Datasets:

botchagalupe

/

ProfoundDeming

like 0

I imagine Deming chose his summer job for the same reason a college kid might intern for free in Silicon Valley: to have a front-row seat at the cutting edge of innovation, landing a role in research and development. While his Ivy League education may have prepared him to appreciate and absorb the management concepts floating around Hawthorne, it was his rural raising that prepared him for the on-the-ground reality and allowed him to empathize with the harsh conditions of the working-class people all around him. The fact that Hawthorne was the foremost industrial site in the nation didn’t change the horrendous working conditions endured by the thousands of people employed in the factories. It was still a hardscrabble life, just on an industrial scale.

By the time Deming stepped off the trolley line in 1925, Hawthorne Works

employed around forty-thousand people, mostly women. A friend of his had forewarned him to “stay well away from the stairway when the whistle blew at the end of the day. ‘Those women will trample you to death. There won’t even be an oil slick.’”

Later, Deming would reflect, “It was hot. It was dirty. No wonder they wanted to get out.”

Ed held a particularly low view of a mainstay of American factories at the time: piecework, where a worker got paid according to the number of units produced or tasks performed. Knowing Hawthorne operated this way, it’s likely Rose was paid according to how many telephone assemblages she inspected.

This type of pay scheme incentivized workers to focus on quantity, not quality. It is harder to take pride in workmanship if you knew that each person working on the unit before you got to it did a rush job. No wonder managers working under Taylorism were so suspicious and antagonistic toward the line workers. Deming would later observe, “Piecework is man’s lowest degradation.”

Although Hawthorne workers took pride in the bigger picture, they still operated under a system where shoddy workmanship was incentivized from the beginning. Ever the philomath, Ed was curious about everything around him. It was his good fortune to wind up at Hawthorne Works, where he could be exposed to the latest in production processes, the social experiment that was the town of Hawthorne, and the production and management experiments that were being conducted by Mayo and others. And I cannot imagine that he worked there for months without indirectly coming across the work of Dr. Walter Shewhart. Unbeknownst to Ed, his relationship with the AT&T physicist researcher would become a defining factor in his life, as we’ll soon see.

Ed’s time at Hawthorne Works exposed him to new ideas about manufacturing and labor. Workers’ autonomy outside of the factory led to a community not dictated by company patronage but one led by the community itself. The result was independent-minded individuals like Rose Cihlar. Without her experience at Hawthorne Works, I don’t know that she’d go on to work for an electrical manufacturer as a married woman with children in those early decades when women were still expected to adhere to their traditional domestic roles. Without her own income, her son wouldn’t have gone to Purdue nor started down the path to becoming the commander of the Apollo 17 mission.

Hawthorne Works’ progressive arrangement prepared Ed to appreciate how companies and their employees could work together for the common good. When he landed in Japan decades later, he, more than any of the other Americans there, immediately grasped how crucial the special arrangement between Japanese companies and their employees was and why it led to superior quality. This line of thinking would become what I believe is the striking difference between the System of Profound Knowledge and Western management: Ed discovered a human-centered approach to systems, in general, and business, in particular.

But before he could begin to fully articulate his System of Profound Knowledge, he had to learn from the master of variation . . . and to understand variation, we have to first appreciate the history of quality control.

Deming’s Journey to Profound Knowledge - How Deming Helped Win a War, Altered the Face of Industry, and Holds the Key to Our Future - Part 1 - Chapter 3: The Birth of Quality Control and Standardization

Violins and violas, in varying stages of progress, hang from the rafters, drying. On the workshop floor, apprentices toil away at their workbenches. An older man unscrews wood furlings from the cello he’s crafting. Behind him, a younger man intently files the scroll piece of a viola.

Master Stradivari comes into the workshop and begins to inspect their work. He strolls over to a young man carefully sanding the upper bout of a nearly finished violin. The young man lays down his sandpaper and steps back with his head down and his hands clasped. Stradivari carefully picks up the instrument to look it over.

“Very good,” he says. “Exquisitely worked. You’ve crafted a jewel, my boy.” He takes a step as he continues. “Perfect for a courtesan or a priest to pluck after supper or polish Sundays after mass. In other words, . . .”—his tenor and countenance change as his gaze moves from the instrument to the young man’s face, holding out the violin as if he were presenting it to the apprentice—“. . . this

violin will never bear my name.”

He spins around and—WHAM!—slams the violin against the bench, breaking it into splinters.

Stradivari shouts in the young man’s face, “Put your anger into your work, boy!” Then he angrily strides out of the workshop. As he does so, he shouts, “Stay with me and learn!”

This scene from the movie The Red Violin is, more or less, the story of tools, quality, and humankind for the last three million years. One person learned how to make a certain type of tool from “a master.” Then they made said tools one at a time by hand. A potter made one pot at a time. One blacksmith made one plow. One cooper, one barrel. One luthier, one violin.

The quality of anything humans made, by and large, depended on the skill of the craftsman who made it. Everything built, crafted, or made was unique. Craftsmen might get decently good at consistently churning out high-caliber products, but each one was still one of a kind. And by and large the quality of anything humankind made depended on the skill of the craftsman who made it.

The eternal question of quality has always been this: “How good is good enough?” Carving a walking stick to hike the Appalachian Trail? Quality isn’t much of an issue. Carving the wood for a Stradivarius? Nothing less than absolute perfection will do.

If you were a soldier on the battlefield, you prayed that your new sword wasn’t made by the village blacksmith when he was blackout drunk. If there were barbarians at the gate, a feudal lord might have a moment of panic remembering that he’d gone with the lowest bidder to build said gates.

History records a few times when we progressed from the craftsman model of production. About six hundred years before and a hundred and forty miles east of Stradivari’s workshop, the city-state of Venice created a massive assembly line called the Venetian Arsenal.

The shipyard there could assemble an entire seaworthy vessel from prefabricated pieces in as little as a day. (I doubt the shipwrights in Venice knew it, but they had rediscovered an idea from over a thousand years earlier and a thousand miles south, when Carthage used the ship assembly-line method during the First Punic War with the Romans.)

Around roughly the same time, the Chinese state of Qin mass-produced some pieces of a crossbow, which played a role in conquering their neighbors and establishing the first Chinese imperial dynasty.

But outside of a handful of examples like these, basically everything we made for thousands of years was one at a time. Stradivari’s workshop was the story of humanity’s progress and civilizations’ development. But the history of production and quality rounded a corner thanks to Thomas Jefferson passing around a pamphlet—the result of which would come to necessitate Walter Shewhart’s statistical process control and the theory of variation.

Standardization & the American System

In 1785, the United States had been a country for only nine years. Jefferson wouldn’t become president for another sixteen years. In the meantime, he was ambassador to France, where he met a gunsmith named Honoré Blanc. Blanc had unknowingly copied the Qin of China: weapons that used interchangeable parts.

With Blanc’s invention, if the flintlock of your musket broke, you didn’t need to return it to a gunsmith to handcraft another one. Instead, you could pick up a new flintlock from a pile of parts and be back in the fray before you could say, “Wait, tell me again why I’m dying for some schmuck I don’t know?”

Jefferson saw the importance of the invention but couldn’t convince Blanc to move from France to America, which was still being carved out of the wilds. Instead, Jefferson wrote to the first Secretary of War, General Henry Knox, explaining Blanc’s ingenious system and urging its adoption.

In 1798, some ten years later, the US government granted a contract to Eli Whitney to manufacture ten to fifteen thousand muskets. (This was some-

what ironic, seeing as how he’d never created a musket in his life.) About ten months into Whitney’s government contract, the secretary of the treasury sent him a “foreign pamphlet on arm manufacturing techniques”—almost certainly French and, therefore, almost certainly Honoré Blanc’s. By 1801, Whitney had not only missed the contractual deadline but the quantity as well. By a magnitude of a thousand. However, with the ten guns he had, he demonstrated to Congress that the parts from any one musket could be switched out with another. If the gun broke, the Army wouldn’t have to buy a whole new gun—just a replacement part. The legislators quickly mandated that all such equipment be standardized.

I imagine he failed to mention that it took more money to manufacture ten muskets with interchangeable parts than it did for a gunsmith to craft ten muskets. It did, however, buy him more time and earn him political support.

While Whitney didn’t invent interchangeable parts, he did a successful job of evangelizing the idea. More and more companies, and especially armories,

began implementing the idea during the 1800s. On top of this, US manufacturers began shifting from hand labor to relying more heavily on mechanization.

They went from skilled craftsmen using hand tools to semiskilled laborers operating machines. These two developments led to what came to be known as the American System. While the Industrial Revolution had already begun in Britain, the American System was a profound evolution in industrial development. By 1880, the US, Europe, and elsewhere had entered what historians term the Machine Age. While these same historians might point to interchangeable parts as a key development, I think they’re missing the point. Interchangeable parts were the result, but standardization was the catalyst.

To illustrate my point, consider the ubiquitous cargo ship container. It doesn’t matter whether you’re in New Orleans, New Plymouth, or Newfoundland, they look exactly the same. However, it didn’t used to be that way. As The Box: How the Shipping Container Made the World Smaller and the World Economy Bigger so insightfully reveals, before standardizing the size of containers, maritime trade vessels had to load cargo by hand.

Crates of fruit, trunks of clothes, sacks of potatoes, individual automobiles, raw lumber—you shipped it however you wanted to. It was up to the longshoremen on the piers to figure out how to most efficiently load it onto ships. And each ship was different. Your goods might arrive in the hold of a cruise liner or belowdecks of a barely seaworthy trawler.

Once the world settled on a standard cargo container size, everything could be planned for. Ships were built to hold the exact dimensions. Cranes could be computerized because they needed to work with only one type of container. Rates could be standardized because transporters knew the dimensions of their loads; cost became a simple matter of weight, distance, and priority.

Trucks and trains could be configured to all carry the same size box, meaning you could load a container of your products onto a railcar, see it lifted onto the bed of a tractor-trailer, set on a ship, unloaded, and transported to your customer . . . without ever opening its doors. Maritime trade went from relying on specialized skill sets to standardized processes—making everything far, far cheaper to transport and trade.

When I worked at the IT company Docker, The Box and its underlying principles were the founder’s bible. By standardizing the way we created data servers, we could format thousands of servers in the same amount of time it would take to conventionally format one.

Standardization: That’s what changed the world. That’s what spurred the Machine Age and everything that came after. Factories standardized their products and processes. Manufacturing quality had evolved from one craftsman’s skill to an era of standardization. Standardization improved production, but despite all the technological innovations and progress, production still hinged on the same problem Stradivari had: sometimes the product didn’t come out right. You’d have to scrap the whole thing and start over. Interchangeable parts were, in fact, the turning point in the history of quality control and led to the theory of variation.

But even with interchangeable parts, producers soon discovered that exact specifications were unrealistic. No matter how precise the machines and the processes, the outputs all slightly varied from each other. This spawned a need to allow for variance in product specifications. Think of the notion of an exact fit as a deterministic approach (as we learned about in Chapter 1). Specifications that allow for a certain variance—or tolerance—are more in line with a non-deterministic approach.

The people in charge of a manufacturing process had to decide the limits of what was acceptable. How much variation would they allow or tolerate in the finished products? They called these “tolerance limits.” Industrial producers switched from trying to achieve an exact fit to allowing products to be manufactured within certain tolerance limits. In the beginning, these limits were simply named “go/no-go.”

Go or No-Go?

Back at Hawthorne Works, this was essentially Rose Cihlar’s job. She was a quality inspector. It was her job to act like Stradivari. As telephone systems rolled off the assembly line, Rose carefully inspected each one. Unlike Stradivari, she had some tools to test the specification tolerance. If the telephone fell within those tolerance limits, it was a “go.” Otherwise, she marked it as a defect—a “no-go”—and tossed it in the reject bin. Out of the forty thousand or so workers at Hawthorne, five thousand of them were inspectors like Rose, inspecting and rejecting all day long. Over one hundred thousand individual parts and pieces composing the individual telephone were scrapped just like that. While the factory manufactured telephones, its second-biggest output was scrap.

With Stradivarius violins, each one was the work of a master craftsman. Master Stradivari allowed no defects for instruments coming out of his workshop. However, the idea of zero defects is realistically impossible. No two violins are exactly the same. Each one has some flaw, no matter how tiny and insignificant.

Now, imagine a huge factory making one hundred thousand different components to be assembled into one telephone. Imagine mass manufacturing thousands of telephones. That means millions of factory systems and processes. The idea of creating every single component with absolute perfection is ludicrous. Stuff happens. Every single finished product slightly differs from all the others. As Deming noted in his final years, variation is a part of life.

Imagine the cannons on a pirate ship. When iron foundries first made cannons, they would create a vertical clay mold. To make the shape of the cannon’s cylinder, they would create the mold of a long shaft. Next, they stood a long piece of clay in the middle of the shaft. (Otherwise, the result would be a thick iron rod.) They’d pour the iron into the mold and let it cool. Then, they’d break all the clay out of the middle of the shaft. The result was a cylinder.

As you might imagine, this was not a precise process. Sometimes, workers would make the clay column a little too thick, resulting in thinner cannon walls. Too thin and the cannon could explode like a bomb, killing the pirates and maybe sinking the ship. Sometimes workers would make the clay column too thin, resulting in thicker cannon walls. Too thick and it wouldn’t hold any cannonballs. It was just an expensive piece of useless metal. The foundry manager had to decide how thick was too thick and how thin was too thin. These tolerance limits dictated if the cannon was a go or a no-go. Despite these limits, there was still considerable variation between cannons.

For the sake of explanation, let’s say the “perfect” size for the mouth of the cannon was 84 mm across. But because nothing was ever perfect, the foundry owner would allow anything between 83 mm to 85 mm to pass. Anything inside that range was a go; anything outside was a no-go and sent to the scrap heap.

A Swiss engineer in the 1700s named Jean Maritz came up with a better way to manufacture cannons. He did away with the interior clay column altogether and forged what was essentially a huge, thick iron rod, then used a drill to bore out the inside. His method resulted in much more precise cannon sizes.