio.

At first, discrete transistors were treated the same as vacuum tubes. The real breakthrough in quality came with the planar process developed by Fairchild Semiconductor, that employed photolitholography to "print" solid state devices onto silicon wafers.

Unlike a discrete transistor, that could be tossed if a single unit didn't meet the right specs, a flaw in a silicon wafer ruined the whole batch. Producing literally perfect wafers became an economic necessity. And that, Sarasohn argues, is what lit the fire.

The problem is, there's nothing proprietary about quality. It took a while, but Detroit caught on, and the Koreans did too, taking over the DRAM business by 1991. And two decades later had grabbed the bulk of the consumer electronics business from Sony and Panasonic.

The job Japan has ahead of it is not only to iterate and improve but to truly create, to somehow (frankly, it might be impossible at this late date) rekindle the white-hot passion for innovation that propelled Japan, Inc. to greatness in those golden postwar years.

Related links

How Homer Sarasohn Brought Industrial Quality to Japan
Twilight of the Zero
The rebirth of Japan's mass media

Labels: , , ,

Comments: