Japanese Quality Management Systems and the Taguchi Approach


Taguchi Methods are a collection of design principles and methodologies aimed at improving product and process quality. These constitute a body of knowledge variously known as quality engineering, robust engineering, or, particularly in the United States, Taguchi Methods, after its exponent, Dr. Genichi Taguchi (see Sidebars 2.1, 2.2, and 2.3).

Sidebar 2.1: The Life and Times of Dr. Genichi Taguchi[8], [9]

Dr. Genichi Taguchi has had a profound effect on the emergence of design-focused quality management methodologies. He was born in 1924 in Japan. After serving in the Astronomical Department of the Navigation Institute of the Imperial Japanese Navy in from 1942 to 1945, he worked in the Ministry of Public Health and Welfare and at the Institute of Statistical Mathematics, Ministry of Education. He learned much about experimental design techniques and the use of orthogonal arrays from the prize-winning Japanese statistician Matosaburo Masuyama, whom he met while he was working at the Ministry of Public Health and Welfare. This also led to his early involvement as a consultant to Morinaga Pharmaceuticals and its parent company, Morinaga Seika.

In 1950 Taguchi joined the newly founded Electrical Communications Laboratory of the Nippon Telephone and Telegraph Company with the purpose of increasing the productivity of its R&D activities by training engineers in effective techniques. He stayed for more than 12 years, during which time he began to develop his methodology, which is now known as Taguchi Methods or robust engineering (see Sidebar 2.2). While working at the Electrical Communications Laboratory, he consulted widely in Japanese industry. Accordingly, Japanese companies began applying Taguchi Methods extensively from the early 1950s, including Toyota and its subsidiaries. Taguchi's first book, which introduced orthogonal arrays, was published in 1951.

In 1954 and 1955, Taguchi was a visiting professor at the Indian Statistical Institute in Calcutta (now named Kolkata), India. During this visit he met the famous statisticians Ronald A. Fisher and Walter A. Shewhart. In 1957 and 1958 he published the first version of his two-volume book Design of Experiments. His first visit to the United States was in 1962 as a visiting research associate at Princeton University, during which he visited AT&T Bell Laboratories. At Princeton, Taguchi was hosted by the eminent statistician John Tukey, who arranged for him to work with the industrial statisticians at Bell Laboratories. In 1962 Taguchi received a PhD from Kyushu University.

In 1964 Taguchi became a professor at Aoyama Gakuin University in Tokyo, a position he held until 1982. In 1966 Taguchi and several coauthors wrote Management by Total Results, which was translated into Chinese by Yuin Wu, his collaborator on a number of later publications. At this stage, Taguchi Methods were still essentially unknown in the West, although applications were taking place in Taiwan and India. In this period and throughout the 1970s, most applications of Taguchi Methods were on production processes; the shift to product design began later. In the early 1970s Taguchi developed the concept of Quality Loss Function. He published two other books in the 1970s and the third edition of Design of Experiments. By the late 1970s Taguchi had earned widespread recognition in Japan, having won the Deming Application Prize in 1960 and Deming awards for literature in 1951, 1953, and 1984.

In 1980 Taguchi was invited to the United States and arranged to revisit AT&T Bell Laboratories. Despite communication problems, successful experiments were run that established Taguchi Methods within Bell Laboratories. Following his 1980 visit to the United States, more and more American manufacturers implemented Taguchi's methodology. Despite an adverse reaction to the Taguchi Methods among a few American statisticians, possibly because of how they were being marketed, major U.S. companies became involved in Taguchi's methodology.

In 1982 Taguchi became an advisor at the Japanese Standards Association. He has received numerous recognitions for his contributions to industries worldwide:

  • The Deming Prize on three separate occasions for his contributions to the field of quality engineering

  • The Willard F. Rockwell Medal for combining engineering and statistical methods to achieve rapid improvements in cost and quality by optimizing product design and manufacturing processes

  • The Shewhart Medal from the American Society for Quality Control

  • The Blue Ribbon Award from the Emperor of Japan in 1990 for his contribution to industry

  • Honorary membership in the American Society for Quality Control

  • Induction into the Automotive Hall of Fame and the World Level of the Hall of Fame for Engineering, Science, and Technology


Sidebar 2.2: Quality Engineering Methodology at a Glance[8], [9]

Taguchi methodology is concerned with optimizing product and process at the design and R&D stages before manufacturing. It is a quality methodology applied at the early stages of product or process development, rather than focusing on achieving quality through inspection. This is an efficient technique involving design tests before entering the manufacturing, fabrication, or assembly phase. Quality thus becomes a function of sound design rather than of test and inspection, however rigorous. Taguchi's approach can also be used as a troubleshooting methodology to address production and process problems in manufacturing and increasingly in other industries, including software.

In contrast to the Western concept of quality, Taguchi methodology perceives quality in terms of quality loss rather than quality itself. Quality loss is defined as "loss imparted by the product to society from the time the product is shipped." This loss includes loss to the company through costs of reworking or scrapping, maintenance costs, and downtime due to equipment failure and warranty claims. It also includes costs to the customer through poor product performance and reliability, leading to further losses to the manufacturer as its market share falls. Taking a target value for the quality characteristic under consideration as the best possible value of this characteristic, Taguchi associates a simple quadratic loss function with deviations from this target. The quality loss function shows that a reduction in variability about the target leads to a decrease in loss and a subsequent increase in quality.

With this concept of quality loss, a loss will occur even when the product is within the allowed specification limits but is minimal when the product is on target. In real life, users do not care for product specificationsdo they? All they want is for the product to perform even when the voltage is slow, the road is slippery, or the operator at the terminal is new on the job. The product must perform on target under varied user conditions. In other words, design for variations in customer use. It is in the producer's interest to attain product or process performance as close to target as economically possible. The loss function may be used to evaluate design decisions on a financial basis. It helps you decide whether additional costs in improving tolerance and production improvement are justified and whether they will actually prove worthwhile in the marketplace.

Taguchi Methods can be applied "offline" in design or "online" in production.

Taguchi breaks offline quality control into three stages:

1.

System Design involves creating a design concept by using brainstorming, research, and other techniques. System design as a whole also includes other tools, techniques, and methodologies, particularly Analytic Hierarchy Process (AHP), Quality Function Deployment (QFD), Theory of Inventive Problem Solving (TRIZ), and Failure Modes and Effects Analysis (FMEA) (see Chapters 8, 11, 12, and 13, respectively).

2.

Parameter Design is at the heart of Taguchi Methods. This is where the Japanese have traditionally excelled and have achieved high quality levels without an increase in costsa major source of their competitive advantage. The nominal design features or process factor levels selected are tested. The combination of product parameter levels or process operating levels that are least sensitive to changes in environmental conditions and other uncontrollable (noise) factors are determined (see Chapters 16 and 17).

3.

Tolerance Design applies to product or process design if the variation reduction from parameter design is insufficient. It tightens the tolerance further on factors shown to have a large impact on variation. Using the loss function, more money is spent only if necessary. You may tighten tolerance or buy better materials or equipment if necessary, which emphasizes the Japanese philosophy of invest last, not first, as has been the practice in the West.


Sidebar 2.3: Taguchi on Taguchi Methods[10]

  • Quality losses result mainly from product failure after sale; product "robustness" is more a function of product design than of online control, however stringent, of manufacturing processes.

  • Design products not to fail in the field; you will simultaneously reduce defectives in the factory.

  • You gain virtually nothing in shipping product that just barely satisfies the corporate standard over a product that just fails. Get on target; don't just stay in-spec.

  • Quality Engineering (Taguchi Methods) is a technology to forecast and prevent quality problems at the early stages of product development and product design, including the troubles associated with a product's function, pollution, and other costs that occur downstream in manufacturing and in the market place.

  • Do not use customer-based quality measures (such as fraction of defects or reliability) as the upstream measure of quality in R&D. Instead, use the dynamic SN ratio as the performance index to evaluate the robustness of a product's function.

  • Robust products deliver a strong "signal" regardless of external "noise" and with a minimum of internal "noise." Any strengthening of a design, that is, any marked increase in signal-to-noise ratios of component parts, will simultaneously improve the robustness of the product as a whole.

  • Work consistently to achieve designs that can be produced consistently; demand consistently from the factory.

  • Catastrophic stack-up is more likely from scattered variation within specifications than from consistent deviation outside. Where deviation from target is consistent, adjustment to the target is possible.

  • Ambient conditions in the factory are rarely as damaging as variations in customer use.


Taguchi Methods have been hailed as being among the major engineering accomplishments of the 20th century. Although the statistical techniques employed by Taguchi have their origins in experimental design practices developed by the English statistician Sir Ronald Fisher, their philosophical underpinnings are unmistakably Japanese. Taguchi Methods and other Japanese quality management systems such as Kaizen (continuous improvement), Kanban (Just in Time), Total Quality Control, and Lean Manufacturing all were inspired by the teachings of American quality guru W. Edwards Deming, as propounded in his 14 Points for Management, Seven Deadly Diseases, and Obstacles to Quality Products. (Richard Zultner's adaptations of Deming's principles to software development are presented in Chapter 5.) Taguchi Methods and other quality management systems provided the foundations for Japan's remarkable rise as an industrial power in the decades after the World War II.

Deming's impact on Taguchi's work can hardly be overestimated. Like other Japanese quality gurus, Taguchi was deeply influenced by Deming. To understand Taguchi's particular approach, it is important to examine its context and roots. Taguchi Methods and the modern Japanese quality systems all had their beginnings in the aftermath of World War II. Deming, often referred to as the father of the modern quality movement, first visited Japan in 1946. He continued to work with the Japanese government and industries during the next decades and trained thousands of Japanese managers and engineers. The Japanese managers were interested in contemporary American management principles. However, Deming offered them something totally innovative that he believed would help transform Japan into a prosperous society and rebuild it as a major industrial power.

The essence of Deming's management principles are well known (see Sidebar 2.4), although they are not as widely assimilated outside Japan. These principles include voice of the customer, reduction of variation, use of statistical measures, winning the confidence and respect of coworkers, and continual improvement in processes as well as in products and services. Deming's approach was enthusiastically studied and applied in Japan. It has had a profound impact on the Japanese industries. In 1951, the Japanese Union of Scientists and Engineers (JUSE) honored Deming by naming its prestigious quality award the Deming Prize. In the United States, however, Deming's theories were largely ignored for almost 30 years. It is widely believed that this might have led to loss of competitiveness in a score of American industries such as automobiles and consumer electronics, where Japanese corporations have made huge strides.

Sidebar 2.4: The Essence of Deming's 14 Points[11]

1.

Create constancy of purpose toward improving products and services. The aim is to become competitive, to stay in business, and to provide jobs.

2.

Management must awaken to the challenge of quality, must learn their responsibilities, and must take on leadership for change.

3.

Cease dependence on inspection to achieve quality. Eliminate the need for inspection on a mass basis by building quality into the product in the first place.

4.

End the practice of awarding business on the basis of price. Instead, minimize total cost. Move toward a single supplier for any one item, creating a long-term relationship of loyalty and trust.

5.

Improve constantly and forever the system of production and service, to improve quality and productivity and thus constantly decrease costs.

6.

Institute training on the job. If people are inadequately trained, they will not all work the same way, and this will introduce variation.

7.

Institute leadership. Deming makes a distinction between leadership and mere supervision. The latter is quota- and target-based. The aim of supervision should be to help people, machines, and gadgets do a better job. Supervision of management is in need of overhaul, as well as supervision of production workers.

8.

Drive out fear so that everyone may work effectively for the company.

9.

Break down barriers between departments. People in research, design, sales, and production must work as a team to foresee problems of production and problems with usage that may be encountered with the product or service.

10.

Eliminate slogans, exhortations, and targets for the workforce that ask for zero defects and new levels of productivity. Such exhortations only create adversarial relationships. Most causes of low quality and low productivity belong to the system and thus are beyond the workforce's control.

11a.

Eliminate work standards (quotas) on the factory floor. Substitute leadership.

11b.

Eliminate management by objective. Eliminate management by numbers (numerical goals). Substitute leadership.

12a.

Remove barriers that rob the hourly workers of their right to pride of workmanship. The responsibility of supervisors must be changed from sheer numbers to quality.

12b.

Remove barriers that rob people in management and engineering of their right to pride of workmanship. This means, among other things, abolishing the annual merit rating and managing by objective.

13.

Institute a vigorous program of education and self-improvement.

14.

Put everybody in the company to work to accomplish the transformation. The transformation is everyone's job.



Much of Taguchi's work was inspired by Deming's 14 Points for Managementin particular the dictum Cease dependence on inspection to achieve quality. Taguchi took a step back in the product development process from emphasis on inspection to R&D, design, and engineering. He emphasized the value of a product's performing consistently and on target, rather than being merely within specifications. As illustrated in Figures 2.3, 2.4, and 2.5, this is achieved in two steps. First, you reduce the variability, and then you adjust an appropriate design factor to achieve the performance as close as possible to the target customer requirement, taking into account cost, design, and other constraints.

Figure 2.3. Performance Distribution Within Specification Limits but Inconsistent and Off Target


Figure 2.4. Performance Distribution Consistent but Off Target


Figure 2.5. Performance Distribution Consistent and Around Target


Furthermore, Taguchi underscored the merit of making design robust against variability in both production and user environments. It will be useful to introduce here the major imperatives of the Taguchi quality philosophy:

  1. Continual quality improvements and cost reductions are necessary for business survival.

  2. An important measurement of product quality is the total loss generated by that product to societythe quality loss function.

  3. Change the preproduction experimental procedure from varying one factor at a time to varying many factors simultaneously. (This is known as Statistical Design of Experiments [SDE] or simply Design of Experiments [DOE].) Thus, quality can be built into the product and the process.

  4. The customer's loss due to poor quality is approximately proportional to the square of the deviation of the performance characteristic from its target or nominal value. Taguchi changes the objectives of the experiments and the definition of quality from achieving conformance to specifications to achieving the target and minimizing the variability.

  5. Product (or service) performance variation can be reduced by examining the nonlinear effects of control factors (parameters) on its performance characteristics. Any deviation from a target leads to poor quality.

One of the major Taguchi objectives is to improve product and process design by identifying controllable factors and their settings, which minimize variation from a target response. By setting controllable factors to their optimal levels, a product can be made more robust to changes in operation, usage, and environmental conditions. A major element of Taguchi Methods is removing the bad effect of the cause rather than the cause of a bad effect. Thus, you obtain a higher-quality product at the lowest possible cost. This strategy of neutralizing just the effect rather than the cause is the smart thing to do, because it may be easier as well as more cost-effective and time-saving. The Taguchi Methods have two key design objectives:

  • Reduce and minimize product and process variability and achieve the target economically.

  • Ensure that the product robustness measured at the design and prototype stages is maintained downstream in the manufacturing and user environments.




Design for Trustworthy Software. Tools, Techniques, and Methodology of Developing Robust Software
Design for Trustworthy Software: Tools, Techniques, and Methodology of Developing Robust Software
ISBN: 0131872508
EAN: 2147483647
Year: 2006
Pages: 394

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net