[SysOp note: no graphics or figures were included in the electronic version of this article. Tom Glenn, TQM BBS.] ----------------------------------------------------Business Index & ASAP------ AUTHOR(s): Kano, Noriaki TITLE(s): A perspective on quality activities in American firms. illustration chart Summary: Dr. Kano reviews the progress of the American quality movement as seen through his experiences with American companies over the last fifteen years. He identifies the movement's strengths and weaknesses and compares its development to that of the Japanese quality movement. In particular, he elaborates the appeal of management by policy (hoshin kanri) to Americans and describes the specific characteristics of its development in American firms. Overall, although progress may be described as slow and uneven, he finds considerable cause for optimism about the progress of selected American firms and industries. (Reprinted by permission of the publisher.) California Management Review p12(20) Spring 1993 v35 n3 DESCRIPTORS: Corporations, American_Quality control Total quality management_Technique FILM NUMBER: 71Y0290 Since 1963, when I was an undergraduate being supervised by Professor Kaoru Ishikawa at the University of Tokyo, I have been a student of quality management and engineering. Over the years, I have done research and taught in the field, widely disseminating its findings. I visited the U.S. for the first time in 1977. In over thirty subsequent visits, I have engaged in plant visits, seminars, conferences, and presentations; counseled nearly twenty firms; and taught and lectured at several universities. This article draws upon these experiences in evaluating recent quality trends in the United States. My Personal View On TQM In order to evaluate TQM in the U.S., I need to clarify how I arrive at my judgements. The following three sections lay out my analytical framework.(1) Generalizing an Individual Observation--It is reasonable to ask how an observer synthesizes a specific observation from Workplace A, Factory B, or Company C into a view that allows generalization from the specific area to the whole unit. I adopted the following approach for generalizing my observations. I would explain to the managers of the factory what I observed in a workplace and what generalizations I derived from these observations. Then I would ask them whether each of these views was applicable only to the workplace I visited or whether it applied to the factory as a whole. If the views coincided, then I would conclude that they were characteristic not only of specific workplaces but also of the entire factory. For example, in one case I observed that in the final inspection area, all the inspections were properly conducted, with the detected defectives being properly reworked. Then I visited the production workplace where one of the defectives had been produced. While the workplace supervisor had been informed of the defectives, I observed that the supervisor provided little more than a warning to the worker who had assembled the defective component. Moreover, he failed to analyze why this defective had been produced. As a result, the worker had difficulty knowing what preventive action to take in the future. I generalized these observations into a view, noting that the final inspection had been properly conducted, with the detected defectives being properly reworked, but the feedback was not properly analyzed to prevent future defects. If the managers agreed with my view that this was a common problem, then it could be accepted as a generalized pattern for the factory. I would then sum up the views of the factories that I visited in a particular area, such as a region or country, and would incorporate their common elements into a hypothesis, a tentative general view, for the area. I would then ask the local quality specialists whether the hypothesis was generalizable to the factories in the area or applied only to the specific factories visited. If they agreed with me about the common elements, the generalization would be considered valid for that area. The views selected did not necessarily encompass every feature of the area. It can be said, however, that these features were at least partially represented in each portion. The degree to which this is true can be determined by local experts with a thorough knowledge of the factories in the area. The House of TQM: Structure of TQM(2)--I use "The House of TQM," which shows the structure of TQM, as the touchstone for identifying the evolution of TQM's implementation in the United States. The portion from the floor up to the roof is TQM, where the floor symbolizes "motivational approach" and the roof shows "customer satisfaction/quality assurance"--which is the purpose of TQM. The base is the "intrinsic technology," which refers to the driving technology specific to an industry. For example, electrical engineering is the technology intrinsic to the electric industry; chemical engineering is intrinsic to the chemical industry. Intrinsic technology provides the necessary foundation upon which TQM is built. Given the existence of intrinsic technology, it is still necessary to carry out "sweating work" (hard work such as promoting standardization, educating and training, and collecting and analyzing data) to obtain good quality. The problem is this: how to create the conditions that will impel management and employees to take up such sweating work. This "motivational approach" is shown as the floor in Figure 1. "Concepts," one of the pillars of the house, shows how to proceed from a particular perspective when a given intrinsic technology and motivation already exist. "Concepts" consists of both a theory of quality (such as "quality is the satisfaction of the customer" or "the next processes are our customers") and a theory of management (such as the "PDCA Cycle" !Plan, Do, Check, Act^, "build quality into the processes," or "management by facts"). Figure 1 shows this as the column on the left. When actual activities based on these concepts begin, some "techniques" (the second pillar) for collecting and analyzing data become necessary. The seven QC tools and "The QC Story" procedure are typical techniques for this purpose, and statistical methods can also be applied here. At this point, some methods for effectively and efficiently promoting all of these activities within the organization become necessary. Management by Policy, Daily Management, and QC Circles are methods that can be called "vehicles" (the third pillar) since they quicken and facilitate promotion. Quality Sweating Theory--The process of introducing TQM can be explained by the "Quality Sweating Theory."(3) It states that TQM is an effective tool for improving quality, but its success depends on making many employees sweat. The quality sweating theory, encompasses two alternative approaches: CLSQ (Crisis consciousness and Leadership make people Sweat for Quality) and VLSQ (Vision and Leadership encourage people to Sweat for Quality). Change of Quality Activities in the U.S. over the Last Decade(4) In 1980, NBC aired a T.V. program, "If Japan can, why can't we?" It might be said that this program symbolized the historical shift of the American quality movement. Around 1980, quality circle activities began booming in the United States. By the middle of the 1980s, however, the quality movement had come to emphasize the necessity of management commitment, stressing the importance of top management's leadership. Subsequently, many companies tried to promote company-wide quality activities by motivating executives and managers. This contrasts with past American practice where quality professionals played the primary role in quality activities. In recent years, this trend has been increasing. Consequently, some companies have successfully introduced and implemented Japanese TQC. A nation-wide quality improvement campaign has been spearheaded by the Malcolm Baldrige National Quality Award (MBNQA). "TQM" is called "TQC" in Japan although its activities can be better explained by TQM (Total Quality Management) than by TQC (Total Quality Control) because both "management" and "control" are foreign terms for the Japanese, who understand them as synonymous.(5) Hence, in this article, the term TQM is used in order to avoid confusion except when specifically discussing Japan, in which case the term TQC will be used. Where They Were 10 Years Ago--On my desk, I have my 1982 lecture notes on "Overseas Quality Activities"(6) as well as an article of mine based on an interview with Dr. J.M. Juran at his New York office during July 1979.(7) They show the state of quality activities in the U.S. during that latter half of the 1970s and the beginning of the 1980s. My views on that period are as follows: * The education of top management is less focused on quality. * Some proponents of the quality circle movement simplistically believe that quality circles can solve all the quality problems of a company. This characteristic tends to make management neglect the necessity of committing itself to quality. * Cross-functional communication is so weak that it is difficult to solve quality problems, which demand cooperation among departments. * The cost of quality is too often over-emphasized. * Strong resistance to the introduction of quality improvement activities is manifested by the "NIH" syndrome (resistance to that which was "Not Invented Here") even though these quality improvement activities seem to be effective in solving quality problems. * The "AQL" (Acceptable Quality Level) practice of parts procurement as well as a short-term view for "achieving best quality at lowest cost" become obstacles to improving quality. Instead, managers need to make purchase decisions based on a long-term approach to continuous improvement. Changes over the Last Decade: Two Episodes--In 1977, I visited the United States for the first time, staying for a month. In the production shop of a certain factory well-known for its good quality management, I found a percent-defective graph on the bulletin board. As I hadn't seen such a graph in any other factory which I had visited during the trip, I had a closer look at it. I found that the last data point plotted was in July 1976, eight months before my visit to the factory. It was not until 1984, when I visited an American factory of a multinational corporation, that I felt things had begun to change in the United States. To my surprise, a considerable number of quality slogans and quality indicator graphs were displayed throughout the factory. One of the graphs carried data points for a control indicator at a certain process plotted hourly; the last data point was plotted at 11:00 of the day I was visiting. (I happened to see that graph around 11:30.) Several other factories that I visited later also had many quality indicator graphs, including control charts. They were practicing "SPC" (Statistical Process Control), and data points were accurately plotted. Of course, I am not going to say that simply plotting data on graphs makes quality better, but this observation does illustrate a change in quality consciousness in the U.S. by the mid-1980s. The Impact of the Malcolm Baldrige National Quality Award (MBNQA)--It could be said that the MBNQA precipitated the broad-based promotion of quality improvement within American companies. More importantly, it presented national quality leaders with opportunities to discuss the direction of the American quality activity systems through the formulation or revision of the guidelines for the MBNQA. They could do this during both the intensive training course for examiners and during the examiners' site visits as they audit the finalists. The successive revisions of the guidelines illustrate this development. Through this process, I believe that the institutional consensus among the experts is deepening and the American approach to quality is solidifying. Before the MBNQA was established, the consensus for nation-wide quality activities had neither been very strong nor well-organized, even though there were many excellent quality experts. Change In The Quality System--In any country or organization, total quality activity can begin only if top management is conscious of the critical need for company-wide commitment to quality and its own responsibility for introducing such activity. As testimony to the leadership that top management is exercising today in some American companies, in lectures at quality-related conventions, top managers have begun to include specific examples of their own practices, instead of just sprinkling their speeches with such phrases as "quality first," "customer satisfaction," or "do it right the first time." Also, the top managers of leading U.S. companies were influential in establishing the MBNQA. Furthermore, although formerly only a few companies had Vice Presidents in charge of quality, in recent years their number has increased. In some companies, Senior Vice Presidents and Executive Vice Presidents in charge of quality have appeared. Finally, top managers of many American companies are visiting Japan to study TQM, a reflection of their positive attitude toward the study of quality. In summary, over the last decade in the U.S., what has changed most is the leadership of top management. Whereas before, quality activities were only implemented by the quality department, top management is now initiating corporate-wide quality activities. As mentioned, a criterion that formerly dominated quality efforts in the U.S. was quality cost. Today, this criterion is increasingly less relevant and is being superseded by "customer satisfaction." In the audit for the MBNQA, this has become evident. Although "customer satisfaction" is only one of many audit categories, it is given the largest weight of all: 30%. In U.S. companies that have introduced quality activities in accordance with Japanese methods, their implementation differs slightly from the original Japanese version because of their different backgrounds. For example, there is an emphasis on team activities in the U.S. version, which can be seen as an effort to break through departmental barriers resulting from a strong chain of command. The term "cross-functional," which is frequently heard in the U.S., reflects the consciousness of barriers between departments. General Trends of Quality Activities in the United States(8) Based on the House of TQM, the current status of quality activities in the U.S. can be analyzed using the categories of motivation, concepts, techniques, and promotion. Motivation--As Dr. Juran said in 1979, "Top management believes that the situation is not so bad yet, and that is a problem of American firms."(9) After the economic recession of the early 1980s, however, America came to face growing trade and financial deficits. This seems to have given many people, especially top management in some leading companies, a crisis consciousness. As a result, top management has begun to play an important role in leading the quality movement. This explains why American firms have become so enthusiastic about quality in recent years. Concepts--There are four criteria for implementing TQM: * Customer satisfaction; putting quality first. * The PDCA cycle; Process-oriented production; Doing it right the first time. Although the PDCA Cycle is well-known, many people have difficulty applying it to their jobs. Therefore, many facilitators try to implement this concept via management by policy and daily management. Often, when they do this, employees ask managers to improve their own way of working. Some resistant employees play with the acronym PDCA, turning it into "Please don't change anything]" * Emphasis on the use of data. Along with the diffusion of the SPC method, the concept of management based on data is widely used. It is easy just to collect data and apply the method mechanically. In each case, however, what physical phenomena the data represent and what the analytical results actually mean need to be individually investigated and discussed, to determine whether the method used adequately reflects the significance of the data. In general, people are not taught to recognize the tangible phenomena based on the data. In most areas, managers and workers alike tend to do their jobs depending on their gut reactions, not on the data. Consider the following example. On my first trip to Florida Power and Light, we discussed the reduction of service interruptions. The FPL people made a presentation with a Pareto diagram of interruptions by causes showing that lightning was the largest cause. To my question of why the groundings or arresters had not prevented the interruptions, they answered that Florida had the most frequent and severe lightning attacks in the United States and that groundings or arresters would not have prevented service interruptions produced by strong lightning. I again asked, "Please show me the data which explain that intense lightning caused interruptions so severe that groundings or arresters could not prevent them." They could not show me any data, but promised to collect data by my next visit. Nearly half a year later when I visited again, they had collected the data and found that interruptions occurred even when they did not have such strong lightning. In addition, they found a certain percentage of poles with absent or insufficient groundings, which they had not recognized until they collected the data. * Employees' commitment; management is not a monopoly of managers. Techniques--The most popular techniques are the Pareto diagram, cause & effect diagram, and the histogram which are among the "Seven QC Tools." This is the same phenomenon which I observed in Japan. Some American companies are also starting to show an interest in applying the "QC Story," which is a standard procedure for problem-solving developed in Japan. Due to the popularity of "SPC seminars," production lines are often flooded with control charts. In many cases, however, I observed that actions had not always been taken when the charts showed out-of-control situations. When I asked supervisors how they examined such facts and what actions they had taken, in most cases they could not explain their specific measures. Nevertheless, I have heard from the quality facilitators of the companies that have introduced control charts in their workplaces that their quality has been rapidly improved by the introduction of SPC. One explanation for this is that the very process of collecting data and plotting points on the charts causes foremen and workers to pay more attention to quality. Thus, even poorly used control charts can contribute to the process of improving quality by making workers more conscious of quality. In addition to control charts, other statistical methods such as analysis of variance and multiple regression analysis have been taught to engineers and managers. I have not, however, found many successful applications of these techniques in U.S. firms, yet. In the factories I observed, a remarkable number of problems were solved by combining intrinsic technology with simple methods such as the Seven QC Tools. In the Toyota Motor Company, which is well-known for its enthusiastic devotion to quality, when management discusses how much quality has been improved and how much they should stress further improvement, it is often said, "People make efforts to squeeze a dry towel drier." This towel metaphor means that in its early stages, a company with a dripping towel gets rid of a lot of water with just a little squeeze, without any tools. As the company evolves, a wringer might be needed. In an advanced stage, an electric vacuum drier is required. If asked how much the U.S. has improved its quality so far, I explain that at present the U.S. is using a wringer, such as the Seven QC Tools, to bring about effective results. I assume that more sophisticated methods, on the level of the electric vacuum drier--such as analysis of variance or multiple regression analysis--shall be required for U.S. firms to progress further. The Taguchi Method and "QD" (Quality Deployment, sometimes called "QFD," Quality Function Deployment) also seem to have become somewhat popular. People are also beginning to show an interest in the "Seven New Management Tools for QC." Some people seem to find somewhat flamboyant ways to demonstrate certain tools, rather than using them to solve problems. This is a common phenomenon at the early stage of TQM's introduction into any country. As TQM develops further in the U.S., this phenomenon should disappear. It is nothing to worry about. Promotional Vehicles Promotional Organization--Who shall ultimately be responsible for promoting company-wide quality? The answer to that question is the key factor to successful implementation, especially in the United States. In companies which promote company-wide quality activities, I find the rank of the executive responsible for this job differs by company. In one company, the Executive Vice President is in charge; in another it is the Vice President or Director. Those companies which elevate someone from personnel to become the responsible Director seem to have a great deal of difficulty. The new Director faces such problems as how to persuade high-level line managers such as the Vice President of Manufacturing or the Division General Manager to take necessary action. I recall the late Dr. Kaoru Ishikawa mentioning that he used to think that the staff function remains strong in Western companies. But, in reality, the situation is the same, as in Japanese companies, where the line function is stronger than the staff function, so that inter-departmental barriers are very thick. When it comes to promoting quality in implementation units--such as a division, factory, and branch office--successful results depend on whether or not their heads, general staff, and managers are actively leading the promotion. On this point, I found no differences between American and Japanese companies. Education And Training--In classrooms and workshops, American companies promote very extensive, and sometimes very intensive, training programs in quality management philosophies, management tools, and statistical methods. It seems to me, however, that they have difficulty in adapting these philosophies and techniques to their workplaces. This is partially due to the shortage of able and seasoned quality experts who can support and encourage both executives and managers to promote company-wide quality management by counseling them in actual situations as they occur. In the U.S., the exchange of ideas between companies has increased along with the number of companies which promote company-wide quality activities. Given the brief history of company-wide quality promotion, dependance on information from other U.S. companies suffices neither qualitatively nor quantitatively. I assume, however, that the exchange of quality information will improve in the future. Quality Circles--In the mid-1970s, Mr. Wayne Rieker at Lockheed Aircraft Co. first introduced quality circles to the United States. By the 1980s, the "IAQC" (International Association for Quality Circles), which was organized to promote the concept, was actively diffusing it. At the end of 1987, IAQC was reorganized and renamed "AQP" (Association for Quality & Participation). In addition to promoting quality circles, it holds numerous seminars on matters relating to quality and participatory management. Quality circle activities seemed to peak around the middle of the 1980s. This peak points to the end of one phase of the quality movement. Companies had initially defined quality problems in relation to workers' morale and workmanship, so they believed those problems could be solved by quality circles. In the early of the 1980s, many American management study teams came to Japan. Most of them were interested in nothing more than quality circles. Gradually, however, many of them learned that promoting quality circles alone was not sufficient to achieve quality. So they began, sometimes hesitatingly, to emphasize company-wide implementation of quality activities. These movements may also have been affected by the article "Quality Circles After the Fad," published in the Harvard Business Review by Prof. Edward Lawler III and associates.(10) The authors sharpened my own reflections on what happens to a company if it introduces quality circles without promoting company-wide quality activities. Although the quality circle boom has been declining, it is clear that group activities of workers, such as quality circles, are essential to a comprehensive implementation of quality. I believe that U.S. companies will need to reactivate quality circles as part of the future development of TQM. Team Activity--There are several different types of team activities in the United States. One type is similar to quality circle activity and involves a voluntary problem-solving group made up of workers in the same workplace. Upon completing one theme, the same team will work on the next one in an on-going activity. These teams are sometimes called functional teams since each group at the bottom of the organizational hierarchy shares one operational function. In a second type of team activity, a Japanese and American joint venture company in the U.S. defines the team as the unit to which a set of jobs is assigned. In typical American companies, a job is assigned to each worker. In this case, a team of 10 workers is given a set of jobs to be shared, flexibly, among team members. In the unionized sector, generally, a detailed job classification is developed jointly with the labor union, and it is not unusual to see over 100 different kinds of jobs classified in just one factory. To adopt an alternate type of team approach, job classifications need to be reduced to allow workers to do multiple jobs. Since this change is linked to the wage system, it necessitates negotiating with the unions. Because these teams are formed primarily to build quality into a product during the assembling process as specified rather than to improve it, they can be called "quality building-in process teams." Several workers in a plant where this team method has been adopted told me that their transformation from a single-skill to a multiple-skill workers has made them more interested in the job and more aware of quality. A third approach involves initiating a team comprising staff or line managers, usually formed on orders from management. It is often called a task team or a project team. Once the task is completed, the team is disbanded. Participation in such teams, just like participation in quality circles, depends in principle on the free will of individual employees. Such teams are often formed across different departments; the nature of the task given sometimes demands the creation of two other types of teams: a problem-solving team (a team to address a particular problem arising in the process of promoting quality activity) and a task-achieving team (a team formed to implement a particular project, such as developing a new product). Of the three types, the third one is the most standard team activity. It seems to me, however, that its content and nature varies from company to company, or even from department to department within the same company. In some cases, members of the team work on the task part-time while they do their regular job back in their own section. In other cases, members work full-time on the team activity. I also heard that the freedom to go beyond departmental barriers to do the activity required by the objective helped revitalize both staff and line managers who participated. This method is certainly useful in temporarily removing the barriers between departments. It could, however, produce new problems, such as a need for accumulated technical expertise and communication between different teams. Whatever the merits or demerits may be, this type of teamwork should not be overlooked when talking about quality activity in the U.S. today. Management By Policy (Hoshin Kanri)--Of the various elements of Japanese TQC, what most interests American management is customer satisfaction activity and management by policy. "MBO" (Management by Objective) has long held a central place for American managers. But, in actuality, in many companies it has become a mere skeleton, without much substance. In contrast, management by policy looks very fresh. It has the following features: * The Four Stages Management By Policy: policy setting, policy deployment, policy implementation, and evaluation and feedback. * Linkage of Annual Company Policy with Longer-Term Plans: Annual company policy is linked with long-term and mid-term plans. * Policy Elements: Each policy consists of an objective and the strategies to achieve it. * Setting Up Annual Company Policy: Annual company policy (frequently called presidential policy in Japan) is nominally a top-down approach. In reality, however, top management collects the opinions of people inside the company about chronic major problems and their wishes for their company's future and uses this information to formulate the annual plan. * Policy Deployment Conducted before the Fiscal Year Starts: Each annual company policy is incorporated into each department manager's annual policy. That way, implementing departmental policies helps to realize the company policy. Additionally, in each department, the manager's policies are coordinated with those of its related functions. Whether the deployment mentioned above is vertical (between a superior and his or her subordinates) or horizontal (cross-functional), communication within the organization should be fluid. * Policy Implementation: After company policy is deployed throughout each department, each function prepares the program and schedule needed to implement specific items to achieve those targets assigned to it. After the fiscal year begins, the items are implemented according to schedule. This process is conducted through the PDCA cycle, usually quarterly or monthly, but occasionally weekly or even daily. * Evaluation and Feedback: At the end of the fiscal year, the implementation results are evaluated for each function, at each department and company level. This is done whether or not the implementation items have been realized and the targets achieved, and whether or not department managers' policies and company policies have been achieved. If they haven't been achieved at a particular level, the causes should be investigated. The results should be reported to policy makers. That would allow some compensatory action and prevent the recurrence of similar failures in the coming year. This system was developed to implement the concept of PDCA cycle for company policies. Before introducing the system, a company's policies are often dream-like, with no systematic actions taken. In the 1960s, the "Integrated Management" system was created by Nippon Denso, Sumitomo Electric Industries, Nippon Kayaku, and Komatsu. It came to be named "Management by Policy" and was initiated in earnest by Bridgestone Co. around 1967. In some American companies, Management by Policy has already been introduced and implemented. I observe that American top executives are involved with it in the following ways: * Two Categories of Strategy: So far, U.S. executives have been involved in preparing various strategies. But when asked to implement them, they have not been necessarily successful. Gradually, these executives have begun to understand that excellent strategies are not always successful. Hence, in addition to a system for preparing strategies, they need to install a system for realizing them. In general, executives deal with two kinds of strategies. The first kind, which is effective immediately after decision making, involves personnel, budgeting, or merger and acquisition. The second kind is effective only with a company-wide effort, such as a quality strategy. Management by policy is the ideal vehicle for realizing the second kind of strategy. When it comes to the introduction and implementation of this second kind of strategy, however, U.S. companies have encountered a variety of difficulties inherent in management by policy, just as many Japanese companies did at a similar stage. * English Translation Problems:(11) American companies have encountered the problem of how to translate "Hoshin Kanri." The Japanese Society for Quality Control committee for the English translation of Japanese TQC terminology, which I chaired, decided to translate the Japanese term "Hoshin Kanri" into "Management by Policy." But this translation alone is insufficient. What makes an exact translation difficult is that a considerable semantic gap exists between the English word "policy" and the Japanese word "hoshin." In English, "policy" implies something rather permanent. The phrase "annual policy" makes English speakers uncomfortable, for it implies that a policy might change from year to year. A detailed explanation is also in order because of the common assumption that "hoshin" or "policy" implies an objective and its methods. Because of this confusion, some American companies such as Hewlett Packard have begun using the Japanese word "hoshin." They feel it is better to use the original word untranslated and free from the misleading connotation of "policy." Some of the other names selected by American companies which have introduced management by policy are "policy deployment" and "hoshin process." * Chase Too Many Rabbits: It is very common for a company introducing Management by Policy to set up too many policies to implement, and I often caution companies about this. This caution, however, is generally neglected or forgotten during the first year; and as a consequence, the results are inadequate. Then management pays attention to focusing its efforts from the second year on. * Inadequate Analysis of Data about the Current Status, A Preference for Pursuing Dreams: When top executives decide company policy, they are apt to chase their dream rather than analyze data about the current status. Dreams are necessary, but top executives should understand that it takes many years to realize them. I recommend that part of the introductory year's annual policy include overcoming weaknesses--reducing failure cost and market claims, for example--rather than enhancing strengths, such as further increasing market share. * Insufficient Cross-Functional Coordination Can Undermine Company Policy: One company, for example, set up a medium-term policy of developing a new product along with specific target for sales and profit. At the end of this period, however, the company achieved its sales target, but not its profit target. This was because of the unexpectedly high warranty costs incurred by the new product. The warranty problems resulted from poor design review and poor reliability which, in turn, arose from product failures under certain usage conditions, which had not been predicted. If service engineers had participated in the design review, the problems in all likelihood would have been minimized because these engineers would have recognized similar kinds of failures in the past. This product development policy was deployed under the auspices of the product development department, which had no contact with the service department. This is an example of poor policy deployment. * Development of a Project: I have noticed that in American organizations, policy is deployed down to lower levels and finally, in most cases, to project teams. At the implementation stage for each project, analysis is conducted to solve problems, countermeasures are developed, and the results are replicated at different departments or sections. This process seems appropriate for American culture. In some companies, in fact, the task teams execute these projects. In the case of Japanese companies, however, the team projects are deployed within the ordinary organization; the team members come from departments and sections whose chiefs have key roles in the promotion of the policy objective. * Presidential Diagnosis: Some companies have introduced presidential diagnosis--sometimes called "presidential audit" or "presidential review"--to ensure that top management's policy is deployed and implemented at each department, and to provide top management with an opportunity to learn what is happening on the first line of operations. In introducing presidential diagnosis, one company made the following argument: Top management's role is to make decisions on various kinds of important issues, including policy making. Implementation is mainly the job of managers who report the results to top management through the chain of command. Top management, therefore, macroscopically knows what is happening in the field. Under such a system, of what use is a presidential review? Would not top management's visit at each department or workplace create confusion in the chain of command? Once such an argument was overcome, however, and the process implemented, the company discovered the merits of the diagnosis. In particular, top management became aware of very different kinds of microscopic information. This company still continues to use the presidential diagnosis. In another company, a division general manager's review was implemented on a trial basis for about two years before the process became company-wide and was eventually upgraded to presidential diagnosis. * Few Companies Are Conducting a Combined Evaluation for Objectives and Methods of Implementation: In the evaluation stage, we need to check whether objectives are achieved as well as how they are realized. The evaluation results can be categorized into the following four cases: These cases can be compared to students who: Case A--fully attended the class and passed the exam; Case B--rarely attended the class but passed the exam; Case C--fully attended the class but failed the exam; Case D--rarely attended the class and then failed the exam. Student A and Student D are easily understood, and Student D may be successful in the next semester if he fully attends the class, so he is hopeful about next semester. Student B and Student C are not as easily understood. Student B is overjoyed that God blessed him, but he needs to be careful, for God will not always bless him. Student C may be hopeless next semester, even though he fully attends the class as he did last semester. Student C should analyze why he failed even though he fully attended the class. He needs to improve his approach to learning in order to pass the exam. Similarly, in evaluating Management by Policy at the end of the fiscal year, both its objectives and the methods for realizing objectives should be evaluated. The reasons for each individual case should be analyzed. In many cases, the problems come from the mode of policy deployment. The above combination of evaluating both the objective and its methods of implementation is an orthodox approach. Yet in the U.S., there are very few companies that conduct such an evaluation. In addition, some companies emphasize objectives; others, on the contrary, emphasize only methods of implementation (process improvements). Daily Management--This approach is less marked among American companies than is Management by Policy, but it is so important that it could be defined as fundamental management. It is encouraging to find a few companies in the U.S. that have introduced this system in addition to Management by Policy, for it strengthens the capabilities of the first-line managers and supervisors. Florida Power and Light calls this system "Quality in Daily Work" (QIDW). In this activity, these are the major tools: a control item or a performance indicator for a job, process standards (including process flow charts and forms/files), and control charts/control graphs. American companies seem to have the following strengths and weaknesses in this activity: * In regard to daily management, the U.S. should be proud of its extensive use of computer systems, although there is still room for improvement from the standpoint of their effective utilization. I have been impressed by the considerable number of computer terminals that have been installed and the reduction of paper documents in offices. In contrast to this, in the offices of Japanese companies, many documents are still piled up on desks. Japanese companies are over 10 years behind the U.S. when it comes to promoting office automation. * Process standards have been prepared by staff departments in headquarters in the United States. Therefore when something needs to be improved in the standards, the workplace manager cannot revise them by himself/herself. In Japan, many companies operate under a corporate standardization rule. It classifies standards into categories according to the responsible department or managerial level and authorizes each of the responsible departments or management levels to prepare and revise the category of standards they are responsible for. Workplace management can revise most operational standards on their own, and this helps management to rotate the PDCA cycle quickly. * In the U.S., the training in SPC or control charts is not being linked with the training for process standardization. SPC is taught by statisticians who have limited experience in workplace management. They base their training on the notion of population, which is just a hypothetical statistical model. What prerequisites are essential to apply this model to process control in the workplace? The answer is process standardization, which bridges the statistical model and the actual job process in the workplace. Without specific standards and their operation, we can neither identify the population nor use the information provided by control charts such as "under control" or "out of control." * It is encouraging to see that the judgement "out of control" is based on control charts. This shows that people have based their actions on "fact," rather than "intuition." In the process of finding the causes of "out of control," however, many people still depend on "intuition" over "data." In order to recover from this dependence, companies need executives and managers who earn the nickname "Mr. Show-Me-Data." * One of the popular activities observed in American companies is the implementation of "5 WHY's"; namely, if a problem occurs, employees are encouraged to ask "Why" five times, on the assumption that the answer to this question will unearth a root cause for the problem. Although this is a very good practice, in many workplaces, I found employees who ask the questions without fully understanding the nature of the problem. Therefore, I recommend asking "5 WHAT's" before asking "5 WHY's," repeating "What is our problem?" five times. Quality Activities in Each Stage of the New Product Development, Production, and Marketing Processes New Product Development--I was very curious about the process U.S. companies use to develop epoch-making new products. I found that they take a "genius approach," in which a company, when it discovers a very outstanding person, perhaps a genius, provides him/her with the necessary resources, conditions, and work environments. The company makes it easy for the person to do his/her best, without bringing into play age, wealth, education, or experience, as is popularly done in Japanese companies. In contrast to this, the method for new product development in Japanese companies which implement TQM could be termed the "systematic approach." We are frequently surprised with the outstanding originality at the core of American new products--the result of giving a genius free rein--but we sometimes encounter problems with the subsidiary parts of the product which are handled by ordinary employees. In contrast to this, Japanese products may have fewer problems as well as less originality, although Japanese companies have made great efforts to downplay this weakness. To sum up, America is strong in its genius approach, but poor in its systematic approach. Japan is just the opposite. Companies in both countries need to overcome their weakness and further develop their strength. Although some companies are already examining this issue, it might take a considerable number of years to achieve this state. (I assume it could be done very well if a Japanese-American development center were established somewhere in the middle of the Pacific Ocean.) Production--The discipline on the shop floor in American firms seems to have been improved considerably during the last decade. People work harder and housekeeping has been much improved. Quality circles, SPC, and team activities are becoming common tools for quality improvement. Some companies are eager to promote JIT as well. Marketing Share--The management of American companies characteristically puts a great deal of emphasis on "amount of profit," rather than on "turnover growth," "profit ratio to sales turnover," or "market share." Based on this concept, U.S. manufacturing firms have cut off the less profitable business lines or transferred their manufacturing base overseas. It might be said that this is one of the negative elements that caused the "hollowing out" of U.S. manufacturing firms. Recently, however, some companies have begun to become more aware of market share and have made comparative studies of their competitors' product quality. Conclusion With regard to the promotion of quality, there appear to be American companies that have been implementing quality activities enthusiastically and effectively during the last decade and are beginning to take further steps for their steady promotion. Although the pace of their efforts is sometimes slow, they are having successful results. At the beginning of 1993, as usual, various statistical reports about the past year were released by the Japanese press. Three of the articles caught my attention. One, dealing with semiconductor production, said that NEC, which had been in first place for the past seven years, was replaced by INTEL, whose sales had increased to five billion dollars (up 26 percent from the previous year). The second article, relating to total car sales, showed the Ford Taurus surpassing the Honda Accord in the U.S. as the best selling passenger car. The third article, using data from Dataquest, deals with personal computers and reports that, worldwide, American companies (IBM, Apple, and others) are regaining market share. These facts point to a trend: during the past ten years, American improvement efforts, including TQM, have gradually led to good results in some industries. We seem to be at the beginning of a new era in which international competition is based on quality. These facts are indeed impressive, and I want to keep my eye on how things develop after this. Because America is such a big country, it will take much more time until it gets macroeconomic results, such as contributing to the elimination of the trade deficit. In the case of Japan, it took about 20 years after the introduction of QC before the trade balance turned positive. As I have repeatedly said, similar problems arise in Japanese and American firms when they first implement TQM. I have often heard some American specialists insisting that it is difficult to implement TQM in the U.S. due to its culture. I disagree. A more important factor, I would suggest, is the degree of top management's motivation to be part of TQM, and the greatest enemy of TQM is top management's doubt. "Is there a need to do so much?" or "Do I have to do so much?" This shows that they still base their professional self-evaluation too much on traditional measures of their companies' standing. Quality will be improved if top management will assert its leadership to smoothly implement TQM corporate-wide--and from the long-term viewpoint. When I consider how American firms will develop TQM activities in the future, one influence is mass media. It tends to discuss the short-term effects of quality. In addition, American consumers' tend to be uncritical of poor quality products and services. Let me elaborate on both these points. One characteristic in the development of Japan's TQC is that Japanese mass media paid little attention to the TQC movement, both its merits and demerits, until around 1980. As a result, Japanese top management could concentrate on TQC from a long-term perspective without worrying about mass media commentary. A drawback is that it took so much time to diffuse TQC activities throughout the country. In the case of the U.S., at the beginning of the 1980s, American mass media began to pay attention to the quality movement, so information about it has been diffused rapidly. A considerable number of articles about it appeared in various media. There are some firms, however, that implement TQM simply because it is the fashion. So quality activities in such companies tend to be perfunctory or short-sighted. It might be said of these companies that they promote TQM for the sake of TQM. So it is not surprising that the media can find easy targets to criticize. A friend of mine, a leading person in the American quality world, once showed me a critical article that appeared in one of the leading newspapers and asked for my comment. The article criticized the absolute amount of funds the firm expended for its TQM implementation. I answered that my friend should calculate the average cost per employee per year as well as the rate per gross sales. Typically, such articles give little consideration to the company's size and time over which the budget was spent; rather, they take the total amount as a reference, even when it is budgeted over five years. This often shocks people into thinking, "How expensive]" According to my experience, it costs approximately 0.1-0.3% of gross sales, referring to total expenses, to run a department for education, training, and quality promotion. The cost can vary depending on how intensively a firm implements it or what related costs are included. So in the case of a mammoth enterprise whose gross sales amount to ten billion dollars, the total costs could be between ten and thirty million dollars a year. In a word, that's how much cost should be permitted when a firm implements quality activity to improve its product/service and to realize the use of scientific methods. It seems to me that American consumers are quite sensitive to the prices of products. On the other hand, they are charitable about quality unless a product becomes a product liability issue. People seem to feel that it is no use complaining because the country is so large and there is too little competition. For example, this past summer when I crossed the continent on a red-eye flight, my plane took off from San Francisco Airport one hour behind schedule. The delay was due to the late arrival of a pilot. As a result, I missed my connection at Chicago and wasted three hours. I did not notice any passengers complaining to the ground staff of the airline at the Chicago terminal. Since the delay was caused not by weather but by the airline's lack of attention, one might assume that the ground service staff in Chicago would have been waiting for us at the arrival gate, eager to offer service for in-transit passengers. On that occasion, however, I could not find any ground service staff at the gate. The airline's indifference led me to wonder if passively accepting the world as it is undermines the quality improvement of American products and services. In conclusion, there are many similarities between the development of quality activities in American TQM and Japanese TQC. Although some differences exist, they are not critical. Therefore it permits me to assert that the experiences and knowledge of both countries is to some extent interchangeable. References Acknowledgement: I am grateful to Prof. Robert E. Cole for helping to clarify the English language usage in this article as well as the content of various ideas and also to Ms. Noriko Kusaba for assisting me in preparing this article. 1. N. Kano, "Business Trip to Europe after an Interval of Six Years," Hinshitsu !Quality, Journal of the Japanese Society for Quality Control, in Japanese^, 12 (1982): 24-28. 2. N. Kano and K. Koura, "Development of Quality Control Seen Through Companies Awarded the Deming Prize," Rep. Stat. Appl. Res., JUSE, 37 (1990-91): 79-105. 3. N. Kano, "Quality Sweating Theory: Crisis Consciousness, Vision, and Leadership," Hinshitsu !Quality, Journal of the Japanese Society for Quality Control, in Japanese^, 19 (1989): 32-42. 4. Ibid. 5. N. Kano, "English Translation of the Technical Terms of TQC Implementation," Hinshitsu !Quality, Journal of the Japanese Society for Quality Control, in Japanese^, 22/3 (1992): 91-99 and 22/4 (1992): 87-93. 6. N. Kano, "A Comparative Study between the Western and Japanese QC," memorandum, the 207th JSA (Japanese Standards Association) COSCO Research Meeting !in Japanese^, 1982. 7. N. Kano, "The Situation Is Not So Bad Yet, and That Is a Problem of American Firms," Hinshitsu !Quality, Journal of the Japanese Society for Quality Control, in Japanese^, 10 (1980): 227-232 and 11 (1980): 87-93. 8. Kano (1989), op. cit. 9. Kano (1980), op. cit. 10. E. Lawler III and S.A. Mohrmann, "Quality Circles After the Fad," Harvard Business Review, 63 (January/February 1985): 65-71. 11. Kano (1992), op. cit.