Deming's Way

MYRON TRIBUS, Director Center for Advanced Engineering Study
Massachusetts Institute of Technology

The ultimate curse is to be a passenger on a large ship, to know t h a t
the ship is going to sink, to know precisely what to do to prevent it,
and to realize that no one will listen! This is the curse that has been
visited for a quarter of a century on W. Edwards Deming, revered i n
Japan as the Father of Quality Control, the man who taught t h e
Japanese how to produce goods of high quality at low cost.
Deming has known for at least twenty-five years that as t h e
Japanese developed their skill in manufacturing, no American
producer using conventional American approaches would be able t o
survive the competition. He knew that no amount of American
cleverness in innovation would make up for the fact that t h e
Japanese knew how to make goods of superior quality for less
money.

Deming had taught the Japanese that higher quality meant lower
cost, an idea foreign to most American managers. He foresaw what
would happen. Japanese entrepreneurs, observing a successful
business in America, such as textiles, steel, autos, or consumer
electronics to name but a few, could study the products, reverse
engineer them, and produce them at lower cost and higher quality. I f
one nation has access to another's technologies and is better at t h e
arts and sciences of mass production, it follows that the first nation
will invade the markets of the second. It is just a matter of time.
Thus it is that one by one, elements of our economy have suffered.
Deming has known for a quarter century that unless Americans
changed their approach to productivity and quality, the economy
would be destroyed. Deming has had to endure the ultimate curse. No
one would listen.

Americans do not appreciate the reverence with which Deming is
regarded by the Japanese. The annual Deming Prize is the most
coveted industrial award in Japan. Winning it is a national event,
reported on national television, with parades and ceremonies a t
which children present flowers to Dr. Deming and monuments a r e
erected in his honor. The Deming medal, with his profile on it, is
displayed prominently and proudly.

Japanese industrialists telephone him regularly at his home in
Washington, D.C. They invite him to Japan as often as he can come.
They listen. They apply his methods. And they have succeeded, a s
the world knows all too well. But in the United States, if managers
have heard of him at all (and most have not), they are likely to say,
"Oh yes. The quality control expert." Very few Americans
understand what Deming's way is all about. The president of t h e
United States appoints a blue ribbon panel to advise on problems of
productivity, and Deming's name is not even mentioned!
Too many people believe that Deming merely teaches simple
statistical quality control. They miss the point. American managers
travel to Japan, marvel at the behavior of the factory workers, a n d
conclude that it is something inherent in the Japanese culture. They
come home convinced that it is not their fault. They blame their
problems on the American worker, on taxes, on government
regulation, on the decay in society, in short, on anything except their
own managerial philosophies.

They do not realize that Deming has developed an entirely n ew
concept of how to manage systems of machines and people. He h a s
evolved a fundamentally new idea. It is revolutionary. It works.
Deming's way is not taught in any school of management in America.
Indeed, many things taught to managers in our schools and seminars
about how to manage enterprises are actually contrary to Deming's
way.

Deming's way has an interesting history. During World War 11 t h e
statisticians of America, under the leadership of such men as Deming,
Juran, and Shewhart, pioneered in new methods of control in t h e
wartime industry America's ability to produce large quantities of
high-quality armaments using an unskilled labor force was t h e
miracle that won the war. This miracle was due in no small measure
to the methods introduced by these men. They were not just
methods of statistical analysis. They represented the beginnings of
an entirely new way to look at the operation of a factory.

But when the war ended, the mass markets of America were waiting
to be filled, and skillful production management was not required. By
1950 many of the lessons of the war were discarded. New managers
came to run new factories. They had little need, they believed, for
methods to increase quality and productivity. They did not study
them. Business schools did not teach them.

The American market was a powerful deterrent to anyone wh o
wanted to concentrate on high quality. Apparently unsaturable
markets swallowed goods of inferior quality. Americans accepted
appliances of inadequate performance as though they represented
the best that could possibly be done. Americans firmly believed t h a t
to increase the reliability of their appliances would have required a n
increase in cost. And until Deming's way showed that this was false,
they had no choice but to believe the myth. By and large they still
do.

The years 1941-1945 had taught Deming a profound lesson. He
realized that a whole new philosophy of management was possible
based on his experiences. And he saw that the basic idea could b e
put to work in any industry, even the service industries and in ways
not thought of during World War 11. But Americans would not listen.
In retrospect we can understand why. With the rest of the world i n
disarray, with America providing help to the rest of the world
through relief programs and the Marshall Plan, exploding hydrogen
bombs, winning Nobel Prizes, expanding its economy explosively,
believing it faced the limitless frontier, Americans did not believe
they needed advice on how to produce quality goods and increase
productivity. After all, weren't we Number One? Our problems were
not productivity or innovativeness. We had no competition. I f
anything, our problems were going to be what to do to fill o u r
leisure-time.

The Japanese faced a different story. Their country was conquered,
the economy was in ruins. Gone were the dreams of a "Greater East
Asia Co-Prosperity Sphere," based on Japanese military conquest.
Their island, smaller than California but with a population half t h e
size of the United States and with very few natural resources, faced a
momentous challenge. In 1950, with the aid of the MacArthur
government, the Japanese Union of Science and Engineering invited
Dr. Deming to come to Japan and tell them about quality control. For
them this was a monumental bit of good luck .
During his first visits to Japan, Deming examined what they were
doing, studied their work force and its habits, and became convinced
his methods could be applied there. With his characteristic audacity
he issued an invitation to the top forty-five industrialists in Japan t o
come to a meeting the next Tuesday; forty-five came. At the meeting
he told them about his methods and promised them that if they
would apply them, within five years Japan would be an important
factor in international trade.

In later years some of the attendees were to write about t h a t
meeting and say that they and their associates did not believe this
strange American who was proposing an impossible scheme. The
Japanese industrialists had a much more limited objective. They just
wanted to bring Japan back to the level it had attained prior t o
World War II. They did not believe Deming, but having attended t h e
meeting, they felt they would lose face if they did not at least give
his idea a try.

Within six weeks some of the industrialists reported productivity
gains of as much as 30 percent without purchasing any n ew
equipment. When the industrialists compared notes, they realized
that Deming's way really worked. They then devoted their time a n d
energies to implementing it. The Japanese Union of Science a n d
Engineering set up study teams. Deming suddenly found himself with
a development laboratory with 90 million subjects in it. The whole of
Japan's industry became his proving ground Today Dr. Deming ha s
had over thirty year's experience with this laboratory.

He has had a chance to test his ideas. When he says he knows what
to do to increase quality and at the same time cut costs, he knows
whereof he speaks. His credibility is incredible! What Is Deming's Way?
Consider a trucking firm managed by a man educated according t o
current management methods taught in our schools of business
management. He will consider his job to be to run the company a s
profitably as he can and to expand its business. To do so he may call
on the best consultants he can get to help him design the best
possible system. He may set up work standards for the drivers a n d
institute computer-based procedures to keep track of t h e
performance of the drivers, trucks, and dispatchers. He will study hi s
markets and their opportunities. And he will keep extensive records
of income and expenses, ever on the alert for opportunities to profit.
Of course, he will not be able to do these things alone, and as hi s
organization grows, he will institute methods to see that his desires
for efficiency and performance are carried out. Perhaps he will adopt
management by objectives and teach it to his subordinates. He ma y
assign as much as 5 percent of his work force to data gathering a n d
performance monitoring, ever searching for possible profit
opportunities.

In short, his idea of a good manager is one who sets up a system,
directs the work through subordinates, and by making crisp a n d
unambiguous assignments develops a basis to set standards of
performance for his employees. He sets goals and production targets
for his people. He rates the employees as objectively as he can,
sometimes even calling on consultants to help him do so. He
identifies poor performers and gives them further education to meet
the work standards, or he replaces them He hopes thereby to create
the most efficient system possible.

Contrast this with the behavior of a manager who follows Deming's
way. This manager sees his job as requiring him to provide a
consistency and continuity of purpose for his organization and t o
seek ever more efficient ways to meet its purpose. For him, making a
profit is necessary for survival but is by no means the main purpose
of his organization. His view of the purpose of his organization is t o
provide the best and least-cost transportation system for hi s
customers and continuity of employment for his workers He does not
view the concepts of "best" and "least-cost" as contradictory .
He will consider that he and the workers have a natural division of
labor. They are responsible for doing the work within the system,
and he is responsible for improving the system. However, he realizes
that the potentials for improving the system are never ending so h e
does not call on consultants to teach him how to redesign the "best"
system for he knows that it doesn't exist. Any system can b e
continuously improved on. He knows that the only people who really
know where the potentials for improvement lie are the workers
themselves. He knows that the system is subject to great variability.
Traffic conditions change, trucks break down, shipping docks are not
always ready to discharge or receive goods, mistakes are made i n
routing or addressing. There are countless ways for the system to go
wrong and out of control, decreasing quality and increasing cost. He
knows that these ways occur randomly. To make it possible for h im
and the workers to work together, he knows they must regard t h e
system in the same way and speak a common language. Therefore h e
learns elementary statistics and teaches it to the workers, engaging
an expert consultant in statistics if necessary to help him and t h e
workers when they come to a problem beyond elementary statistics.
All of his employees learn to keep their own statistics. Truck drivers
keep track of how long they have to wait at docks and study t h e
circumstances at each event. They develop their own control charts
and look for trends, for correlations with other events, usually events
beyond their control. The drivers meet with each other a n d
sometimes with the dispatcher and compare notes. They keep data
on the performance of their trucks and discuss their statistical charts
with the purchasing agent and each other. Based on these data t h e
manager, who is responsible for the system, makes the changes, a n d
the workers, based on their statistical information, help him to learn
how effective the changes have been. When the manager instructs
the purchasing agent to buy on "quality," not just on first cost, t h e
purchasing agent has the information from the drivers with which t o
do just that, and to demonstrate that he has done so. Everyone in t h e
system is involved in studying it and proposing how to improve it.
Everyone spends about 5 percent of their time in this pursuit. No one
spends 100 percent of the time, except the company statisticians. The
employees will see the setting of work standards as a dumb idea
since it inhibits their ability to improve the system. They will not
need to "manage by objectives" because they will be engaged i n
consistently redefining their objectives themselves and recording t h e
performance of the system.

The workers and the manager will be aware of the results reported
by Juran--that in most systems 80 to 85 percent of the problems a r e
with the system and 15 to 20 percent are with the worker. This is a n
important understanding, for it can free the workers to speak out
without fear, a quality of the work place that the manager
assiduously cultivates.

Under Deming's way, the manager understands that he needs t h e
workers not only to do the work but to help him to improve t h e
system. Thus he will not regard them simply as robots made of flesh
and bone, but he will rather consider them as thinking, creative
human beings. No one will have to teach him to be nice to people. He
will not try to motivate them with empty slogans, such as "Zero
Defects!" Because they will be measuring and counting the defects
themselves and helping him to remove them, there will be no need
for the slogans. He will not ask them to sign pledges to be polite t o
customers. Nor will he select the "Polite Trucker of the Week."
Instead, he and they will have been studying the records of repeat
orders and asking what they can do to improve the statistics.
From time to time he asks for volunteers from his work force to take
time out to interview customers and vendors, to understand what
they want or can supply to provide better service. They report back
to him and the rest of the work force on what they have found,
statistically analyzed.

In short, the Deming-trained manager will have a natural basis for
building a team and will not have introduced adversarial relations.
Under currently taught methods of management it is presumed t h a t
the relation between boss and worker is inherently adversarial. The
result is that bosses who wish to fit the understood image must b e
careful not to develop too intimate a relationship to the worker, lest
they lose their objectivity in judging and rewarding performance.
(Recall the restrictions on officers in the military against mixing
socially with the enlisted men. This is probably a good idea for a
system in which no one is supposed to propose improvements!)
Under Deming's way the boss and worker work together naturally
and can even afford to like each other!

Deming's way is therefore more than just attention to quality control.
It is a managerial philosophy for achieving lower cost and higher
quality. And it works not only in the factory, but in hospitals, i n
service industries, and even in the office.

It is in seeing how a changed managerial self-image could lead t o
such phenomenal successes that Deming had one of those brilliant
flashes of insight that few of us are privileged to have. As Newton
with the apple, Einstein with relativity, Freud with the subconscious,
Deming saw a new way for management.

The basic idea that Deming had is this If management is to b e
responsible for improving something as complicated as a modern
assembly of machines and people (whether in the factory, t h e
hospital, the office, or anywhere else), managers must have a way of
learning (1) which parts of the problems are due to the workers a n d
(2) which parts are due to the system. Deming understood that this
can happen only if the two circumstances are fulfilled. The worker
and the management can speak the same language.

The management uses the workers as essential "instruments" i n
understanding what is happening at the place where the work gets
done.

Given the complexity of modern systems, there is no way t h e
managers can begin to understand what is happening without t h e
full cooperation of the workers. And even given a cooperative spirit,
there is no way they can work together if they do not have a
language suitable for discussing the inherent randomness in such
systems.

And what is that language? It is the language of statistics. Deming, a n
established and esteemed statistician, understood this immediately.
We all use statistics every day. It dominates the reporting of t h e
sports announcers. We gamble. We handicap the horses. We listen t o
Jimmy the Greek. But for some reason or other, among "educated"
people, there is a tendency to shun statistics and to avoid even t h e
simplest taking of averages, the calculation of how much variation t o
expect, or when to decide that something is "unusual." Among
"educated people" (for example, hosts on TV talk shows) as soon a s
something in elementary mathematics is introduced, one hears, "Oh, I
was never very good with numbers," spoken as a badge of honor, a s
though the ability to do so marked one as a drudge Among t h e
"common folk" there may be "mathaphobia" but no one is proud of it!
Uncontrolled variations in a factory or other place of work lead t o
low productivity, poor quality, and increased need for capital
equipment to obtain high rates of production. If the management is
to control the variation, there is no escape from learning how to u s e
statistics. Furthermore, if the cooperation of the workers is to b e
obtained, they, too, must learn the language of statistics.
Most American managers today tend to take the system as given a n d
to try to get the most out of it as is. Thus they believe the problems
of the factory are to be found in better understanding of morale, t h e
work ethic, work standards, job definitions, more communications,
slogans, exhortations, personnel selection, better record keeping,
better union bargaining, and so on. They equate increased
productivity only with increased capital spending (e.g., o n
automation) and have an easy out in blaming the tax laws, interest
rates, and labor rates. It never occurs to them that people do not
have to work harder, just smarter. And they do not comprehend t h a t
they must provide the leadership in working smarter. "Overmanaged
and Underled" is the way Don Alstadt, CEO of the Lord Corporation,
describes American industry.

Of course, better communications, better equipment, b e t t e r
understanding of each job can contribute to efficiency. But these
contributions need to be seen as part of an overall managerial
framework, not isolated elements. When seen as part of an attempt
to improve the system and not just attempt to get more out of t h e
system as it is, these techniques take on a much different meaning.
In the typical corporate environment, the conventional system often
works in a self-defeating way. How often is someone promoted
because he or she knows how to get things done "in spite of t h e
system"? And how often is someone passed over because he or s h e
spent too much time developing corporate relations in a setting i n
which departmental infighting was the order of the day? How many
managers make exorbitant demands on their workers as a way of
"keeping them on their toes" rather than taking the time to discuss
with them how to change the system so that the workers can do their
jobs better?

If the workers and the bosses both speak the language of statistics,
they will have something to discuss at their quality circle meetings.
If they do not, quality circles can be an invitation to controversy a n d
misunderstanding.

If managers adopt Deming's way, they will understand that they
need the workers - not just as arms and legs to do what they are bid,
but as intelligent human beings who can provide insights into how t o
improve the output and efficiency of the place. And we know t h a t
the improvement in management labor relations can come from such
an attitude, honestly felt and experienced.

What is happening to Deming's way in America today? Dr. Deming
has undertaken to teach Americans because he senses that n ow
people are ready to listen to him. Despite his age (he is in hi s
eighties) he has undertaken to give at least eighteen four-day
seminars in 1982 plus many days of individual consultation. He will
have spoken toapproximately 2500 executives this year alone. Of this
experience he says, "I light a number of fires. Some of them ar e
continuing to burn."

One of the fires that has burned more brightly is in Nashua, New
Hampshire where William Conway, president of the Nashua
Corporation, has been implementing Deming's methods for almost
two years. He credits Deming with the turnaround in his company's
fortunes. With the example of the nearby Nashua Corporation a n d
Bill Conway's help, twenty-two companies in nearby Lawrence,
Massachusetts, have banded together to learn to apply Deming's
methods.

Probably the biggest success of all is at the Ford Motor Company,
where Don Peterson, Ford's president, has declared that "the Ford
Motor Company is committed to the adoption of the methods of W.
Edwards Deming." The recent contract settlement with the UAW
indicates how serious Ford is. In a videotape prepared to convince
Ford's employees, especially its middle management, case examples
of the success of Deming's methods are given.

Similar pronouncements have been made by representatives of t h e
Pontiac Division of General Motors. From the experiences at Ford, Nashua Corporation, and several Japanese-managed companies in the United States, we now have
substantial evidence that Deming's methods work wherever they
have been seriously tried.

We may say with confidence that the poor showing of American
industry in competition with the Japanese is entirely due to t h e
management The attitude of the government plays a role, of course,
but in many cases this is also the result of management's approaches
in the past. It is the culture of the managers that must change, a n d
quickly.

There are some who say that ultimately Dr. Deming will b e
recognized as the Father of the Second Industrial Revolution. Be t h a t
as it may, he has this to say about future competition There will b e
no room for managers who do not know how to work with their
people to produce high quality goods at low cost. High reliability
cannot be secured without worker cooperation. Complex systems
cannot be understood without statistics. In the competitive world of
the future, companies which have not mastered these ideas will
simply disappear. There will be no excuses."