Statistical and operating techniques were developed to control quality of manufactured products

 

In the early decades of the twentieth century, considerable research took place at AT&T's Bell Laboratories to understand and improve underlying processes supporting business operations. Walter Shewhart of Bell Labs published the results of this work in 1931 under the title Economic Control of Quality of Manufactured Product. The challenges to be met at AT&T were built around the felt need to assure quality and consistency in telephones and manufactured products related to telecommunications infrastructure. The object was to eliminate any product or service inconsistencies in the experience of the public. Dr. Shewhart's work consisted of a sophisticated, yet pragmatic approach to the use of statistics to bring processes into a state of control. Such control was not dependent on vague interpretations of the term--nor of motivational slogans, accounting classifications, or other obscure characteristics--but straightforward application of specific statistical techniques adapted from similar techniques applied in natural and scientific settings.

 

 

This work represented a seminal breakthrough. With the consolidation of the telephone industry as a regulated monopoly and the aggregation of research facilities at Bell Laboratories, there was a considerably high level of capital to support such research. The resulting techniques enabled AT&T to benefit further from its monopoly position--reducing costs considerably and establishing a reputation for quality and consistency that carries through in current times.

 

To the general benefit, AT&T published the work of Dr. Shewhart so that it could be used by other economic institutions. W. Edwards Deming--an graduate student intern who developed a friendship with Dr. Shewhart--became an advocate of these techniques in the 1930s and thereafter. One result was a companion book by Dr. Shewhart in 1939 published by the United States Department of Agriculture titled Statistical Method from the Viewpoint of Quality Control--which further explained the phenomenon.

 

 

Manual statistical quality control methods proven out by U.S. in WWII and Japan in 1950s and 1960s

 

With the outbreak of World War II, Dr. Deming spearheaded an effort to teach these techniques to U.S. production managers as a means of supporting the wartime production effort. Under the sponsorship of Stanford University, the techniques were taught to over 30,000 managers--measurably assisting in the effort to provide reliable armaments in large quantities to both theaters in that war.

 

Shewhart's technique is designed to measure characteristics of an established process and to transform those measurements using specific formulas. The purpose of this exercise is to provide guidelines to workers and management to prevent two kinds of errors--those of commission and those of omission. The techniques are used to separate out elements of a process that are in control from a statistical standpoint and those that are out of control from that perspective. By so doing, managers and workers can choose between actions that will not interfere with those parts of the process that are in control while correcting elements of the process that are out of control.

 

As was later documented in his 1982 book, Out of the Crisis, Deming presented these techniques and his implementation principles as an antidote to many of the problems associated with industrial specialization, arbitrary job definition, motion studies, and other techniques of the 'scientific management' movement. In the Shewhart/Deming world, there was a high degree of collaboration among specialists. Deming clearly makes reference to two classes of employees in a working environment--the managers and the workers. Managers are responsible for the overall system in Deming's opinion and the workers must work within the system provided by the company (the managers).

 

Deming's philosophy is that all managers and workers are to roll up their sleeves and apply Shewhart's techniques to solve problems that they uncover. This is the basis for what Deming termed Profound Knowledge--something that should be developed and used by managers and workers alike. He was not an advocate for emphasis on career specialization--considering such to result in sub-optimization of the organization. In his view, by collectively mastering the control requirements of an institution's key processes, companies could achieve further gains in terms of cost, quality, and delivery times--with increased flexibility, less waste, and improved morale.

 

Deming was the great teacher of these principles. He registered great pride at how they were used in support of the United States in World War II, but he also felt rejection--to the point of disgust--because of the speed with which U.S. industrialists abandoned the techniques with the end of the war. With the postwar shift back to industrial and consumer markets, managers of most companies--faced with the prospects for global expansion without competition from Europe or Asia--did not feel a need to use the techniques in question, which did require time, attention, and effort. At this point, the mass production model grew significantly--allowing U.S. firms to not only expand domestically, but to actually dictate market trends and product features to end users worldwide--which were nonetheless eager to accept their new products' features based on availability alone.

 

Based on his experience with the techniques--by now referred to as Statistical Process Control (SPC)--Deming was convinced that they could be used to support service-related processes as well as manufacturing processes. In fact, in his experience, service-related activities offered even more promise for improvement in this regard than would normally occur in manufacturing environments.

 

After the general postwar abandonment of SPC by U.S. industry, Deming concluded that the techniques could only be effectively employed with fundamental support at the board of director level on down. He concluded that the failure of U.S. manufacturers to continue to benefit from the techniques was due to a lack of understanding of the techniques that had been largely administered by the government and by temporary combinations of people and organizations quickly assembled to deal with wartime challenges.

 

 

Deming's story and the ongoing saga of SPC took an interesting turn in postwar Japan--as has been recounted by Deming and others. In short, in Japan to assist in their efforts to assemble a post-war census, Deming mentioned that he was aware of certain techniques that could be useful to Japan in its efforts to rebuild. When his contacts in Japan indicated that there was interest in learning and applying the techniques, Dr. Deming indicated that he would only teach these techniques to top management of Japan's industrial combines. He did not want a repetition of what had happened in the U.S. after the war. In short, they agreed and in the summer of 1950 he taught these techniques and their implications to groups that have been later calculated to have represented approximately seventy-five percent of the industrial capacity of Japan.

 

Deming's 'flavor' of SPC served as the conceptual foundation for the Japanese recovery which made major inroads in Western markets for capital and consumer goods in subsequent decades. This development has been widely acknowledged in Japan and in the West--though Deming was largely unknown in the United States until he was almost inadvertently 'discovered' by a U.S. reporter in the process of researching other issues. Dr. Deming subsequently was in great demand as a teacher of SPC techniques, consultant, and commentator until his death in 1993. Through his efforts, several Western industries and companies were able to answer to the Japanese challenge in the 1980s and 1990s. Interestingly, Ford Motor--earlier a leader in mass production techniques--also led out in the movement toward Deming's techniques, becoming a leader in lean production.

 

Many important considerations were derived from Deming's work and the demonstrated success of SPC. The procedures and philosophies behind the movement literally forced institutional decision-makers to take a broader view of overall institutional performance than historically had been the case. There was nothing in the work of Frederick W. Taylor that encouraged or facilitated a global view. Scientific management under Mr. Taylor concentrated largely on efficiency and task optimization. Deming's work--on the other hand--focused on optimization of the delivery system overall, not based on improvement of individual tasks at the expense of the whole system.

 

Deming directed his attentions largely to people-oriented issues--with an effectiveness that caused many to misunderstand his purpose. He was not a cheerleader. His purpose was to persuade managers and workers alike to assume a new spirit of cooperation that itself was in tune with the concept of statistical control. His purpose was to simplify, not complicate working relationships and emphasize similarities, rather than differences between specialists. His teachings--with the purpose of reducing variation in processes through use of specific statistical techniques--were also very helpful in bringing the firm into top-down control. He taught concepts that would bring more cooperation and harmony into the process of defining and delegating institutional responsibilities.

 

William Glasser, M.D.--a prominent expert in Control Theory--indicates that in so doing, Deming has introduced a critical factor to future development of how organizations operate.

 

Deming talks extensively about the need to understand psychology and points out clearly that he believes human beings are intrinsically, rather than extrinsically, motivated. In doing so, he shows that he understands the basic premise of control theory.  

Control Theory provides important insights into the requirements and benefits of Dual Control. In order to establish an environment of trust and cooperation sufficient to reduce variation in processes according to the Shewhart model, certain conditions of cooperation and trust must exist. Glasser indicates that the main idea of Control Theory is that managers must lead their workers to accomplishing their assigned tasks--not boss them. Control is thus achieved through cooperation. Deming's observations about the positive example of symphony orchestras refer to this and other principles and conditions.

 

By the same token, Deming's techniques ignored institutional boundaries in favor of statistics that reflected the combined performance of however many partner companies participated in the business activity at hand. Thus, using SPC techniques as outlined by Shewhart and Deming, system optimization was possible--with unforeseen and unprecedented benefits.

 

 

Mass production--the assumed environment supporting Mr. Taylor's concepts and procedures--was challenged by Deming's work. The new industrial phenomenon was described as an outgrowth of the wide application of SPC in Japan and later in the West that formed a combination between mass production and the craft movement that was largely replaced in the Industrial Revolution. This new industrial type has been referred to in a number of ways--including flexible manufacturing, just-in-time manufacturing, agile manufacturing, lean production, and under other descriptive terms. One outgrowth of this new structure has been greater information sharing and collaboration among partners. Deming's work, however, was not focused on such collaboration. It's primary objective was that of reducing process variation by means of the application and mastery of specific statistical techniques--leading to selective collaboration and information sharing based on the conclusions derived from those techniques. Deming did not advocate collaboration for collaboration's sake--in fact warning against such a development.

 

Deming's work also resulted in a newfound emphasis on process as a defined term--as a specific industrial phenomenon. Prior to popularization of his work in the late 1980s, there was no such distinction in the literature or vocabulary of Western business management. In the world of Shewhart and Deming, an industrial process was defined and evaluated as much as possible as would a such process found in nature. Such phenomena were to be studied in a specific way in order to support certain decision-making options.

 

With its popularization, the concept of an 'industrial process' to be studied and analyzed in its state of natural evolution was largely converted in the business lexicon to that of a 'business process' to be mapped out with the objective of radical revision of the operational structure of an enterprise. This was largely the result of the short-lived, but widespread reengineering movement--which was used as rationale for the development of many large-scale computing projects, but which was found to be an impractical means of dealing with the reality of institutional operations. There was selective borrowing of concepts from the Deming world for the reengineering model, but the heart of the Shewhart/Deming model was seldom used.

 

Shewhart's work--popularized, evangelized, and explained by Deming--has stood the test of time in similar fashion to that of the symphony orchestra model. The physics and science behind the symphony is so correct as to have stood the test of time for almost two centuries. The science behind Shewhart's SPC--though proven and unchanged for almost a century--remains outside of the reach of mainstream business. Its implications are not widely known--though Deming himself reached a measure of celebrity in his last years.

 

Of course, there is considerable use of SPC in many companies and many industries. Many consultants and managers were trained by Deming himself, and the Quality Movement is an active phenomenon in the business world. The techniques are typically not principal management tools in any but a few firms. For the most part--to be used and to be useful--they must be applied by hand or by manually transferring underlying data to ancillary personal computers or secondary systems. They are not principal elements of enterprise systems such that the elements of primary processes can be defined, evaluated, and readily modified--as is necessary for successful application of the Shewhart/Deming tools.