Six Sigma Lessons from Deming, Part 2

Opinion

Published: Tuesday, March 4, 2008 - 22:00

In Part 1 of this article I discussed Six Sigma failures and the fundamental flaws in the term Six Sigma, as seen in figure 1 below, which highlights the flaws in the hundreds of Six Sigma web sites displaying the nonsense of out-of-control processes that are the result of the claimed “unavoidable” 1.5 sigma drift.

Figure 1: The consequences of Six Sigma’s +/–1.5 sigma shift that Mikel Harry claims for every process—a process wildly out of control.

Perhaps the most easily recognized difference between Six Sigma and Deming’s teachings, is Six Sigma’s belt system. Black Belts are given responsibility for assigning improvement projects. Deming suggested that quality “was everyone’s responsibility,” and at the same time most companies had quality experts in the form of quality engineers. A quality engineer had several years training compared to Six Sigma’s highest level of belt, the Black Belt, with typically four weeks of training. With this meager level of training, massive responsibility for achieving hundreds of thousands of dollars cost savings is assigned to the Black Belts.

Six Sigma’s system of belts—Black, Green, Yellow, and more recently, White Belts—has been criticized because it builds elitism. Deming stressed the importance of “breaking down barriers.” He stressed the importance of people working together as a team rather than being directed by an artificial hierarchy of poorly trained people holding belts. All team members should be encouraged to educate themselves. The operator who spends all day every day at a work station has the greatest knowledge of the machine and the greatest potential contribution to make to quality improvement and cost savings.

Deming placed great importance on pride of workmanship in achieving good quality. He pointed out how management and workers had become a commodity to be bought or disposed of. How can employees take pride in their work when the numbers are more important than quality? Deming stressed the importance of learning the psychology of individuals and groups. Six Sigma has returned quality thinking to the old days before Deming. As Mikel Harry states, “In short, numbers-oriented thinking applies to people as much as it applies to processes and products.” Treating people as numbers rather than individuals is indeed short sighted. It may or may not give short-term gains but a lack of caring for people and treating employees as commodities will lead to a company’s downfall.

Deming stated that everyone should work toward improving quality. He stressed the importance of individuals to a company’s success. This may seem like common sense, but Six Sigma focuses on an elite band of people holding belts and disregards what Mikel Harry refers to as “the masses.” Such an approach alienates the workers. It is contradictory to modern thinking about incorporating an emotionally intelligent approach to management. Every employee is important to a company’s success and every employee should be appropriately trained and supported. As Daniel Goleman, the world’s leader in the study of emotional intelligence, states: “To the degree your organizational climate nourishes these competencies, your organization will be more effective and productive. You will maximize your group’s intelligence, the synergistic interaction of every person’s best talents.”

A defining characteristic of Six Sigma has been its wealth of superlatives and exaggerations, such as “breakthrough strategy” and “transformation.” The basis of this “breakthrough” is described by Mikel Harry: “In contrast with TQM [total quality management], Six Sigma operates on a very simple principle. Whatever you do to improve quality should simultaneously and immediately improve the business in a visibly quantifiable and verifiable way.” This is decidedly deceptive. Deming states, "Productivity increases as quality improves.” He clearly described how business success was based on quality. This has been demonstrated by the rapid rise of Deming-orientated companies such as Toyota. “Everyday I think about what he meant to us,” says Shoichiro Toyoda, Ph.D., founder and chairman of Toyota Motor Corp. “Deming is the core of our management.”

Motorola’s Malcolm Baldrige Quality Award in 1988 is often cited as an example of Six Sigma’s success, but in reality it was eight years of TQM programs that led to this award. Incidentally, TQM is often used synonymously with Deming, although Deming didn’t use the term in his teachings.

Six Sigma’s superlatives and wild claims themselves are in marked contrast to Deming’s "eliminate slogans and exhortations.” Strangely, Mikel Harry claims that slogans such as “zero defects” are “devoid of meaning,” and at the same time claims that “3.4 dpmo” is not. In reality, neither Phil Crosby’s “zero defects” nor Six Sigma’s 3.4 defects has any real meaning, because defect levels depend on where specification limits are chosen. Deming stressed “Focus on outcome (management by numbers, zero defects, meet specifications ) must be abolished.…”

Numbers-orientated thinking is central to Six Sigma, and a major difference with Deming’s teachings. Six Sigma preaches management by objectives with a target of 3.4 dpmo. This is a retrograde step to a management style of the 1950s, first outlined by Peter Drucker. Deming gives dozens of examples of how this simplistic style of management doesn’t work. “A numerical goal leads to distortion and faking, especially when the system is not capable of meeting the goal.” A person must have numbers to show and churns out the required numbers by whatever means is most convenient. Pride of workmanship and quality disappear. Perhaps it’s more difficult for simplistic managers to understand that if the system is corrected, good numbers will flow automatically. However, only by understanding this basic principle, can there be an outcome of real quality and business success.

A shortcoming of Six Sigma has been its focus on defects without consideration of waste reduction. This has led to the growth of lean or lean Sigma. Again we should turn to Deming. While it’s widely known that Deming stressed the importance of reduction of variation, he states quite clearly that the reason productivity increases as quality improves is that there is less waste. This means less waste of man hours and machine hours, leading to the manufacture of improved products and services.

Deming’s 8th point in his 14 Management Points is “Drive out fear.” By contrast, Six Sigma promotes an environment of fear. Modern thinkers such as Goleman describe fear as a negative emotion and a long-term demotivator, despite possible short-term gains. Fear is the easiest of motivations to instill, yet the least effective, resulting in inner anger and resentment. The most effective form of motivation is causal, where people are motivated to work for a cause or something they believe in. This can only happen when all employees are respected, trained, and working toward a common goal. This can only happen by driving out fear, by allowing employees to feel secure, to ask questions, to increase their knowledge, to feel respected.

Six Sigma and Deming describe the importance of leadership. Six Sigma’s approach is to establish a hierarchy of belts, with management by objectives, fear, and numbers. Deming pointed out how management is responsible for “the system,” which is responsible for 90 percent of problems—”Don’t blame the individual, fix the system for them.” Deming described an ongoing cycle of continual improvement whereby this can be achieved, rather than Six Sigma’s single cycle.

Most Six Sigma programs include hypothesis testing. In stark contrast, Deming uses this as an example of “poor teaching of statistical methods.” He stated that hypothesis testing has no application in analytical problems in science and industry. The reason for this difference is that Six Sigma fails to differentiate between what Deming called enumerative studies and analytic studies. The aim of an enumerative study is a description of a fixed frame of material. The material is sampled randomly and assumed to fit some particular distribution. Hypothesis testing is a valid technique in enumerative studies, such as surveys or psychological tests. It’s perhaps no coincidence that the latter field is where Six Sigma’s prominent proponent, Mikel Harry, has his tertiary education.

An analytical study aims to improve the process that creates the material, usually while material is being produced continuously. Shewhart pointed out that the form of the distribution of data will always be unknown, and random sampling doesn’t have the same meaning as in an enumerative study. Without a probability model, hypothesis tests become meaningless. For example, consider that historical data for two processes, A and B, are compared and A appears better than B at a 90 percent confidence level. However at a 95 percent confidence level, there’s no apparent difference. This is confusing, and it gives no indication whatsoever as to future behavior of the processes.

Enumerative studies and hypothesis tests dispose of the data’s time element. The results are purely historical and make no prediction as to future behavior. Shewhart’s control charts are quite different in that they provide an inference to the future. The future behavior of an in-control process is predictable, while an out of control process is not.

Six Sigma’s lack of differentiation between analytical and enumerative studies has lead to a false belief that control charts are based on the normal probability model. Common statements taught in Six Sigma classes such as “99.7 percent of points lie within control limits” are quite false. Deming described how this "derails effective study and use of control charts.” While having validity in statistics, Shewhart control charts are based on economics. Control limits are intended to give signals as to when it’s most economic to investigate special causes.

Six Sigma has introduced other dubious statistical practices such as overlaying normal distributions on histograms of process data, something that Deming would have scoffed at. Wheeler has shown that such attempts at distribution fitting are meaningless. He has shown that 3,200 data points are needed to fit a distribution out to only +/–3 sigma and the distribution is likely to change with time anyway. Even worse, attempts at distribution fitting distract users from the real purpose of process data histograms: as a tool to gain a better understanding of the process.

The “Seven Tools of Quality” are widely associated with TQM. The group of seven tools are attributed to Kaoru Ishikawa, “The seven QC Tools, if used skillfully, will enable 95 percent of workplace problems to be solved.” The seven grew to 14 with the seven “new” tools and has continued to creep to up to the 40 or so seen in Six Sigma programs today. Deming mainly used the key tools: cause and effect analysis, Pareto, flow charts, histograms, run charts, and control charts. More tools don’t imply better quality. Given the widespread misuse of statistics, it’s surely better to learn to use the primary tools correctly.

In summary, Six Sigma programs have a great deal to learn from Deming. While Six Sigma may be “80 percent TQM,” the remaining 20 percent needs a great deal of improvement in terms of statistics and management. As long as companies are managed on the basis of a poor understanding of the analysis of data and a poor approach to working with people, Six Sigma companies will continue to fail.

References:
1. Lean Six Sigma—An Oxymoron? by Mike Micklewright
2. “New rule: Look out, not in.” Betsy Morris, Fortune, July 11, 2006
3. “The ‘Six Sigma’ Factor for Home Depot” Karen Richardson, The Wall St Journal
4. “Six Sigma Stigma” Martin Kihn, Fast Company, September 2005
5. Deming, The New Economics, 2d edition, p.31
6. Error rate studies: http://panko.shidler.hawaii.edu/HumanErr/index.htm
7. “Making War on Defects,” Bill Smith, Motorola. IEEE Spectrum 1993
8. “Sick Sigma” Dr. A. Burns. Quality Digest. April 2006 : http://qualitydigest.com/IQedit/QDarticle_text.lasso?articleid=8819
9. “Sick Sigma Part 2. Tail Wagging Its Dog” Dr. A. Burns. Quality Digest. February 2007: http://qualitydigest.com/IQedit/QDarticle_text.lasso?articleid=11905
10. Mikel Harry http://www.mikeljharry.com/story.php?cid=6
11. Daniel Goleman. “Working With Emotional Intelligence,” http://www.danielgoleman.info/blog/emotional-intelligence/
12. “The Mist of Six Sigma,” Alan Ramias, BPTrends October 2005.
13. The Practice of Management, Peter F Drucker, 1954.
14. Out of The Crisis Edwards Deming. 1982
15. The Nielson Group http://www.nielsongroup.com/articles/articles_climateformotivation.shtml
16. Advanced Topics in Statistical Process Control, Donald Wheeler. SPC Press 1995
17. Normality and The Process Behaviour Chart, Donald Wheeler. SPC Press 2006

About The Author

Anthony D. Burns

Anthony Burns, Ph.D., has a bachelor of engineering and a doctorate in chemical engineering from the University of New South Wales in Sydney, Australia. He has 36 years of experience and his company, MicroMultimedia Pty. Ltd., is responsible for the development of the e-learning quality product Q-Skills and its support tools.