QUALITY MANAGEMENT. Zsuzsanna Eszter Tóth Tamás Jónás. teaching material

Transcription

1 Budapest University of Technology and Economics Faculty of Economic and Social Sciences Institute of Business Department of Management and Corporate Economics Zsuzsanna Eszter Tóth Tamás Jónás QUALITY MANAGEMENT teaching material Budapest 014

2 Table of contents I. Introduction...5 I.1. Evolution of quality concepts...5 I.. Defining quality...5 I.3. Evolution of quality management I.4. Comparing schools of quality: Europe, Japan, USA I.5. Gurus of the quality movement II. Development of quality management systems... 1 II.1. Conformance to requirements... 1 II.. Good manufacturing practice GMP... 1 II.3. Military standards... 1 II.4. BS5750 and ISO II.5. HACCP... II.6. QS II.7. VDA II.8. ISO/TS II.9. AS II.10. ISO/IEC II.11. Six Sigma... 6 II.1. ISO II.13. EFQM... 8 II.14. Common Assessment Framework... 8 III. ISO 9000 standards III.1. Quality management system III.. ISO III.3. ISO 9000 family of standards III.4. Quality management principles III.5. ISO 9001 requirements - Quality management system (Clause 4) III.6. ISO 9001 requirements Management responsibility (Clause 5) III.7. ISO 9001 requirements Resource management (clause 6) III.8. ISO 9001 requirements Product realization (clause 7)... 4 III.9. ISO 9001 requirements Measurement, analysis and improvement (clause 8) IV. Total Quality Management... 51

3 IV.1. Introduction to TQM IV.. Applying quality concepts IV.3. Customer focus IV.4. Mechanisms for understanding customers... 7 IV.5. Managing key processes IV.6. Six steps to process improvements IV.7. Measuring performance IV.8. Leadership IV.9. Empowered workforce V. EFQM Excellence Model V.1. Fundamental concepts V.. Enablers V.3. Results V.4. Radar logic VI. Six Sigma VI.1. DMAIC Process VII. Quality management tools and methods VII.1. What causes defectives? VII.. Diagnosis of processes VII.3. How to obtain data... 1 VII.4. Check sheets VII.5. What are Pareto Diagrams? VII.6. How to make Pareto diagram? VII.7. Cause-and-effect diagram VII.8. Distributions and histograms VIII. Application of Statistical Process Control VIII.1. What is SPC? VIII.. Selecting the Appropriate SPC Method VIII.3. Overview of Control Charts VIII.4. Defining Control Charts IX. Capability Assessment IX.1. Short-term data vs. long-term data IX.. Capability Assessment for Quantitative Process Characteristics

4 X. Introduction to measurement system analysis mathematical basis X.1. Notation X.. Analysis of Variance (ANOVA) XI. Measurement Systems XI.1. Categories of Measurement Systems XI.. Analysis of Quantitative Measurement Systems XI.3. Analysis of Qualitative Measurement Systems (Attribute Gauge Study) XII. Appendix XII.1. The F Probability Distribution XIII. References

5 I. Introduction The word quality has many different definitions, ranging from the conventional to those that are more strategic. Conventional definitions of quality usually describe a quality item as one that wears well, is well constructed, and will last a long time. Still another definition conveys the image of excellence, first-rate, the best. However, managers competing in the fierce of the international marketplace are increasingly concerned with the strategic definition of quality: meeting the needs of the customers. I.1. Evolution of quality concepts David Garvin, in his book Managing Quality, describes five maor approaches to quality: 1. Transcendent: quality is understood only after exposure to a series of obects that develop its characteristics. For example, the quality of a particular artist only becomes apparent when a number of his or her works have been viewed. The idea here is that quality cannot be defined, and you recognize it only when you see it.. Product-based: quality is based on the presence or absence of a particular attribute. If an attribute is desirable, greater amounts of that attribute, under this definition, would label that product or service as one of higher quality. 3. Manufacturing-based: quality in manufacturing is defined as the conformance of a product or service to a set of predetermined requirements or specifications. Failure to meet these requirements is, by definition, a deviation, as such, represents a lack of quality. This approach assumes that the specification is a valid surrogate for a customer requirement and that, if met, it would satisfy the customer. 4. User-based: Quality lies in the eye of the beholder. The ability to satisfy the customers requirements, expectations or wants is the sole criterion by which quality will be determined. The ultimate aim of the organization is the complete satisfaction of the customer. 5. Value-based: quality under this definition consists of offering product or service to a customer with certain characteristics at an acceptable cost or price. This definition combines the idea of worth or value with the offering. I.. Defining quality The common element of the wide-ranging definitions used in business is that the quality of a product or service refers to the perception of the degree to which the product or service meets the customer's expectations. Quality has no specific meaning unless related to a specific function and/or obect. Quality is a perceptual, conditional and somewhat subective attribute. The business meanings of quality have developed over time. Various interpretations are given below: 1. American Society for Quality: "A combination of quantitative and qualitative perspectives for which each person has his or her own definition; examples of which 5

6 include, "Meeting the requirements and expectations in service or product that were committed to" and "Pursuit of optimal solutions contributing to confirmed successes, fulfilling accountabilities". In technical usage, quality can have two meanings: a. The characteristics of a product or service that bear on its ability to satisfy stated or implied needs; b. A product or service free of deficiencies.". Philip B. Crosby: "Conformance to requirements." Requirements may not fully represent customer expectations; Crosby treats this as a separate problem. 3. W. Edwards Deming: concentrating on "the efficient production of the quality that the market expects," and he linked quality and management: "Costs go down and productivity goes up as improvement of quality is accomplished by better management of design, engineering, testing and by improvement of processes." 4. Peter Drucker: "Quality in a product or service is not what the supplier puts in. It is what the customer gets out and is willing to pay for." 5. ISO 9000: "Degree to which a set of inherent characteristics fulfils requirements." The standard defines requirement as need or expectation. 6. Joseph M. Juran: "Fitness for use." Fitness is defined by the customer. 7. Noriaki Kano and others, present a two-dimensional model of quality: "must-be quality" and "attractive quality." The former is near to "fitness for use" and the latter is what the customer would love, but has not yet thought about. Supporters characterize this model more succinctly as: "Products and services that meet or exceed customers' expectations." 8. Six Sigma: "Number of defects per million opportunities." 9. Genichi Taguchi gives two definitions: "Uniformity around a target value." The idea is to lower the standard deviation in outcomes, and to keep the range of outcomes to a certain number of standard deviations, with rare exceptions. "The loss a product imposes on society after it is shipped." This definition of quality is based on a more comprehensive view of the production system. 1 1 In the followings we give some interesting examples how the readers of Quality Digest magazine define quality: The word "quality" represents the properties of products and/or services that are valued by the consumer. Reducing the variation around the target". The degree to which something meets or exceeds the expectations of its consumers. Definition of quality: "WOW" When the customer returns and the product doesn't. Quality is the expression of human excellence. Never having to say you're sorry. That we shall get the right product to the right place at the right time while exceeding our customers expectations. 6

7 Examples for dimensions of quality in different areas of society: Airlines: on-time, comfortable, low-cost service Health care: correct diagnosis, minimum wait time, lower cost, security Food services: good product, fast delivery, good environment Postal service: fast delivery, correct delivery, University: proper preparation for the future, on-time knowledge delivery Insurance: payoff on time, reasonable costs, Automotive: defect free IT: clearer, faster, cheaper service I..a. What are the difficulties in defining quality? You know quality when you see it You probably know of some things products, services, etc. that you feel are excellent, because they meet your needs, do what they are supposed to do, make your life easier, taste great, whatever. These things are identified by you as being very positive and if you stop to think about why they are positive, the word "quality" probably comes to mind. This view of quality lacks any concrete way of measuring the quality of something, because it is based on a specific person's udgment. People udge quality on many factors; most of these factors are specific, and many of those are probably being used without the person ever being aware of it. Quality is a function of brand Brand names can be a sign of quality, either as a result of personal experience with the brand or from udgments driven by advertising. Most people have faith in some brands or companies. A brand is a great way to set expectations, so for example you can go to any McDonald's restaurant in the world and get a Big Mac that's going to taste ust about exactly the same. The problem with this definition of quality is that you may have a hard time understanding ust what about the brand marks quality. If you don't love McDonald s Big Mac, you will be hard-pressed to explain ust what it is about them that makes Big Mac such a good sandwich. Quality means providing customer with innovative products or services characteristics/attributes and defects free which provide fitness for use. Quality is in the eyes of the beholder. And in a business environment, the beholder is always the customer or client. In other words, quality is whatever the customer says it is. A degree of excellence. Quality is a "system" which produces a product, service, information or delivery, on target with minimal variance which meet or exceeds the customers needs, now and in the future. Quality is meeting customer expectations. Quality is not something extraordinary. It is something ordinary extraordinarily well. Quality is not achieved by doing different things. It is achieved by doing things differently. 7

8 And if you disagree with the quality signaled by a specific brand, perhaps because of bad experiences with the product or brand, then that brand sure won't signal quality to you. Quality is a passing grade Perhaps the most common understanding of quality is that quality corresponds to a passing grade. If you've ever bought a piece of clothing that had one of those little paper tags saying that this piece had passed inspector something-or-other, you know that your clothes were tested and received a passing grade. Your experience with class grades is similar. You spend a semester studying a topic, attending lectures, doing your reading, turning in papers, and taking tests, and your understanding of the topic is evaluated by the teacher. Your teacher assigns you a grade, and this grade is used to help determine whether you pass the class and move ahead. This definition has several problems, though, when it comes to understanding the meaning of quality. First, you can't always be certain about what exactly the test and the passing grade actually measure. With the class example, the teacher could be measuring your comprehension of the class's topic or your ability to master the class itself; haven't you ever taken a class where somebody who doesn't understand the material passes because they meet the minimum work requirement? Second, the teacher is often using his/her subective views to set the milestones measured by the tests, so the tests are not necessarily obective evaluations. Third, this view doesn't help you understand the differences in grades that pass. Say you have two students who pass a class, one of whom gets a grade of "5", the other gets a "3". Both pass the class, but one apparently did better than the other. If quality is passing this class, is there any difference between the performance and understanding of the satisfactory (3) student and that of the excellent (5) student? From a testing point of view, passing a test would seem to imply quality, but you need to know a lot more about the big picture to get any value from a passing grade. Quality is perfection Perfection is good; actually, perfection is better than good, it's the best. If something is the best, then it must be overflowing with quality. But who decides that something is perfect, and who decides what perfection means? Who or what is it perfect for? The problem with this definition is that it tells you where you want to go, but not how to get there. Quality is the absence of problems Something doesn't cause me any problems, and I have no complaints about it, it doesn't get in the way of work or play, it's always held up, never breaks, never dies, it's "old reliable" that's a sure sign of quality. Saying that quality is the absence of problems doesn't go far enough, because it doesn't address the big picture around problems: problems for whom? Take something like a software program you can expect different people to be different kinds of users, with different needs that involve using the software with different goals in mind, and probably using different tools within the software. Say the software has a function that most people will never need to use, like the old "convert the file to Sanskrit" command; if that function does not work, but most users won't come across that failure, does the problem nonetheless exist? 8

9 Quality is zero defect code Quality is code that has no bugs that's a great goal to have, but it's usually impractical. Quality is acceptable performance You have something that overall does what it's supposed to do, and failures are within acceptable limits. The maority of students pass the class, the spread of grades looks like it's supposed to. It may not be perfect, but it's certainly good enough. Quality is meeting goals This definition also describes quality as reflecting what something does. You set goals for something, say a web site, and if the web site meets these goals then it has quality. But is there a scale, so that there is an amount or level of quality that is a function of how much or how well the goals are met? Quality is meeting requirements If you define quality as meeting requirements, then you have specific indicators of quality. If the requirements are testable, then you can test the success of meeting the requirements. And you can verify the quality repeatedly, by testing at intervals. You can say that whatever you're testing is good, and that it was good the last time you ran the tests, and the time before that, etc. You can measure quality, and measure it over time. A widely used definition of quality is: "The totality of features and characteristics of a product or service that bear upon its ability to satisfy stated or implied needs" This highlights that the service/product must satisfy a given need, and that is the need of the customer. Quality is providing a product or service that is "fit for the purpose". In many areas the popular choice of purchase is not the cheapest but it is chosen because its quality and reliability are perceived by the customer as being the best value for money. An alternative definition could be: "Quality is the sum of all the factors that enable ownership satisfaction and bring customers back to buy a product or service again and again." Quality could therefore include: a) Knowing the customer's needs: If the customer's needs are not identified they can only be achieved by accident, and any amount of trial and error causes delay and additional cost. b) Designing to meet them: If the customer's needs are to be satisfied, they must be documented to clarify the extent of the contract and, if appropriate, agreed with the customer. c) Reliable bought-in equipment and materials: The equipment and materials must be fit for the intended purpose if the customer is to be satisfied and the company is to avoid rectification and warranty costs. The choice of suppliers is an important factor in avoiding these costs. d) Clear and precise instructions: Many people and organizations are involved in fulfilling a contract, and each one requires communication of clear instructions. Production personnel cannot be expected to achieve the correct results first time if they do not possess clear instructions. Anything less causes unproductive rectification costs. 9

10 e) Punctual Delivery: Poor delivery is a maor cause of customer dissatisfaction. f) Faultless production: Quality of production not only requires good instructions and equipment, it demands staff capable of performing a professional service. Training and experience is vital to ensure faultless production. g) Effective support services: This is the company operating as a team! If one person is unsure of their obectives, others come to their aid in order to avoid lost time due to shortage of correct information, tools, materials, equipment, etc. h) Feedback of field experience: If things go wrong they should be regarded as an opportunity to learn for the future, and represent the best method for identifying improvements in the company's performance. I.3. Evolution of quality management I.3.a. Quality inspection The main principle is that prefixed quality requirements should be fulfilled all the time. The primary goal was to unfold defects through measurements and standardization. In this period of quality evolution the department of quality inspection was responsible for quality in the whole organization. During the early days of manufacturing, an operative s work was inspected and a decision was made whether to accept or reect it. As businesses became larger, so too did this role and full time inspection obs were created. In the era of Taylor s scientific management (in the early 1900 s) the function of planning and the function of production were separated. It was the first time when the early initiatives of different management methods were used in order to increase efficiency. This meant that the work or process leaders were got rid of controlling end products in the production processes. Instead of them qualified and independent quality controllers were responsible for this duty. This was the time when controlling quality became a separate discipline and profession. The main disadvantage of pure quality inspection was that it didn t assure the improvement of the process as there were no feed-backs into the production process itself. There were always conflicts of interest between the production and quality department. Top managers got very far away from the issue of quality, they got less and less information, and their knowledge about quality was quite dissatisfactory. These changes led to the birth of a separate inspection department with a chief inspector, reporting to either the person in charge of manufacturing or the manager. With the establishment of this new department the question of new services and other issues, e.g, standards, training, recording of data and the accuracy of measuring equipment arose. It became clear that the responsibilities of the chief inspector were more than ust product acceptance and the need to address defect prevention emerged. Hence the quality control department evolved, in charge of which was a quality control manager with responsibility for the inspection services and quality control engineering. 10

11 Accompanying the formation of inspection functions, other problems arose: More technical problems occurred, requiring specialised skills, often not possessed by production workers, The inspectors lacked training, Inspectors were ordered to accept defective goods to increase output, Skilled workers were promoted into other roles, leaving less skilled workers to perform the operational obs such as manufacturing. Until the 190 s quality was essentially inspection, the know-how of inspection was the know-how of quality. Better quality meant a better and extensive inspection activity. I.3.b. Quality control In the 190 s statistical theory began to be applied effectively to quality control, and in 194 Shewhart made the first sketch of a modern control chart. His work was later developed by Deming and the early work of Shewhart, Deming, Dodge and Romig constitutes much of what today comprises the theory of statistical process control (SPC). However, there was little use of these techniques in manufacturing companies until the late 1940 s. After mass-production had started up, most industries arrived at the period of statistical quality control. Technologies, the way of working, the profession of quality inspection developed, and the spread of mass production laid the foundation for the application of statistical based quality control. During the production process measures were put through, and with this kind of in-process control it could be assured that the features of products would meet the requirements. By controlling a random sample taken from the population (i.e. the production process) it was possible to collect enough information based on few pieces of products rather than controlling all products as the random sample taken out of the process could represent the specific features of the whole process in a quantitative way. Quality was therefore the correct application of statistic methods in production in this period of quality evolution. The main aim of this period was controlling and regulating the process. The production and engineering departments were responsible for quality. The process control based on statistical methods assured that quality was not only controlled into the products, but the whole production process (and not only end products) was under control. I.3.c. Total quality control At that time Japan s industrial system was virtually destroyed and it had a reputation for cheap imitation products and an illiterate workforce. The Japanese recognised these problems and set about solving them with the help of some notable quality gurus Juran, Deming and Feigenbaum. In the early 1950 s quality management practices developed rapidly in Japanese plants and become a maor theme in Japanese management philosophy, by 1960 quality control and management had become a national preoccupation. By the late 1960 s Japan s imports into the USA and Europe increased significantly, due to its cheaper, higher quality products compared to the Western counterparts. 11

12 In 1969 the first international conference on quality control, sponsored by Japan, America and Europe, was held in Tokyo. In a paper given by Feigenbaum, the term total quality was used for the first time, and referred to wider issues such as planning, organisation and management responsibility. Ishikawa gave a paper explaining how total quality control in Japan was different, meaning companywide quality control, and describing how all employees, from top management to the workers must study and participate in quality control. Companywide quality management was common in Japanese companies by the late 1970 s. I.3.d. Quality management systems The assurance of quality became a separate function in most European countries and organizations stepped towards the era of quality management systems. All departments and functions participating in the fulfillment of customer needs and expectations take part in the establishment and operation of quality management systems. The main aim is to manage the whole quality management system and to reach an operational optimum. The 1960 s and 70 s are fundamental years for quality management; quality broke through the complete structure and hierarchy of organizations. It was cheaper to manage the quality of an organization than control the quality of single products. At that time quality assurance was born. There are several ways of building up a quality management system, one of the most widespread systems is ISO 9000, but there are several others (see in later chapters). The common feature of these systems is that they give only basic guidelines, milestones to establish and manage such a system. At the end of the 1970 s quality becomes customer satisfaction. The main concept at the beginning of the 1980s is that competitiveness means higher quality at a lower cost, the goal is the continuous improvement of organizational processes. This approach to quality leads to TQM. I.3.e. Total Quality Management The quality revolution in the West was slow to follow, and did not begin until the early 1980 s, when companies introduced their own quality programmes and initiatives to counter the Japanese success. Total quality management (TQM) became the centre of these drives in most cases. In a Department of Trade & Industry publication in 198 it was stated that Britain s world trade share was declining and this was having a dramatic effect on the standard of living in the country. There was intense global competition and any country s economic performance and reputation for quality was made up of the reputations and performances of its individual companies and products/services. The British Standard (BS) 5750 for quality systems had been published in 1979, and in 1983 the National Quality Campaign was launched, using BS5750 as its main theme. The aim was to bring to the attention of industry the importance of quality for competitiveness and survival in the world market place. 1

13 Since then the International Standardisation Organisation (ISO) 9000 has become the internationally recognised standard for quality management systems. It comprises a number of standards that specify the requirements for the documentation, implementation and maintenance of a quality system. TQM is now part of a much wider concept that addresses overall organisational performance and recognises the importance of processes. There is also extensive research evidence that demonstrates the benefits from the approach. As we move into the 1st century, TQM has developed in many countries into holistic frameworks, aimed at helping organisations achieve excellent performance, particularly in customer and business results. In Europe, a widely adopted framework is the so-called Business Excellence or Excellence Model promoted by the European Foundation for Quality Management (EFQM). I. Table: Main features of the different eras in the evolution of quality management Feature Quality inspection Statistical quality control Quality management (assurance) systems Total quality management Primary goal detection of defects control harmonization strategic pressure Way of reaching quality The aim of activities Methods The tasks of quality professionals and quality function Responsible quality for The approach of quality, orientation Solving problems Homogeneous products quality Standardization and measurement control, sorting, calculuses, qualification Homogeneous products with less control Statistical tools and techniques trouble shooting, applying statistical methods quality inspection production and other engineering functions Focusing on preventing defects and on the production process in a wider sense Quality proects and systems quality planning, quality programs, valuation of quality system Inspection Control Establish and manage a quality system Assuring competitive position stable Market and customer needs and expectations Strategic planning, subsuming goals, mobilization of the total system education, trainings, supporting other departments, planning quality proects, setting quality goals and obectives all departments Every employee with the contribution and total commitment of management Prime management system 13

14 1. Figure: Evolution of quality management I.4. Comparing schools of quality: Europe, Japan, USA Based on the quality evolution presented above three different schools of quality can be separated: the Japanese, the North-American and the European quality culture. Though, these schools of quality have much in common, important differences have to be highlighted when comparing them regarding the spread of quality approach within the organization, the base of quality movement in the organization, the special features and the key elements. The following table summarizes the main features of the three different schools of quality. II. Table: Main features of the different schools of quality Feature Japan USA Europe spread multitudinous, bottom-up strategy top-down strategy, snowball principle production and technology management bearing group quality circles top management middle management special features totality, simple tools and techniques management environment standardization, regulation key element quality circles management climate documented shadowing One of the basic features of the American school of quality is that the management is quite strong and manages organizations self-confidently with tough methods. The American answer to the unbelievable development of Japan was the TQM which aimed at gaining the top management over quality issues and followed a top-down approach. The base of TQM is a long term strategic perspective and an integrated approach typical of management. Besides the powerful management environment, TQM delegates responsibilities and decisions down in the organizational hierarchy. This kind of empowerment was such a new element that has had an impact on the development of other schools as well. In contrary to the American approach, the Japanese school s main characteristic is the bottomup approach which means a high involvement of employees in the form of quality circles. 14

15 These quality circles are initiatives for solving quality problems and improving the quality of products and processes. Individualism is behind collectivism, the Japanese share success which is the cultural reason for the effectiveness of quality circles. In a Japanese organisation quality is a collective issue which means that every employee is empowered and feel free to take part. Another main feature is the widespread application of quite simple management tools and techniques. Continuous improvement is an everyday issue. The European school of quality is based on punctuality, documentation, standardization and on the observation of rules and execution of duties. This school evolved at the latest and took over totality from other schools, but the focus is on quality assurance. The main goal is to cover the whole system and processes with systematic exploration of defects and corrective actions. This philosophy quantifies the level of control and capability of processes which needs a well-established measurement background. The European model is strongly formalized; the middle management is the flagship of quality issues within the organization. The focus is on traceability and on corrective and preventive actions, the model prefer standardized systems which are regularly revised and certified by independent institutions. The main conclusions are the followings. The momentous milestones of the evolution of quality management can be linked to the world s developed industrial regions. Not only do the special features of a given school owe to the impacts deriving from the global economic competition, but the economic, political and cultural circumstances of a region influences them as well. The quality philosophies, the schools and applied methods are strongly influenced by cultural aspects. While the top management and success-orientation is strong in the USA, in Japan the culture-rooted public spirit and in Europe the skills, qualifications and the follow of formal rules are dominant. The presented quality schools influence each other which has never been a simple copying, rather the integration of new elements into their own culture which has been fulfilled by the application of management and motivation tools and techniques that characterize a given region. The TQM philosophy has spread in Japan and in Europe as well, both the Japanese and European systems converge to TQM, but at the same time TQM enriches with the application of the Japanese and the European quality toolbar as well. I.5. Gurus of the quality movement I.5.a. Taylor Frederick W. Taylor ( ) is credited with being one of the firsts to implement new approaches to improve the work of unskilled workers in industrial organizations. Taylor as a chief engineer developed a series of concepts that laid a foundation of work improvement for the 0 th century. The systematic approach of analysis and the application of some basic concepts to manual work earned Taylor the title of father of scientific management. In his book, The Principles of Scientific Management, Taylor reveals a few elements of his management theory: 15

16 A daily task: each person in every organization should have a clearly defined, large task which should take one day to complete. Standard conditions: the worker should have standard tools and conditions to complete the task. High pay for success: significant rewards should be paid for the successful completion of the task. High loss for failure: failure for completing the tasks should be personally costly. Tasks in large sophisticated organizations should be made difficult so as to require skilled, accomplished workers. Taylor's scientific management consisted of four principles: 1. Replace rule-of-thumb work methods with methods based on a scientific study of the tasks.. Scientifically select, train, and develop each employee rather than passively leaving them to train themselves. 3. Provide "Detailed instruction and supervision of each worker in the performance of that worker's discrete task". 4. Divide work nearly equally between managers and workers, so that the managers apply scientific management principles to planning the work and the workers actually perform the tasks. He also isolated the activity of planning from work improvement. This initiative deprived the worker of his responsibility to improve work. This resulted in the set-up of a separate department of inspectors to monitor the quality of the output. At the same time this lateral thinking diffused the responsibility of quality within the organization. This group of inspectors, reporting to a chief inspector, became known later as the quality assurance department. I.5.b. Shewart Walter A. Shewart ( ) was a statistician employed by Bell Labs during the 190s and 30s. His book The Economic Control of Quality of Manufactured Products was considered by statisticians as a landmark contribution to the effort to improve quality of manufactured goods. Shewart reported that variations exist in every facet of manufacturing but that variations could be understood through the application of simple statistical tools such as sampling and probability analysis. Shewart s techniques taught that work process could be brought under control by defining when a process should be left alone and when intervention was necessary. He was able to define the limits of random variation that occur in completing any task and said that intervention should occur only when these limits had been exceeded. He developed control charts to monitor performance over time, thereby providing workers with the ability to monitor their work and predict when they were about to exceed limits and possibly produce waste. 16

17 Shewart s work in sampling and control charts attracted the interest of another statistician, W. Edwards Deming. I.5.c. Deming As a statistician, W. Edwards Deming ( ) trained Japanese engineers in the 1950s. He significantly assisted in Japan s remarkable recovery from the devastation of World War II. He learned that quality is not determined on the shop floor but in the executive suite. In 1950, the Union of Japanese Scientists and Engineers invited Deming to come to Japan and deliver a series of lectures on quality. Deming has summarized his concepts and principles in a series of fourteen points and seven deadly diseases. His approach can be described as follows: Quality is primarily the result of senior management s actions and decisions and not the result of actions taken by workers. Deming stresses that it is the system of work that determines how work is performed and only managers can create the system. Only managers can allocate resources, provide trainings to workers, select the equipment and tools that workers use, and provide the plant and the environment to achieve quality. Only senior managers determine the markets in which the firm will participate and what products or services will be sold. The worker, in turn, is responsible for the solution of those special problems caused by actions or events directly under his or her control. Deming attempts to separate the common from the special causes that contribute to the variation in product or service quality and thereby allocate correctly the task of improving quality between the manager and the worker. He advocates the use of statistical quality control, since he believes it is the statistical understanding of systems that allows accurate diagnosis and solution of problems. Seven deadly diseases: 1. Lack of constancy of purpose;. Emphasis on short term profits; 3. Evaluation of performance, merit rating or annual review; 4. Mobility of management; 5. Management by use of visible figures; 6. Excessive medical costs; 7. Excessive costs of liability. Deming s 14 points: 1. Create and publish to all employees a statement of the aims and purposes of the company or other organization. The management must demonstrate constantly their commitment to this statement.. Learn the new philosophy, top management and everybody. 3. Understand the purpose of inspection, for improvement of processes and reduction of cost. 4. End the practice of awarding business on the basis of price tag alone. 5. Improve constantly and forever the system of production and service. 6. Institute trainings (for skills). 17

18 7. Teach and institute leadership. 8. Drive out fear, create trust, create a climate for innovation. 9. Optimize toward the aims and purposes of the company the efforts of teams, groups, staff areas, too. 10. Eliminate exhortations for the workforce. 11. Eliminate numerical quotas for production. Instead, learn and institute methods for improvement. Eliminate management by obectives. Instead, learn capabilities of processes and how to improve them. 1. Remove barriers that rob people of pride of workmanship. 13. Encourage education and self-improvement for everyone. 14. Take action to accomplish transformation. I.5.d. Juran Joseph M. Juran ( ) worked within the Bell System until the start of the World War II. Juran was also familiar with Shewart s work and was personally involved in applying these and other statistical approaches in the production of telephone equipment. Juran like Deming also assisted Japanese leaders in taking charge of restructuring their industries so they could export products to world markets. Working independently of Deming (who focused on the use of statistical process control), Juran who focused on managing for quality went to Japan and started courses (1954) in quality management. The training started with top and middle management. He was able to help the Japanese to adapt the quality concepts and tools designed primarily for the factory into a series of concepts that would become the basis for an overall management process. Juran in his famous book titled Quality Control Handbook (published in 1951) documented three fundamental managerial processes that were originally used to manage the finances of an organization financial planning, financial control and financial improvement and has applied this approach to the task of managing quality. The three elements of the Juran Trilogy are as follows: 1. Quality planning: a process that identifies the customers, their requirements, the product and service features the customers expect, and the processes that will deliver those products and services with correct attributes and then facilitates the transfer of this knowledge to the production department of the organization.. Quality control: A process in which the product is actually examined and evaluated against the original requirements expressed by the customer. Problems detected are then corrected. 3. Quality improvement: A process in which the sustaining mechanisms are put in place so that quality can be achieved on a continuous basis. This includes allocating resources, assigning people to pursue quality proects, and in general establishing a permanent structure to pursue quality and maintain the gains secured. 18

19 I.5.e. Crosby Crosby s ( ) approach to quality is also summarized in fourteen steps but built around the following four fundamental beliefs, which he calls absolutes: 1. Crosby defines quality as conformance to requirements, not elegance. This differs from the conventional definition of quality on that it does not reference the manner in which the item is constructed or the method by which a service is provided. Rather, this definition is strategic, in that it focuses on trying to understand the full array of expectations that a customer has and drives organizations to meet these expectations. Clearly, this external view of quality is energizing, because it establishes targets that may be far more demanding and realistic than those established internally.. Do it right for the first time focusing on prevention, not inspection. This notion attempts to correct the problem created by Taylor by ensuring that the worker manufacturing the product or providing the service does not pass defective work. There will be few, if any, inspectors in a quality organization, since everyone has the responsibility for his or her own work. There is no one else to catch errors. 3. The performance standard is zero defects. Crosby has advocated the notion that zero errors can and should be the target. 4. The measurement of quality is the cost of quality. Costs of imperfection, if corrected, have an immediate beneficial effect on bottom-line performance as well as on customer relations. To that extent, investments should be made in training and other supporting activities to eliminate errors and recover costs of waste. III. Table: Crosby s Quality Management Conventional wisdom Quality absolutes Reality Definition Goodness Conformance to requirements System Appraisal Prevention Standard That s close enough Zero defects Measure Indices Price of nonconformance Quality improvement process Management commitment quality improvement team measurement cost of quality quality awareness corrective action zero defects planning employee education zero defects day goal setting error cause removal 19

20 recognition quality councils do it all over again I.5.f. Feigenbaum A. V. Feigenbaum (190-), a former manager of manufacturing operations and quality control for GE, has contributed significantly to the worldwide quality movement by developing the approach that the responsibility for quality extends well beyond the manufacturing department. He developed the concept that quality in manufacturing could not be achieved if the products were poorly designed, inefficiently distributed, incorrectly marketed, and improperly supported in the customer s site. Thus, Feigenbaum s idea that every function within the organization is responsible for quality was developed and became known as total quality control. "Total quality control is an effective system for integrating the quality development, quality maintenance, and quality improvement efforts of the various groups in an organization so as to enable production and service at the most economical levels which allow full customer satisfaction. Feigenbaum also originated the concept known as the cost of quality as a means of quantifying the benefits of adopting a total quality management approach. He also refers to the concept of a "hidden" plant the idea that so much extra work is performed in correcting mistakes that there is effectively a hidden plant within any factory. He emphasized the accountability for quality: because quality is everybody's ob, it may become nobody's ob the idea that quality must be actively managed and have visibility at the highest levels of management. The common thrust behind the teachings of these quality gurus is the concept of continuous improvement. Although their approaches differ in technique, emphasis, and application, the obective is the same continuous improvement of every output, whether it is a product or a service, by removing unwanted variation and by improving the underlying work processes. 0

21 II. Development of quality management systems II.1. Conformance to requirements Products and services have to meet the requirements set by laws, regulations, or standards regulating the specific features, characteristics or parameters of the product or service. In several cases the details of the contract between the supplier and customer include similar requirements. II.. Good manufacturing practice GMP A good manufacturing practice (GMP) is a production and testing practice that helps to ensure a quality product. Many countries have legislated that pharmaceutical and medical device companies must follow GMP procedures, and have created their own GMP guidelines that correspond with their legislation. Basic concepts of all of these guidelines remain more or less similar to the ultimate goals of safeguarding the health of the patient as well as producing good quality medicine, medical devices or active pharmaceutical products. GMP regulations require a quality approach to manufacturing, enabling companies to minimize or eliminate instances of contamination, mix-ups, and errors. This in turn, protects the consumer from purchasing a product which is not effective or even dangerous. GMP regulations address issues including recordkeeping, personnel qualifications, sanitation, cleanliness, equipment verification, process validation, and complaint handling. Most GMP requirements are very general and open-ended, allowing each manufacturer to decide individually how to best implement the necessary controls. This provides much flexibility, but also requires that the manufacturer interpret the requirements in a manner which makes sense for each individual business. GMP guidelines are not prescriptive instructions on how to manufacture products. They are a series of general principles that must be observed during manufacturing. When a company is setting up its quality program and manufacturing process, there may be many ways it can fulfill GMP requirements. It is the company's responsibility to determine the most effective and efficient quality process. II.3. Military standards Defence standards evolved from the need to ensure proper performance, maintainability and repairability, and logistical usefulness of military equipment. The latter two goals favour certain general concepts, such as interchange ability, standardization (of equipment and processes, in general), cataloguing, communications, and training (to teach people what is standardized, what is at their discretion, and the details of the standards). In the late 18th century and throughout the 19th, the American and French militaries were early adopters and long time developmental sponsors and advocates of interchange ability and standardization. 1

22 By World War II ( ), virtually all national militaries and transnational alliances of the same were busy standardizing and cataloguing. The U.S. AN- cataloguing system (Army- Navy) and the British Defence Standards (DEF-STAN) provide examples. For example, due to differences in dimensional tolerances, in World War II American screws, bolts, and nuts did not fit British equipment properly and were not fully interchangeable. Defence standards provide many benefits, such as minimizing the number of types of ammunition, ensuring compatibility of tools, and ensuring quality during production of military equipment. II.4. BS5750 and ISO 9000 ISO 9000 was first published in It was based on the BS 5750 series of quality assurance standards from British Standard Institution (BSI) that were proposed to ISO in However, its history can be traced back some twenty years before that, to the publication of the United States Department of Defense MIL-Q-9858 standard in MIL-Q-9858 was revised into the NATO AQAP series of standards in 1969, which in turn were revised into the BS 5179 series of guidance standards published in 1974, and finally revised into the BS 5750 series of requirements standards in 1979 before being submitted to ISO. ISO 9000 is a series of standards, developed and published by the International Organization for Standardization (ISO), that define, establish, and maintain an effective quality assurance system for manufacturing and service industries. Revised in 1994, 000 and then in 008, the international quality management systems standard has proved a global success with more than 1 million ISO 9001 certificates (000 and 008 combined) issued in 178 countries and economies by the end of 009. The ISO 9000 standards are available through national standards bodies. ISO 9000 deals with the fundamentals of quality management systems, including the eight management principles upon which the family of standards is based. ISO 9001 deals with the requirements that organizations wishing to meet the standard must fulfill. We are going to get acquainted with the ISO 9001 requirements in the next chapter. II.5. HACCP Hazard analysis and critical control points (HACCP), is a systematic preventive approach to food safety and allergenic, chemical, and biological hazards in production processes that can cause the finished product to be unsafe, and designs measurements to reduce these risks to a safe level. In this manner, HACCP is referred as the prevention of hazards rather than finished product inspection. The HACCP system can be used at all stages of a food chain, from food production and preparation processes including packaging, distribution, etc. HACCP itself was conceived in the 1960s when the US National Aeronautics and Space Administration (NASA) asked Pillsbury to design and manufacture the first foods for space flights. Since then, HACCP has been recognized internationally as a logical tool for adapting traditional inspection methods to a modern, science-based, food safety system. Based on risk-

23 assessment, HACCP plans allow both industry and government to allocate their resources efficiently in establishing and auditing safe food production practices. Hence, HACCP has been increasingly applied to industries other than food, such as cosmetics and pharmaceuticals. This method, which in effect seeks to plan out unsafe practices based on science, differs from traditional "produce and sort" quality control methods that do nothing to prevent hazards from occurring and must identify them at the end of the process. HACCP is focused only on the health safety issues of a product and not the quality of the product, yet HACCP principles are the basis of most food quality and safety assurance systems. Seven principles of HACCP: Principle 1: Conduct a hazard analysis. Plans determine the food safety hazards and identify the preventive measures. A food safety hazard is any biological, chemical, or physical property that may cause a food to be unsafe for human consumption. Principle : Identify critical control points. A critical control point (CCP) is a point, step, or procedure in a food manufacturing process at which control can be applied and, as a result, a food safety hazard can be prevented, eliminated, or reduced to an acceptable level. Principle 3: Establish critical limits for each critical control point. A critical limit is the maximum or minimum value to which a physical, biological, or chemical hazard must be controlled at a critical control point to prevent, eliminate, or reduce to an acceptable level. Principle 4: Establish critical control point monitoring requirements. Monitoring activities are necessary to ensure that the process is under control at each critical control point. Principle 5: Establish corrective actions. - These are actions to be taken when monitoring indicates a deviation from an established critical limit. The final rule requires a plant's HACCP plan to identify the corrective actions to be taken if a critical limit is not met. Corrective actions are intended to ensure that no product inurious to health or otherwise adulterated as a result of the deviation enters commerce. Principle 6: Establish procedures for ensuring the HACCP system is working as intended. Validation ensures that the plants do what they were designed to do; that is, they are successful in ensuring the production of a safe product. Verification ensures the HACCP plan is adequate, that is, working as intended. Verification procedures may include such activities as review of HACCP plans, CCP records, critical limits and microbial sampling and analysis. Verification also includes 'validation' the process of finding evidence for the accuracy of the HACCP system (e.g. scientific evidence for critical limitations). Principle 7: Establish record keeping procedures. The HACCP regulation requires that all plants maintain certain documents, including its hazard analysis and written HACCP plan, and records documenting the monitoring of critical control points, critical limits, verification activities, and the handling of processing deviations. 3

24 II.6. QS-9000 QS-9000 is the name given to the Quality System Requirements of the automotive industry which were developed by Chrysler, Ford, General Motors and maor truck manufacturers issued in The QS-9000 Quality System requirements are divided into three sections. Section 1: Common requirements, includes the exact text of ISO 9001 with the addition of automotive / heavy trucking requirements. Section : Additional Requirements, includes requirements beyond the scope of ISO 9001, common to all three manufacturers. Section 3: Customer Specific Sections, contains requirements unique to either Ford, General Motors, or Chrysler. II.7. VDA 6.1 VDA 6.1 is a German quality management system standard initiated by the automobile industry. Based on ISO 9001:1994, the quality management system includes all elements of QS 9000, with an additional four requirements specific to VDA 6.1 as follows: Recognition of Product Risk: This is the risk of the product, failing to fulfil its stipulated function, and its effect on the whole assembly. Employee Satisfaction: This covers the perception of the company employees, as well as their needs and expectations that will be met through the company's quality system. Quotation Structure: A customer or market is offered products for purchase or made available to own or to use. Quality History: This section covers the quality history of customer-supplied products and gives an overview of the situation during a particular period. The VDA standard is broken into two parts: Management Products and Processes. II.8. ISO/TS The ISO/TS16949 is an ISO technical specification (and not a standard) aiming to the development of a quality management system that provides for continual improvement, emphasizing defect prevention and the reduction of variation and waste in the supply chain. It is based on the ISO 9001 and the first edition was published in March 00 as ISO/TS 16949:00. ISO/TS 16949:009, in conunction with ISO 9001:008, defines the quality management system requirements for the design and development, production and, when relevant, installation and service of automotive-related products. 4

25 ISO/TS 16949:009 is applicable to sites of the organization where customer-specified parts, for production and/or service, are manufactured. II.9. AS9000 When the ISO 9000 standard was first introduced, the aerospace industry started reforming their quality management systems to comply with the standard. The ISO standard did not go far enough in addressing the additional safety, risk management, and regulatory requirements of their business, so they augmented it with additional quality standards of their own. AS 9000 (Aerospace Standard) is the aerospace version of ISO AS 9000 contains ISO 9001 embedded verbatim plus 7 clarifications or qualifiers and 8 notes to the existing twenty elements of ISO The intent is to standardize and streamline many other aerospace quality management standards. Compared to most other industries, the aerospace sector is considerably more demanding in terms of safety and quality. The standard embodies a significant streamlining of all the current aerospace quality standards. The final result was the AS9100 standard, published by the Society of Automotive Engineers (SAE) International in It was the first comprehensive standard written for the aerospace industry. Beyond the ISO portion, some of the requirements specific to the aerospace industry included in AS9100 are: o o o o o o o o o o o Subcontractor performance approval and review Maintenance, reliability, and safety Configuration management First article inspection Verification of design, validation and testing processes Purchased product verification Product identification throughout its life cycle Production machinery, tools and numerical control programs compliance Work performed at an outside supplier s facility Special processes Technical documentation reporting and control AS 9100 has undergone two revisions since it was first published. The last rewrite was designed to coincide with the release of ISO 001:008, and is known as AS9100C. The most notable changes in AS9100C are the addition of new requirements for proect management and risk management, two of the biggest concerns facing the industry today. The proect management requirements were included to ensure that stakeholders had a say in the changes proposed by a company, as well as ensure that continuous improvement efforts followed along with the proect management process. As for risk management and assessment, it has now been implemented as its own process, weaving in business, supply chain, cost, technical, program constraint and quality considerations. Each business must undergo a risk management review any time a new proect is undertaken, a process is changed, or a product is modified. Acceptable programs, along with each assessed action item, must include a responsible party, mitigation criteria, completion of the criteria, and final acceptance to each line item. 5

26 II.10. ISO/IEC 1705 ISO/IEC 1705 General requirements for the competence of testing and calibration laboratories is the main ISO standard used by testing and calibration laboratories. There are many commonalities with the ISO 9000 standard, but ISO/IEC 1705 is more specific in requirements for competence. It applies directly to those organizations that produce testing and calibration results. The ISO/IEC 1705 standard itself comprises five elements: Scope, Normative References, Terms and Definitions, Management Requirements and Technical Requirements. The two main sections in ISO/IEC 1705 are Management Requirements and Technical Requirements. Management requirements are primarily related to the operation and effectiveness of the quality management system within the laboratory. Technical requirements include factors which determine the correctness and reliability of the tests and calibrations performed in laboratory. In common with other ISO quality standards, ISO/IEC 1705 requires continual improvement. Regular internal audits are expected to indicate opportunities to make the test or calibration better than it was. Additionally, the laboratory will be expected to keep abreast of scientific and technological advances in relevant areas. In common with other accreditation standards of the ISO series (and unlike most ISO standards for management systems), third party auditing (assessment) of the laboratory is normally carried out by the national organization responsible for accreditation. Laboratories are therefore accredited under ISO/IEC 1705, rather than certified or registered (c.f. ISO 9000 series). II.11. Six Sigma Six Sigma is a set of techniques and tools for process improvement. It was developed by Motorola in Six Sigma seeks to improve the quality of process outputs by identifying and removing the causes of defects (errors) and minimizing variability in manufacturing and business processes. It uses a set of quality management methods, including statistical methods, and creates a special infrastructure of people within the organization ("Champions", "Black Belts", "Green Belts", "Yellow Belts", etc.) who are experts in the methods. Each Six Sigma proect carried out within an organization follows a defined sequence of steps and has quantified value targets, for example: reduce process cycle time, reduce pollution, reduce costs, increase customer satisfaction, and increase profits. The term Six Sigma originated from terminology associated with manufacturing, specifically terms associated with statistical modeling of manufacturing processes. The maturity of a manufacturing process can be described by a sigma rating indicating its yield or the percentage of defect-free products it creates. A six sigma process is one in which % of the products manufactured are statistically expected to be free of defects (3.4 defective parts/million), although, as discussed below, this defect level corresponds to only a 4.5 sigma 6

27 level. Motorola set a goal of "six sigma" for all of its manufacturing operations, and this goal became a by-word for the management and engineering practices used to achieve it. Organizations need to determine an appropriate sigma level for each of their most important processes and strive to achieve these. As a result of this goal, it is incumbent on management of the organisation to prioritize areas of improvement. Six Sigma methodology asserts that: Continuous efforts to achieve stable and predictable process results (i.e., reduce process variation) are of vital importance to business success. Manufacturing and business processes have characteristics that can be measured, analyzed, controlled and improved. Achieving sustained quality improvement requires commitment from the entire organization, particularly from top-level management. Features that set Six Sigma apart from previous quality improvement initiatives include: A clear focus on achieving measurable and quantifiable financial returns from any Six Sigma proect. An increased emphasis on strong and passionate management leadership and support. A clear commitment to making decisions on the basis of verifiable data and statistical methods, rather than assumptions and guesswork. In recent years, some practitioners have combined Six Sigma ideas with lean manufacturing to create a methodology named Lean Six Sigma. The Lean Six Sigma methodology views lean manufacturing, which addresses process flow and waste issues, and Six Sigma, with its focus on variation and design, as complementary disciplines aimed at promoting "business and operational excellence". Companies such as GE and IBM use Lean Six Sigma to focus transformation efforts not ust on efficiency but also on growth. It serves as a foundation for innovation throughout the organization, from manufacturing and software development to sales and service delivery functions. The International Organisation for Standards (ISO) has published ISO 13053:011 defining the six sigma process. II.1. ISO ISO is a family of standards related to environmental management that exists to help organizations (a) minimize how their operations negatively affect the environment (i.e. cause adverse changes to air, water, or land); (b) comply with applicable laws, regulations, and other environmentally oriented requirements, and (c) continually improve in the above. ISO is similar to ISO 9000 quality management in that both pertain to the process of how a product is produced, rather than to the product itself. As with ISO 9000, certification is 7

28 performed by third-party organizations rather than being awarded by ISO directly. The ISO audit standard applies when auditing for both 9000 and compliance at once. II.13. EFQM The EFQM Excellence Model is a non-prescriptive framework for organizational management systems, promoted by EFQM (formerly known as the European Foundation for Quality Management) and designed for helping organizations in their drive towards being more competitive. The Model is regularly reviewed and refined: the last update was published in 013. Regardless of sector, size, structure or maturity, organizations need to establish appropriate management systems in order to be successful. The EFQM Excellence Model is a practical tool to help organizations do this by measuring where they are on the path to excellence; helping them understand the gaps; and then stimulating solutions. The EFQM Model acts as a common reference tool helping organisations move towards Excellence. Thus, the Model provides its users with a set of performance improvement tools in order for them to achieve and sustain results and Excellence. The Model can be used to understand the relations of cause and effects between what organisations do and the results they get. There are 3 components of the Model: Fundamental concepts, representing eight core values or key management principles that drive sustainable success Nine criteria, separated in to categories of enablers and results RADAR logic, continuous improvement cycle used by EFQM. It was originally derived from the PDCA cycle. We are going to get to know the model in details in a later chapter. II.14. Common Assessment Framework The Common Assessment Framework (CAF) is the common European quality management instrument for the public sector. It is a free tool to assist public sector organisations to improve their performance. The CAF helps the organisations to perform a self-assessment with the involvement of all staff, to develop an improvement plan based on the results of the self-assessment and to implement the improvement actions. The model "is based on the premise that excellent results in organisational performance, citizens/customers, people and society are achieved through leadership driving strategy and planning, people, partnerships and resources, and processes. It looks at the organisation from different angles at the same time, the holistic approach of organisation performance analysis." The CAF has four main purposes: To introduce public administrations to the principles of TQM and gradually guide them, through the use and understanding of self-assessment, from the current Plan- Do sequence of activities to a full fledged Plan-Do-Check-Act (PDCA) cycle; 8

29 To facilitate the self-assessment of a public organisation in order to arrive at a diagnosis and improvement actions; To act as a bridge across the various models used in quality management; To facilitate bench learning between public-sector organisations. 9

30 III. ISO 9000 standards III.1. Quality management system An organisation will benefit from establishing an effective quality management system (QMS). The cornerstone of a quality organisation is the concept of the customer and supplier working together for their mutual benefit. For this to become effective, the customer-supplier interfaces must extend into and outside of the organisation, beyond the immediate customers and suppliers. A QMS can be defined as: A set of co-ordinated activities to direct and control an organisation in order to continually improve the effectiveness and efficiency of its performance. These activities interact and are affected by being in the system, so the isolation and study of each one in detail will not necessarily lead to an understanding of the system as a whole. The main thrust of a QMS is in defining the processes, which will result in the production of quality products and services, rather than in detecting defective products or services after they have been produced A fully documented QMS will ensure that two important requirements are met: The customers requirements confidence in the ability of the organisation to deliver the desired product and service consistently meeting their needs and expectations. The organisation s requirements both internally and externally, and at an optimum cost with efficient use of the available resources materials, human, technology and information. QMS enables an organisation to achieve the goals and obectives set out in its policy and strategy. It provides consistency and satisfaction in terms of methods, materials, equipment, etc, and interacts with all activities of the organisation, beginning with the identification of customer requirements and ending with their satisfaction, at every transaction interface. Management systems are needed in all areas of activity, whether large or small businesses, manufacturing, service or public sector. A good QMS will: Set direction and meet customers expectations Improve process control Reduce wastage Lower costs Increase market share Facilitate training Involve staff Raise morale 30

31 III.. ISO 9000 ISO is the International Organization for Standardization, a worldwide federation of national standards organizations, which serves as a link between the standards of various national organizations. It has a membership of 160 national standards institutes from all over the world and ISO s portfolio consists of more than standards and related documents. ISO develops voluntary technical standards which add value to all types of business operations. They contribute to the dissemination of technology and good business practice. They support the development, manufacturing and supply of more efficient, safer and cleaner products and services. They make trade between countries easier and fairer. ISO standards also safeguard users and consumers, and make many aspects of their lives simpler. Published under the designation of International Standards, ISO standards represent an international consensus on the state of the art in the technology or good practice concerned. The benefits of having common international standards are significant. Manufacturers of raw materials, processed materials, parts, components, and subassemblies in one country can much more easily compete against suppliers in the home country of the customer, assuming that they have achieved internationally recognized ISO 9000 quality systems certification. III.3. ISO 9000 family of standards The ISO 9000 quality standards were introduced in 1987 as an umbrella set of quality system standards by the International Organization for Standardization. The ISO 9000 family addresses various aspects of quality management and contains some of ISO s best known standards. The standards provide guidance and tools for companies and organizations who want to ensure that their products and services consistently meet customer s requirements, and that quality is consistently improved. The ISO 9000 family of international quality management standards and guidelines has earned a global reputation as a basis for establishing effective and efficient quality management systems. Standard Title Edition ISO 9000:005 Quality management systems Fundamentals and vocabulary Third ISO 9001:008 Quality management systems Requirements Fourth ISO 9004:000 ISO 19011:00 Quality management systems Guidelines for performance improvements Guidelines for quality and/or environmental management systems auditing Second First The ISO 9000 standard provides the fundamentals and vocabulary used in the entire ISO 9000 family of standards. It sets the stage for understanding the basic elements of quality management as described in the ISO standards. ISO 9000 introduces users to the eight Quality 31

32 Management Principles (see later) as well as the use of the process approach to achieve continual improvement. ISO 9001 is used when you are seeking to establish a quality management system that provides confidence in your organization s ability to provide products that fulfil customer needs and expectations. It is the standard in the ISO 9000 family against whose requirements your quality management system can be certified by an external body. The standard recognizes that the term product applies to services, processed material, hardware and software intended for your customer. There are five sections in the standard that specify activities that need to be considered when you implement your system: Overall requirements for the quality management system and documentation Management responsibility, focus, policy, planning and obectives Resource management and allocation Product realization and process management, and Measurement, monitoring, analysis and improvement. Continual improvement of the quality management system Customers and other interested parties Resource management Management responsibility Measurement, analysis and improvement Customers and other interested parties Satisfaction Requirements Input Product realization Product Output. Figure: Structure of ISO 9001 The requirements in four of the sections are applicable to all organizations Quality management system, Management responsibility, Resource management, and Measurement, analysis and improvement. The Product realization section may be tailored to meet the needs of the organization. The quality manual or other documentation will demonstrate how an organization meets the ISO 9001 requirements. Together, the five sections of ISO 9001 define what an organization should do to consistently provide product that meets customer and applicable statutory and regulatory requirements. In addition, an organization will seek to enhance customer satisfaction by continual improvement of its quality management system. 3

33 ISO 9004 is used to extend the benefits obtained from ISO 9001 to all parties that are interested in or affected by the operations of an organization. Interested parties include employees, owners, suppliers, partners and society in general. ISO 9001 and ISO 9004 are compatible and can be used separately or in combination to meet or exceed expectations of customers and interested parties. Both standards apply a process approach. Processes are recognized as consisting of one or more linked activities that require resources and must be managed to achieve predetermined output. The output of one process may directly from the input to the next process and the final product is often the result of a network or system of processes. The eight Quality Management Principles provide the basis for the performance improvement. ISO 9004 gives guidance on a wider range of obectives of a quality management system than does ISO 9001, particularly in managing for the long-term success of an organization. ISO 9004 is recommended as a guide for organizations whose top management wishes to extend the benefits of ISO 9001 in pursuit of systematic and continual improvement of the organization s overall performance. However, it is not intended for certification or contractual purposes. ISO covers the area of auditing of quality management systems and environmental management systems. It provides guidance on the audit programmes, the conduct of internal or external audits, and information on auditor competence. ISO provides an overview of how an audit programme should operate and how management system audits should take place. Effective audits ensure that an implemented QMS meets the requirements specified in ISO III.3.a. Implementing and maintaining a quality management system based on the ISO 9001 standard The implementation process is important in achieving the full benefits of the quality management system (QMS). Most new users will obtain measurable payback early in the process. For a successful implementation of your QMS, these seven steps are recommended: 1. Fully engage top management to Define why an organization want to implement ISO 9001 Define the organization s mission, vision, and values Define the organization s stakeholders: customers, suppliers, stockholders, employees, society, etc. Define quality policy, and Define and align organizational obectives and related product/service quality obectives.. Identify key processes and the interactions needed to meet quality obectives 3. Implement and manage the QMS and its processes (using process management techniques) 4. Build the ISO 9001-based QMS 33

34 Identify ISO 9001 requirements Map these requirements with the implemented QMS, where applicable Make a gap analysis : identify where in the existing system the requirements are fulfilled, and where they are not Include in the QMS processes the activities, procedures and controls needed. 5. Implement the system, train company staff and verify effective operation of the processes 6. Manage the QMS Focus on customer satisfaction Monitor and measure the operation of the QMS Strive for continual improvement Consider implementing business excellence models in the company operations 7. If necessary, seek third party certification/registration of the QMS or alternatively, issue a self-declaration of conformity III.4. Quality management principles This chapter introduces the eight quality management principles on which the quality management system standards of the ISO 9000 series are based. These principles can be used by senior management as a framework to guide their organizations towards improved performance. The eight quality management principles are defined in ISO 9000:005 (Quality management systems Fundamentals and vocabulary), and in ISO 9004:009 (Managing for the sustained success of an organization A quality management approach). Principle 1 Customer focus Organizations depend on their customers and therefore should understand current and future customer needs, should meet customer requirements and strive to exceed customer expectations. Key benefits: Increased revenue and market share obtained through flexible and fast responses to market opportunities Increased effectiveness in the use of the organization s resources to enhance customer satisfaction Improved customer loyalty leading to repeat business. Applying the principle of customer focus typically leads to: Researching and understanding customer needs and expectations Ensuring that the obectives of the organization are linked to customer needs and expectations Communicating customer needs and expectations throughout the organization Measuring customer satisfaction and acting on the results 34

35 Systematically managing customer relationships Ensuring a balanced approach between satisfying customers and other interested parties (such as owners, employees, suppliers, financiers, local communities and society as a whole). Principle Leadership Leaders establish unity of purpose and direction of the organization. They should create and maintain the internal environment in which people can become fully involved in achieving the organization s obectives. Key benefits: People will understand and be motivated towards the organization s goals and obectives Activities are evaluated, aligned and implemented in a unified way Miscommunication between levels of an organization will be minimized. Applying the principle of leadership typically leads to: Considering the needs of all interested parties including customers, owners, employees, suppliers, financiers, local communities and society as a whole Establishing a clear vision of the organization s future Setting challenging goals and targets Creating and sustaining shared values, fairness and ethical role models at all levels of the organization Establishing trust and eliminating fear Providing people with the required resources, training and freedom to act with responsibility and accountability Inspiring, encouraging and recognizing people s contributions. Principle 3 Involvement of people People at all levels are the essence of an organization and their full involvement enables their abilities to be used for the organization s benefit. Key benefits: Motivated, committed and involved people within the organization Innovation and creativity in furthering the organization s obectives People being accountable for their own performance People eager to participate in and contribute to continual improvement. Applying the principle of involvement of people typically leads to: People understanding the importance of their contribution and role in the organization People identifying constraints to their performance People accepting ownership of problems and their responsibility for solving them People evaluating their performance against their personal goals and obectives 35

36 People actively seeking opportunities to enhance their competence, knowledge and experience People freely sharing knowledge and experience People openly discussing problems and issues Principle 4 Process approach A desired result is achieved more efficiently when activities and related resources are managed as a process. Key benefits: Lower costs and shorter cycle times through effective use of resources Improved, consistent and predictable results Focused and prioritized improvement opportunities. Applying the principle of process approach typically leads to: Systematically defining the activities necessary to obtain a desired result Establishing clear responsibility and accountability for managing key activities Analysing and measuring of the capability of key activities Identifying the interfaces of key activities within and between the functions of the organization Focusing on the factors such as resources, methods, and materials that will improve key activities of the organization Evaluating risks, consequences and impacts of activities on customers, suppliers and other interested parties. Principle 5 System approach to management Identifying, understanding and managing interrelated processes as a system contributes to the organization s effectiveness and efficiency in achieving its obectives. Key benefits: Integration and alignment of the processes that will best achieve the desired results Ability to focus effort on the key processes Providing confidence to interested parties as to the consistency, effectiveness and efficiency of the organization. Applying the principle of system approach to management typically leads to: Structuring a system to achieve the organization s obectives in the most effective and efficient way Understanding the interdependencies between the processes of the system Structured approaches that harmonize and integrate processes Providing a better understanding of the roles and responsibilities necessary for achieving common obectives and thereby reducing cross-functional barriers Understanding organizational capabilities and establishing resource constraints prior to action 36

37 Targeting and defining how specific activities within a system should operate Continually improving the system through measurement and evaluation. Principle 6 Continual improvement Continual improvement of the organization s overall performance should be a permanent obective of the organization. Key benefits: Performance advantage through improved organizational capabilities Alignment of improvement activities at all levels to an organization s strategic intent Flexibility to react quickly to opportunities. Applying the principle of continual improvement typically leads to: Employing a consistent organization-wide approach to continual improvement of the organization s performance Providing people with training in the methods and tools of continual improvement Making continual improvement of products, processes and systems an obective for every individual in the organization Establishing goals to guide, and measures to track, continual improvement Recognizing and acknowledging improvements. Principle 7 Factual approach to decision making Effective decisions are based on the analysis of data and information. Key benefits: Informed decisions An increased ability to demonstrate the effectiveness of past decisions through reference to factual records Increased ability to review, challenge and change opinions and decisions. Applying the principle of factual approach to decision making typically leads to: Ensuring that data and information are sufficiently accurate and reliable Making data accessible to those who need it Analysing data and information using valid methods Making decisions and taking action based on factual analysis, balanced with experience and intuition. Principle 8 Mutually beneficial supplier relationships An organization and its suppliers are interdependent and a mutually beneficial relationship enhances the ability of both to create value. Key benefits: Increased ability to create value for both parties Flexibility and speed of oint responses to changing market or customer needs and expectations 37

38 Optimization of costs and resources. Applying the principle of mutually beneficial supplier relationships typically leads to: Establishing relationships that balance short-term gains with long-term considerations Pooling of expertise and resources with partners Identifying and selecting key suppliers Clear and open communication Sharing information and future plans Establishing oint development and improvement activities Inspiring, encouraging and recognizing improvements and achievements by suppliers. III.5. ISO 9001 requirements - Quality management system (Clause 4) III.5.a. General requirements Establish, document, implement, and maintain a quality management system. Continually improve its effectiveness in accordance with ISO 9001 requirements. Implement the system to: Determine processes needed for the quality management system (and their application throughout the organization) Determine process sequence and interaction Determine criteria and methods for process operation and control Ensure resources and supporting information are available Monitor, measure where applicable, and analyze these processes Implement actions to achieve planned results and continual process improvement Manage these processes in accordance with ISO 9001 requirements. Define the type and extent of control applied to any outsourced processes that affect product conformity to requirements. III.5.b. Documentation requirements III.5.b.1 General requirements Establish, document, implement, and maintain a quality management system. Continually improve its effectiveness in accordance with ISO 9001 requirements. Implement the system to: Determine processes needed for the quality management system (and their application throughout the organization) Determine process sequence and interaction Determine criteria and methods for process operation and control Ensure resources and supporting information are available Monitor, measure where applicable, and analyze these processes Implement actions to achieve planned results and continual process improvement 38

39 Manage these processes in accordance with ISO 9001 requirements. Define the type and extent of control applied to any outsourced processes that affect product conformity to requirements. III.5.b. Quality manual Establish and maintain a quality manual with: Scope of the quality management system Details and ustification for any exclusions Procedures or references to the procedures Description of interaction between processes III.5.b.3 Control of documents Control the documents required by the quality management system. Records are a special type of document and must be controlled. Establish a documented procedure to: Approve documents for adequacy prior to issue Review, update as necessary, and re-approve documents Identify the changes and current document revision status Make relevant documents available at points of use Ensure the documents remain legible and readily identifiable Identify external documents and control their distribution Prevent obsolete documents from unintended use Apply suitable identification if obsolete documents are retained III.5.b.4 Control of records Establish and control records as evidence of conformity to requirements and to demonstrate the effective operation of the quality management system. Establish a documented procedure to define the controls needed for record: Identification Storage Protection Retrieval Retention Disposition Keep records legible, readily identifiable, and retrievable. III.6. ISO 9001 requirements Management responsibility (Clause 5) III.6.a. Management commitment Provide evidence of management commitment to develop and implement the quality management system as well as continually improve its effectiveness by: Expressing the importance of meeting requirements 39

40 Establishing the quality policy and quality obectives Conducting management reviews Ensuring the availability of necessary resources III.6.b. Customer focus Ensure customer requirements are determined and met in order to improve customer satisfaction. III.6.c. Quality policy Ensure the quality policy is: Appropriate to the purpose of the organization Focused on meeting requirements and continual improvement Used as a framework for quality obectives Communicated and understood at appropriate levels Reviewed for continuing suitability III.6.d. Planning III.6.d.1 Quality obectives Ensure quality obectives, including those needed to meet product requirements, are established at the relevant functions and levels within the organization. Ensure quality obectives are measurable and consistent with the quality policy. III.6.d. Quality management system planning Ensure that planning for the quality management system: Meets the general requirements, as well as, quality obectives Maintains the system integrity when changes are planned and implemented III.6.e. Responsibility, authority, communication Ensure responsibilities and authorities are defined and communicated within the organization. III.6.e.1 Management representative Appoint a member of your management who, irrespective of other duties, has the responsibility and authority to: Ensure the needed processes are established, implemented, and maintained Report to top management on quality management system performance Report to top management on any need for improvement Ensuring the promotion of awareness of customer requirements III.6.e. Internal communication Ensure the appropriate communication processes are established and carried out within the organization regarding the effectiveness of the system. 40

41 III.6.f. Management review III.6.f.1 General Review the quality management system at planned intervals to: Ensure a suitable, adequate, and effective system Assess possible opportunities for improvement Evaluate the need for any changes to the system Consider the need for changes to the quality policy and obectives Maintain records of the management reviews. III.6.f. Review input Inputs for management review must include information on: Results of audits Customer feedback Process performance and product conformity Status of preventive and corrective actions Follow-up actions from earlier reviews Changes that could affect the quality system Recommendations for improvement III.6.f.3 Review output Outputs from the management review must include any decisions and actions related to: Improvement of the effectiveness of the quality management system and its processes Improvement of product related to customer requirements Resource needs III.7. ISO 9001 requirements Resource management (clause 6) III.7.a. Provision of resources Determine and provide the resources necessary to: Implement and maintain the quality management system Continually improve the effectiveness of the system Enhance customer satisfaction by meeting customer requirements III.7.b. Human resources III.7.b.1 General Ensure people performing work affecting conformity to product requirements are competent based on the appropriate education, training, skills, and experience. III.7.b. Competence, training and awareness The organization must: 41

42 Determine the competency needs for personnel Provide training (or take other actions) to achieve the necessary competence Evaluate the effectiveness of the actions taken Inform employees of the relevance and importance of their activities Ensure they know their contribution to achieving quality obectives Maintain education, training, skill, and experience records III.7.c. Infrastructure Determine, provide, and maintain the necessary infrastructure to achieve product conformity. Infrastructure includes, as applicable: Buildings, workspace, and associated utilities Process equipment (both hardware and software) Supporting services (such as transport, communication, or information systems) III.7.d. Work environment Determine and manage the work environment needed to achieve product conformity. III.8. ISO 9001 requirements Product realization (clause 7) III.8.a. Planning of product realization Plan and develop the processes needed for product realization. Keep the planning consistent with other requirements of the quality management system and document it in a suitable form for the organization. Determine through the planning, as appropriate, the: Quality obectives and product requirements Need for processes, documents, and resources Verification, validation, monitoring, measurement, inspection, and test activities Criteria for product acceptance Records as evidence the processes and resulting product meet requirements III.8.b. Customer-related processes III.8.b.1 Determination of Requirements Related to the Product Determine customer requirements: Specified for the product (including delivery and post-delivery activities) Not specified for the product (but needed for specified or intended use, where known) Determine: Statutory and regulatory requirements applicable to the product Any additional requirements considered necessary by the organization 4

43 III.8.b. Review of Requirements Related to the Product Review the product requirements before committing to supply the product to the customer in order to: Ensure product requirements are defined Resolve any requirements differing from those previously expressed Ensure its ability to meet the requirements Maintain the results of the review, and any subsequent follow-up actions. When the requirements are not documented, they must be confirmed before acceptance. If product requirements are changed, ensure relevant documents are amended and relevant personnel are made aware of the changed requirements. III.8.b.3 Customer communication Determine and implement effective arrangements for communicating with customers on: Product information Inquiries, contracts, or order handling (including amendments) Customer feedback (including customer complaints) III.8.c. Design and development III.8.c.1 Design and development planning Plan and control the product design and development. This planning must determine the: Stages of design and development Appropriate review, verification, and validation activities for each stage Responsibility and authority for design and development The interfaces between the different involved groups must be managed to ensure effective communication and the clear assignment of responsibility. Update, as appropriate, the planning output during design and development. III.8.c. Design and development inputs Determine product requirement inputs and maintain records. The inputs must include: Functional and performance requirements Applicable statutory and regulatory requirements Applicable information derived from similar designs Requirements essential for design and development Review these inputs for adequacy. Resolve any incomplete, ambiguous, or conflicting requirements. III.8.c.3 Design and development outputs Document the outputs of the design and development process in a form suitable for verification against the inputs to the process. The outputs must: Meet design and development input requirements 43

44 Provide information for purchasing, production, and service Contain or reference product acceptance criteria Define essential characteristics for safe and proper use Be approved before their release III.8.c.4 Design and development review Perform systematic reviews of design and development at suitable stages in accordance with planned arrangements to: Evaluate the ability of the results to meet requirements Identify problems and propose any necessary actions The reviews must include representatives of the functions concerned with the stage being reviewed. Maintain the results of reviews and subsequent follow-up actions. III.8.c.5 Design and development verification Perform design and development verification in accordance with planned arrangements (see 7.3.1) to ensure the output meets the design and development input requirements. Maintain the results of the verification and subsequent follow-up actions. III.8.c.6 Design and development validation Perform validation in accordance with planned arrangements to confirm the resulting product is capable of meeting the requirements for its specified application or intended use, where known. When practical, complete the validation before delivery or implementation of the product. Maintain the results of the validation and subsequent follow-up actions. III.8.c.7 Control of design and development changes Identify design and development changes and maintain records. Review, verify, and validate (as appropriate) the changes and approve them before implementation. Evaluate the changes in terms of their effect on constituent parts and products already delivered. Maintain the results of the change review and subsequent follow-up actions. III.8.d. Purchasing III.8.d.1 Purchasing process Ensure that purchased product conforms to its specified purchase requirements. The type and extent of control applied to the supplier and purchased product depends upon the effect of the product on the subsequent realization processes or the final product. Evaluate and select suppliers based on their ability to supply product in accordance with the requirements. Establish the criteria for selection, evaluation, and re-evaluation. Maintain the results of the evaluations and subsequent follow-up actions. III.8.d. Purchasing information Ensure the purchasing information contains information describing the product to be purchased, including the requirements for: Approval of product, procedures, processes, and equipment 44

45 Qualification of personnel (Also include quality management system requirements in the purchasing information) Ensure the adequacy of the specified requirements before communicating the information to the supplier. III.8.d.3 Verification of purchased product Establish and implement the inspection or other necessary activities for ensuring the purchased products meet the specified purchase requirements. If the organization or its customer proposes to verify the product at the supplier location, state the intended verification arrangements and method of product release in the purchasing information. III.8.e. Production and service provision III.8.e.1 Control of production and service provision Plan and carry out production and service provision under controlled conditions to include, as applicable: Availability of product characteristics information Availability of work instructions, as necessary Use of suitable equipment Availability and use of monitoring and measuring equipment Implementation of monitoring and measurement activities Implementation of product release, delivery, and post-delivery activities III.8.e. Validation of processes for production and service provision Validate any production or service provision where subsequent monitoring or measurement cannot verify the output. This validation includes processes where deficiencies may become apparent only after product use or service delivery. Demonstrate through the validation the ability of processes to achieve the planned results. Establish validation arrangements including, as applicable: Criteria for process review and approval Approval of equipment Qualification of personnel Use of defined methods and procedures Requirements for records Re-validation III.8.e.3 Identification and traceability Identify, where appropriate, the product by suitable means during product realization. Identify the product status with respect to monitoring and measurement requirements throughout product realization. Where traceability is a requirement, control the unique identification of the product and maintain records. 45

46 III.8.e.4 Customer property Exercise care with any customer property while it is under the control of, or being used by, the organization. Identify, verify, protect, and safeguard customer property provided for use, or for incorporation into the product. Record and report any lost, damaged, or unsuitable property to the customer. III.8.e.5 Preservation of product Preserve the product during internal processing and delivery to the intended destination in order to maintain conformity to requirements. As applicable, preservation includes: Identification Handling Packaging Storage Protection Also apply preservation to the constituent parts of the product. III.8.f. Control of measuring and monitoring equipment Determine the monitoring and measurements to be made, and the required equipment, to provide evidence of product conformity. Use and control the monitoring and measuring devices to ensure that measurement capability is consistent with monitoring and measurement requirements. Where necessary to ensure valid results: Calibrate and/or verify the measuring equipment at specified intervals or prior to use Calibrate the equipment to national or international standards (or record other basis) Adust or re-adust as necessary Identify the measuring equipment in order to determine its calibration status Safeguard them from improper adustments Protect them from damage and deterioration Assess and record the validity of prior results if the device is found to not conform to requirements. Maintain records of the calibration and verification results. Confirm the ability of software used for monitoring and measuring for the intended application before its initial use (and reconfirmed as necessary). III.9. ISO 9001 requirements Measurement, analysis and improvement (clause 8) III.9.a. General Plan and implement the monitoring, measurement, analysis, and improvement processes needed to: 46

47 Demonstrate conformity to product requirements Ensure conformity of the system Continually improve effectiveness Determine through planning the need for, and use of, applicable methods, including statistical techniques, as well as, the extent of their use. III.9.b. Monitoring and measurement III.9.b.1 Customer satisfaction Monitor information on customer perception as to whether the organization is meeting requirements (as one of the performance measurements of the quality management system). Define the methods for obtaining and using this information. III.9.b. Internal audit Conduct internal audits at planned intervals to determine if the quality management system: Conforms to planned arrangements Conforms to requirements of ISO 9001 Is effectively implemented and maintained The organization must: Plan the audit program Consider the status and importance of the audited areas Consider the results of prior audits Define the audit criteria, scope, frequency, and methods Select and use impartial and obective auditors (not audit their own work) Establish a documented procedure to address responsibilities and requirements to: Plan audits and conduct audits Establish records and report results Maintain records of the audits and their results. Ensure management of the audited areas takes actions without undue delay to eliminate detected nonconformities and their causes. Verify through follow-up actions the implementation of the action and report the results. III.9.b.3 Monitoring and measurement of process Apply suitable methods for monitoring and, where applicable, measurement of the quality management system processes. Confirm through these methods the continuing ability of each process to satisfy its intended purpose. When the planned results are not achieved, take correction and corrective action, as appropriate. III.9.b.4 Monitoring and measurement of product Monitor and measure product characteristics to verify product requirements are being met. Carry out the monitoring and measuring at the appropriate stages of product realization in 47

48 accordance with planned arrangements. Maintain evidence of conformity with the acceptance criteria. Record the person responsible for authorizing release of product for delivery to the customer. Product release and service delivery cannot proceed until all planned arrangements have been satisfactorily completed, unless otherwise approved by a relevant authority, and where applicable, the customer. III.9.c. Control of nonconfirming product Ensure any nonconforming product is identified and controlled to prevent its unintended use or delivery. Establish a documented procedure to define the controls and related responsibilities and authorities for dealing with nonconforming product. Where applicable, deal with the nonconforming product by one or more of the following ways: Take action to eliminate the detected nonconformity Authorize its use, release, or acceptance by concession Take action to preclude its original intended use or application Take action appropriate to the effects, or potential effects, of the nonconformity when nonconforming product is detected after delivery or use has started Maintain records of the nature of the nonconformity, and any subsequent actions, (including any concessions). When the nonconformity is corrected, re-verify it to show conformity. III.9.d. Analysis of data Determine, collect, and analyze appropriate data to demonstrate the suitability and effectiveness of the quality management system, as well as, evaluate where continual improvement of the effectiveness of the quality management system can be made. Include in the analysis the data generated by monitoring and measuring activities and from other relevant sources. Analyze this data to provide information on: Customer satisfaction Conformity to product requirements Characteristics and trends of processes and products, including opportunities for preventive action Suppliers III.9.e. Improvement III.9.e.1 Continual improvement Continually improve the effectiveness of the quality management system through: Quality policy Quality obectives Audit results 48

49 Analysis of data Corrective and preventive action Management review III.9.e. Corrective action Take corrective action to eliminate the causes of nonconformities and prevent recurrence. Corrective action must be appropriate to effects of the problem. Establish a documented procedure for corrective action that defines requirements to: Review nonconformities (including customer complaints) Determine the causes of nonconformities Evaluate the need for actions to prevent recurrence Determine and implementing the needed action Maintain records of the results of the action taken Review the effectiveness of corrective action taken III.9.e.3 Preventive action Determine the action to eliminate the causes of potential nonconformities in order to prevent their occurrence. Ensure preventive actions are appropriate to the anticipated effects of the potential problem. Establish a documented procedure for preventive action to define requirements to: Determine potential nonconformities and their causes Evaluate the need for actions to prevent occurrence Determine and implementing the needed action Maintain records of the results of the action taken Review the effectiveness of preventive action taken Organizations and companies often want to get certified to ISO s management system standards (for example ISO 9001) although certification is not a requirement. The best reason for wanting to implement these standards is to improve the efficiency and effectiveness of company operations. A company may decide to seek certification for many reasons, as certification may: be a contractual or regulatory requirement be necessary to meet customer preferences fall within the context of a risk management programme, and help motivate staff by setting a clear goal for the development of its management system. To become certified, a firm needs to do an enormous preparation, including assignment of management responsibility; development of detailed operations descriptions; preparation of quality control procedures, product identification and testing, procedures for handling storage, 49

50 packing and shipping of products, training procedures, description of inspection, measuring, and test equipment, procedure for maintaining quality records, and numerous other processes and procedures. Upon completion of the firm s preparatory activities for ISO 9000 quality systems certification, the firm s operations are evaluated by a registrar. The registrar is a third party independent from the firm seeking ISO 9000 certification. The registrar is certified by a national accreditation board. If accredited, the firm must undergo periodic audits by independent registrars. ISO 9000 accreditation is not given permanently but is subect to suspension of the ISO 9000 certified firm does not adhere to its certified process, procedures and quality systems and products. 50

51 IV. Total Quality Management IV.1. Introduction to TQM At the conclusion of World War II, demand for consumer goods was so strong that U.S. producers of goods and services had difficulty meeting market demands. The lessons learned during the war about how to produce quality materials were ignored, and American manufacturers went about the business of meeting the needs of the largest market in the world with quantity, not quality. The huge production capability the US built to produce war material was converted to the mass production of cars, refrigerators and other consumer products. Other countries recovering from the devastation of the war had little infrastructure left and a much smaller market to serve. Starting with a clean state, they designed new, more efficient, more flexible manufacturing capabilities, which enabled them to meet needs of their smaller market and niches in our market. This new flexibility and capability, coupled with a keen understanding of what customers desired, has led to an economic revolution in which smaller countries such as Japan, Korea, Taiwan and Germany have captured large segments of the US market and have also displaced US manufacturers as suppliers in segments of world markets. This postwar reversal in roles has caused US imports to overwhelm our exports, thereby creating a huge trade imbalance and an outflow of US dollars to other countries. As producers in Europe and Asia continued to gain market share in the US and displaced domestic manufacturers in world markets, the question for American was, What they can do about this? What they can do to stem the tide of imports and increase their ability to export their goods and services and to compete internationally? During the thirty-year period ending in 1980, the US dominated a number of industries and world markets. The US supplied the maority of the world s need for cars, radios, TVs, cameras, and copiers because we they were the most efficient producer. Since the 1980s, they have lost their role as supplier to the world. More important, Americans may have lost the one advantage they prized most. They know that Japanese, for example, have supplanted them not only in their traditional markets, such as consumer electronics, but in what was once a unique American competitive advantage know-how. Comparing the Japanese edge in manufacturing to their own capabilities, they find that their Japanese competitor has bested them in a very critical dimension. For every hour of labor the Japanese use to produce a product, the US required more labor to produce the same item. Many studies were completed in the US during the last decade on ways to improve their international competitiveness. Since 1980, the US has been displaced in these markets by foreign competitors that seem to be more innovative, more cost effective, and more capable of meeting the consumer s requirements. 51

52 The result of this superior performance by foreign firms is that the US has been pushed out of supplying the world s maor markets, some of which they developed and once dominated. The post-world War II economy has given way to a new era in which consumers can choose from a full array of good and services. The relative openness of US markets to all foreign firms means that US-managed firms are subected to increasingly intense competition from emerging economies with lower wage rates, from economies that enoy a lower cost of capital, and from economies whose employees have a different work ethic. These conditions, coupled with the fact that American consumers will always choose those goods and services that best meet their full range requirements, mean that American managers must change the way if they are to survive and if US as a nation are to maintain our standard of living. The MIT Commission reported in Made in America that the problems plaguing American industries are not ust random events that, given time, will correct themselves but rather are signs of systematic and pervasive ills that cannot be corrected by working harder to do the same things that failed in the past. Efforts to improve quality were initiated in the US in the early part of the 0 th century by men like Shewart, Deming, Juran, Feigenbaum, Crosby and others. Their approach as to move from the inspection of manufactures products to uncover defects to the prevention of defects. This approach recognizes the expense and waste associated with reworking defective products and emphasizes the need to improve the fundamental processes employed by an organization. Although perfected by the Japanese in the 1960s and 1970s, quality management returned to North America in the 1980s, and the challenge for the USA was to extend these successful manufacturing-improvement concepts to the service sector. IV.1.a. Evolution of the TQM Concept The concept of quality first began in manufacturing organizations producing physical, tangible products. Indeed, it is clear that one worker might receive immediate feedback if the subassembly being passed on were faulty. As a result, much progress has been made in the pursuit of product- and user-based quality as firms pursued either the production or product philosophy in their organizations. But it had to be recognized that the world moved in the 1980s to a new consumer-oriented economy that brings the concept of quality closer to the user-based/value-based approach described by Garvin. The quality gurus introduced above offer a variety of definitions of quality that organizations can employ to describe the management approach they intend to use in order to achieve their organizational obectives. Clearly, such a definition must account for the customer, but we have come to realize that an organization must satisfy many different types of customers. To accommodate the need to meet these varying requirements, TQM offer the following definition: Quality: A basic business strategy that provides goods and services that completely satisfy both internal and external customers by meeting their explicit and implicit expectations. This strategy utilizes the talents of all employees, to the benefit of the organization in particular and society in general, and provides a positive financial return to shareholders. 5

53 IV.1.b. What is a TQM System? It has become clear that quality is not determined by the worker on the shop floor, nor it is determined by the service technician working at the customer s site. Quality is determined by the senior managers of an organization, who by virtue of the positions they hold, are responsible to customers, employees, suppliers, and shareholders for the success of the business. These senior managers allocate resources, decide which markets the firm will enter, and select and implement the management processes that will enable the firm to fulfil their mission and their vision. Combining various teachings of the quality gurus with practical experience has led to the development of a simple but effective model for implementing TQM. This model build on three fundamental principles of total quality focus on the customers (internal and external); focus on improving work processes to produce consistent, acceptable outputs; and focus on utilizing the talents of those with whom we work and six supporting elements. Quality principles Customer focus: quality is based on the concept that everyone has a customer and that the requirements, needs and expectations of that customer must be met every time if the organization as a whole is going to meet the needs of the external customer. This concept requires a thorough collection and analysis of customer requirements, and when these requirements are understood and accepted, they must be met. Process improvement: the concept of continuous improvement is built on the premise that work is the result of a series of interrelated steps and activities that result in an output. Continuous attention to each of these steps in the work process in necessary to reduce the variability of the output and improve the reliability of the process. The first goal of continuous improvement is processes that are reliable reliable in the sense that they produce the desired output each time with no variation. If variability has been minimized and the results are still unacceptable, the second goal of process improvement is to redesign the process to produce an output that is better able to meet the customer s requirements. Total involvement: this approach begins with the active leadership of senior management and includes efforts that utilize the talents of all employees in the organization to gain a competitive advantage in the marketplace. Employees at all levels are empowered to improve their outputs by coming together in new and flexible work structures to solve problems, improve processes and satisfy customers. Suppliers are also included, and over time, become partners by working with empowered employees to the benefit of the organization. Supporting elements Leadership: senior management must lead this effort by example, by applying the tools and language, by requiring the use of data and by recognizing those who successfully apply the concepts of TQM. When installing TQM as the key management process, the importance of the role of senior managers as advocates, 53

54 teachers, and leaders cannot be overstated. Businnes leaders must understand that TQM is such a process and is composed of principles and supporting elements that they must manage in order to achieve continuous quality improvement. Education and training: quality is based on the skills of every employee and his or her understanding of what is required. Educating and training all employees provides the information they need on the mission, vision, direction and strategy of the organization as well as the skills they need to secure quality improvement and resolve problems. This core training ensures that a common language and a common set of tools will be used throughout the firm. Additional training on benchmarking, statistics, and other techniques is also required to pursue and achieve complete customer satisfaction. Supportive structure: senior managers may require support to bring about the change necessary to implement a quality strategy. Such support may be provided by outside consultants, but it is clearly far superior for an organization to be self-sufficient. To gain this self-sufficiency, a small support staff can help the senior management team understand the concepts of quality, assist by networking with other quality managers in other parts of the organization, and serve as a resource on the topic of quality for the senior management team. Communications: communications in a quality environment may need to be addressed differently in order to communicate to all employees a sincere commitment to change. Reward and recognition: teams and individuals who successfully apply the quality process must be recognized and possibly rewarded, so that the rest of the organization will know what is expected. Failure to recognize that someone who achieves success using the touted quality management process will convey the message that this is not the true path to ob success, possible promotion and overall personal success. Recognizing successful quality practitioners provides role models for the rest of the organization. Measurement: the use of data becomes paramount in installing a quality management process. To set the stage for the use of data, external customer satisfaction must be measured to determine the extent to which customers perceive that their needs are being met. The collection of customer data provides an obective, realistic assessment of performance and is useful in motivating everyone to address real problems. IV.. Applying quality concepts The TQM approach relies on understanding all works as a process. The continuous improvement of work processes represents one of the key principles of TQM, but many quality practitioners have struggled when applying this basic principle beyond the manufacturing operations for which it was developed. When comparing various types of work processes, manufacturing stands out as unique in (1) its customers are isolated from production, () its outputs are tangible, and (3) its operations are highly repetitive. By contrast, nonmanufacturing processes differ in one or more of these key features. Customers are usually involved directly in the delivery of services, and the value 54

55 added by nonmanufacturing processes is often characterized as intangible. Further, some nonmanufacturing processes are repeated infrequently, and their outputs can be unique every time. Although it sounds simple, overcoming of these key differences can be very difficult. IV..a. Product vs Process The traditional method focused on the product or output. Quality improvement required tighter inspection of both incoming raw materials and outgoing finished products. With this approach, better quality was achieved at the expense of increased waste and higher costs. Quality is the responsibility of the QA department. Suppliers Traditional method Product management Customers INPUT I N S P E C T Work process I N S P E C T OUTPUT WASTE WASTE 3. Figure: Traditional method of product management This is contrasted with modern, process-centred quality improvement, in which better quality can be achieved without necessarily increasing costs. Improving quality through the process relies on an integrated approach along the entire customer-supplier chain. Process improvement is broader than ust quality assurance or inspection. It is broader than ust operations and production alone. Furthermore, process improvement can be applied to nonmanufacturing as well as manufacturing functions. Suppliers Modern method Process management Customers INPUT Work process OUTPUT INFORMATION People processes INFORMATION 4. Figure: Modern approach of process management The figure above shows the integrations of process management along the customer-supplier chain. At its core are the work processes themselves, the processes on which product quality improvement efforts had traditionally focused. Process management recognizes the value of the line workers who produce and deliver products / services and integrates them into the 55

56 improvement processes. Process management also recognizes the role of customers and suppliers and integrates systems that exchange information with them. Requirements Requirements SUPPLIER WORK GROUP CUSTOMER Product / Service Product / Service Feedback Feedback 5. Figure: Visualization of the work process This figure shows the basic elements of the work process model. The process begins with the flow of information in the form of requirements from the customer to define the characteristics of the desired output. The work group then integrates materials, equipment, methods, and people within an environment to produce these outputs. Customer satisfaction forms the feedback loop that drives corrective action to improve performance. This same model is mirrored upstream to suppliers. This basic model serves as the foundation for understanding and improving processes, regardless of the output. Quality process management offers an inherent competitive advantage over alternative practices, because it permits the improvement of quality while simultaneously reducing waste and costs. This advantage is available to all organizations, regardless of whether their outputs are manufactured products, marketing data, financial services, health care, engineering designs, or administrative services. Since quality improvement techniques were developed for manufacturing, successfully applying them to other functions requires identifying the inherent differences between the work processes and adapting the techniques to the desired function. IV..b. Manufacturing vs Nonmanufacturing processes Nonmanufacturing processes differ from their manufacturing counterparts in a number of ways. IV. Table: Differences between manufacturing and nonmanufacturing processes Comparing typical process attributes Manufacturing Nonmanufacturing Output properties tangible intangible or tangible Production and delivery separate integrated Customer interface focused: sales and marketing spread across line employees Feedback through process through customer 56

57 Organizational focus process efficiency customer relations Process ownership clearly defined multiple Process boundaries defined unclear Process definition documented unclear Control points defined none Quality measures established and obective subective Corrective action preventive reactive Consider the following differences: output properties that are tangible versus intangible; processes that are well documented versus unclear; quality measures that are obective and formally established versus subective; and work processes that are separated from customers versus subect to customer intervention. Quality improvement techniques were designed for applications typified by the first rather than the second characteristic in each pair. Thus, the following can be concluded about the attributes of applications that are best suited to classic quality improvement techniques: 1. Tangible outputs that permit direct physical measurement, determination of obective customer requirements, and translation into definitive engineering specifications.. Processes that are clearly documented, including raw material and equipment specifications, product movement, operating procedures, and performance standards. 3. Functional delineation of production, sale, and delivery that clarifies organizational boundaries, process ownership, and logical control points. Quality measures can be established and controlled within each step of the process. IV..c. Improving nonmanufacturing processes The preceding array of differences in the attributes of manufacturing and nonmanufacturing processes can be simplified to three key characteristics: coproduction, tangibility, and repetition. Coproduction: customer participation in your work Customer participation in the production of the output is the first key characteristic that distinguishes manufacturing from nonmanufacturing processes. Referred to as coproduction, this characteristic is found in most nonmanufacturing functions. As coproducers, customers frequently provide the material input into the service process. Customers supply the automobiles, appliances, buildings on which maintenance and repair service are performed. Customers supply their own bodies to the work processes in the health care, travel, and entertainment fields. Customers supply the money managed and the data processed in financial and information services. Customers supply the conceptual designs and real estate used in construction. The feature of coproduction brings the customer directly into the service process. As a result, the process itself represents an experience of vital interest, importance, and value to the 57

58 customer. Coproduction also exposes a broad array of employees in face-to-face contact with customers. In fact, coproduction influences the basic design of service processes. Tangibility and repetition Tangibility and repetition are the other two key characteristics that distinguish manufacturing from nonmanufacturing functions. Defining specifications and measuring conformance of tangible outputs are relatively straightforward procedures that can rely on physical characteristics such as size, weight, shape, volume, thickness, and material composition. Repetitive processes generate large quantities of data over relatively short periods of time. The combination of tangible outputs with repetitive processes facilitates the measurement, comparison, analysis, and systematic improvement of operations, as well as the inspection, grading, and sorting of outputs. Overcoming obstacles In order to apply quality tools beyond the manufacturing processes for which they were designed, it is necessary to overcome the obstacles imposed by the three key differences. In cases where the outputs are intangible, successful application requires the identification of appropriate measures, either subective or obective. In cases where the outputs are unique or on which customers are coproducers, successful application requires the clarification of the underlying work processes that are repeated. And in other cases, application requires taking both of these steps. IV..c.1 Measurement at three levels The measurement of performance also becomes more difficult as business characteristics diverge from those of the ideal applications. This obstacle to quality improvement can be overcome by taking advantage of measurement at three levels: process, output, and outcome. Unfortunately, when performance parameters are difficult to define or measure, the tendency has been to substitute other measures without recognizing that they represent different levels. 1. Process measures define activities, variables, and operations of the work process itself. Measures at the process level also include the measuring the products and services that suppliers input to the work process. These measures represent parameters that directly control the integration of people, materials, methods, machines, and the environment within the work process. Although frequently understood and used in manufacturing operations, measures at the process level are often absent from other functions. Understanding and applying measures at the process level help to predict the characteristics of the outputs before they delivered to customers.. Output measures define specific features, values, characteristics, and attributes of each product or service. Furthermore, output measures can be examined from two sides. One side represents the output characteristics desired by the customer, and the other side represents the output characteristics actually delivered by the process. The former is referred to as requirements, expectations, or the voice of the customer. The latter is referred to as capability or the voice of the process. Measures at the output reveal what is delivered to customers. 58

59 3. Outcome measures define the ultimate impact of the process on the customer and are dependent on what the customer does with the product or service. Although this is the most important level, outcome measures are the most difficult to define and analyse because they are confounded by the customer s work process. To simplify the array of performance measures, customer satisfaction can be used as the key measurement of outcome. Measures at the outcome level reflect the impact of outputs on the customer s processes and can only be determined after the product has been delivered or the service provided. Training division example of measurement levels Participants in this example process include Carl Thoren, Michael Roberts and the company s 40 service representatives who attend the training program and apply its teachings. Thoren and Roberts are both assigned to the training division. Thoren was responsible for developing and designing the training program, and Roberts then delivered the course to the service representatives. Carl Thoren is an instructional designer in the training division of the customer service department. He developed a four-hour training session to improve service representatives ability to answer maintenance questions on a new household appliance. Although Michael Roberts, the classroom instructor, is the direct customer of Thoren, attention must be paid to the needs of customers all along the chain, and especially to the ultimate end users. Activities in Thoren s work included identifying the training needs of the service representatives, developing the required training program, preparing an instructor s guide, and producing classroom material. Measures at the process level of Thoren s work are somewhat vague and only partly defined. One obvious measure is the length of time required for each activity, but the training division has not clarified any others. Instead, it focuses on output measures and develops internal process measures as needed to understand and improve performance. Outputs from Thoren s work include the design of the four-hour course, a seventy-two-page instructor s guide, a twenty-six-page participant s manual, and thirty-four visual aids for use in the classroom. These items are the tangible outputs from Thoren, but their measurement does not represent all of the most critically important parameters. Other output measures relate to the intangible characteristics required for these outputs, such as accuracy, completeness, clarity, and ease of use. Michael Roberts, a classroom trainer, delivered ten sessions of the course that Thoren designed to 0 customer service representatives. The inputs into Roberts work process include outputs from Thoren. Furthermore, outputs from Roberts work are outcomes of Thoren s process. Activities within Roberts work included lesson planning, classroom preparation, and representation of the actual course. As with Thoren, a complete set of process and output measures has not been defined for Roberts. As an outcome of the training process, service representatives are now able to answer twentyfour additional customer inquiries each week. These additional calls are outputs of customer 59

60 service department. One outcome of these calls is an increase in sales of the new product by $60000 per month. Another outcome is the improvement of customer satisfaction attributable to the new skills of the service reps. V. Table Role Instructional designer Classroom instructor Service representative Example measures Process: time to develop a new course Output: 7-page guide; 6-page manual; 34 visual aids; 4-hour design Outcome: 10 classes delivered Process: time to plan lessons Output: 10 classes delivered Outcome: 5 skills gained by 40 reps Process: 5 skills gained Output: 4 contacts added weekly Outcome: $60000/month higher sales; happier customers The figure shows how the level represented by any particular measurement is defined by its frame of reference. To see how this works, examine one measurement from three reference points. Training 40 service reps are an outcome of Thoren s work as the course designer. This same measure is an activity in the daily process of the service representatives. This moving frame of reference shows how the measurement level is translated as products/services flow along the customer-supplier chain. Even though its set of measurements is incomplete, the training division is able to improve its performance in the development and delivery of training programs through use of data from all three levels. By measuring appropriate parameters and results, experiments can be designed to test performance improvement hypotheses. In order to apply the quality improvement concepts and tools beyond the manufacturing processes for which they were designed, the obstacles imposed by three key differences coproduction, tangibility, and repetition must be overcome. This can require clarifying the underlying work processes and/or identifying appropriate measurements. Capturing the full set of performance measures is often an idealized dream. Complete measurement is not always practical, particularly when working outside of manufacturing and with intangible, subective performance characteristics or unique outputs or processes in which customers are coproducers. Use of measures at the process, output, and outcome levels provides a framework for understanding performance in dimensions beyond those that are obectively measurable. This approach can be thought of as an extension of high-school algebra with three equations and 60

61 three unknowns. The unknown values can be postulated and tested, derived through a systematic application of formulas, or guessed through trial and error. By recognizing the key differences among various types of work processes, and by overcoming the obstacles imposed, the competitive advantage of TQM can be captured by all organizations. IV.3. Customer focus IV.3.a. Identifying the customer The key to gaining a long-term competitive advantage is to continually meet customers expectations in ways that they recognize as adding value. To achieve this advantage, it is necessary to know who your customers are, what they expect, and how well the organization and its competitors are performing from the customers point of view. Focusing on the customer, then, is the first of three basic quality management principles. Who is the customer? We need to know to whom we must talk to assess the level of customer service that we are providing, and we need to identify what we must do in the future to improve. Obviously, to do these things, we need to identify specific people with whom we work within each client organization so that we can become more precise about what we must do to better meet their needs. If the task of identifying external customers is difficult, the task of identifying fellow employees as potential customers may be even more complex, especially when dealing with professional or senior-level employees. The concept of internal customers is significant because it dramatically makes the case that an organization cannot successfully meet the needs of its external customers if each output passed between employees within the company is deficient. Mathematically it is easy to see that the external customers requirements cannot be achieved 100 per cent of the time if each handoff is less than 100 per cent. For example, a chain of only three internal suppliers who each meet 90 per cent of their internal customers requirement may result in a 73 per cent effective delivery from the organization to an external customer (0,9*0,9*0,9=0,73). If however, we can identify the person to whom we pass the output of our work, then we can secure from that person a list of needs, expectations and requirement that we as the supplier must meet. Several guidelines on correctly identifying outputs are listed here: Things that you supervise or approve but that are actually produced by others. Examples are budgets that are developed by a staff but approved by a senior manager. The budget is the output of the financial staff; the senior manager is a co-supplier. Goals or outcomes for the organization. Examples are profits, customer satisfaction, revenues, and market share increases. Like morale, these outcomes are the result of a 61

62 number of outputs produced by a number of individual people and are usually beyond the capability of a single individual to impact or influence. Steps in the work process. Generating a work plan or schedule is a step that an individual takes as part of the effort to complete a proect but is not an output that is passed along others. The overall function described by an individual s ob title or responsibility. The manager of the computer service department, in all likelihood, does not actually repair computers, even though he or she is certainly responsible for the timely, accurate repair of malfunctioning computers. Rather, the manager generates staffing plans, forecasts, training plans, budgets, and a number of outputs that are passed to others. The employees in the department actually perform the repair service because they visit the customers site, diagnose the problems, and effect the repairs. What outputs are: The specific products or services that you produce, as part of your work process, and that you pass to others, who, in turn, use them in their work process. What do customers want? Once we have identified the customers for the output we produce, we must determine from the customer what he or she expects, requires, and needs from us, the supplier. We must accept the concept that quality is defined by the customer and meeting the customer s needs and expectations is the strategic goal of TQM. Sometimes, the customers are unsure of their precise needs and expect the supplier to assist in the clarification of their requirements. This situation can be turned into an advantage, since it creates opportunity to develop a partnership between customer and supplier that is beneficial to both. In many TQM systems, the customer bring the supplier in early in the new product development cycle to take advantage of the supplier s specialized skill. Customer satisfaction The obective of implementing this disciplined approach of determining outputs, identifying customers, and identifying requirements is to enhance the supplier s ability to meet the customers needs and expectations and thereby increase customer satisfaction. Clearly, as we discussed it earlier, the only viable means for organizations to achieve their obectives is to meet the requirements of their customers by continually improving work processes. Customer needs and expectations are constantly escalating as customers have their requirements met and learn of new possibilities from competitors. Any company that is too internally focused and not mindful of the dynamics of the marketplace will eventually lose market share. The task, then, is to pursue customer satisfaction in an organized, disciplined manner. A word of caution While internal suppliers are focusing on meeting the needs of their immediate customer (the next person to whom they are providing a product or service), they should also be mindful of 6

63 the final end user, the external customer who will eventually consume the product or service and pay the bills. This awareness provides a reality check on an organization, in that all individuals in the organization must understand how their products and services are being accepted in the marketplace. Later, when customer satisfaction data are reported, everyone sees his or her value in contributing to the success of the organization. Summary Identifying our customer is best started by determining the output we are producing and then identifying the individual to whom we will pass our output. If we cannot identify an individual who receives our output in his or her work process, we must stop as we are producing a scrap! Once we have identified our customers, and they agree that they are our customers, then we must gather and clarify their requirements and build a complete understanding of what they want, need and expect of us. If we go further as suppliers, and identify latent requirements (features that our customers may not have been aware of but really want), then we begin to become a more valued supplier and possibly, in the eyes of the customer, a partner. Having identified the customers requirements, meeting those requirements every time 100 per cent of time is the essence of achieving quality. IV.3.b. Understanding customer expectations What do customers want? In essence, they want their expectations to be met completely and consistently. Consumers tend to perceive quality of a service by comparing the actual service experienced to what their expectations were before purchasing it. Service is udged to be unsatisfactory when expectations are not met, satisfactory they are met, and more than satisfactory when they are exceeded. Successful organizations are able to diagnose the full set of customers expectations and satisfy them completely, every time. World-class organizations have the uncanny ability of understanding implicit and latent requirements. These latent requirement are features that customers want but do not know are available and hence are unable to articulate in discussions with their suppliers. Once the customers are identified in the targeted market niche for a particular product or service, their expectations can be determined by answering key questions: 1. What product/service characteristics do customers want?. What performance level in needed to satisfy their expectations? 3. What is the relative importance of each characteristic? 4. How satisfied are customers with performance at the current level? Finding answer to these questions begins postulating a set of features and characteristics that customers might want. This list of hypothetical criteria should next be tested by directly asking customers. The process of learning customers needs, requirements, expectations, and level of satisfaction is commonly called listening to the voice of the customers. 63

64 IV.3.c. What are the characteristics of a quality service? Customers expect to receive value in the products or services they purchase or use. In this context, value can be defined as the relationship between what customers get in exchange for what they give. Although this has often been considered as a trade-off between price and quality, a detailed analysis shows that far more is involved. For example, what about convenience? A customer may sacrifice convenience in search of lower price or higher quality. Therefore, in a global sense, the characteristics of quality service are those that enable customers to feel they have made a fair exchange and received value. Scores of models and frameworks have been developed to help clarify how customers define quality or value. In this chapter we introduce five models: 1. Faster, better, cheaper. Eight dimensions of quality 3. Ten determinants of service quality 4. Five rater criteria 5. Compendium of quality characteristics Faster, better, cheaper Value can be viewed most simply as getting things that are faster, better, and cheaper than available elsewhere. The first dimension, time, represents how quickly, easily, or conveniently a product or service can be obtained. The second dimension, cost, equates to how expensive the item is. The third dimension, quality, is the most difficult one to characterize. To begin clarifying this complex dimension of quality, it is useful to subdivide it into two maor sets. The first is product quality and includes the tangible attributes that are retained by the customer. The second is service quality and includes the characteristics observed or experienced by the customer during the transaction. As explained previously, customers are often involved directly in the service process. Therefore, some portion of the value in service consists of how it is delivered. Service quality is udged by the customer during the service. In air travel, for example, service quality involves the behaviour of the cabin attendants and the comfort of the flight. The product represents what is delivered and can be measured after the service has been performed. Continuing with the air travel example, product quality includes safe delivery of passengers and their luggage to the expected destination at the scheduled time. Eight dimensions of quality Garvin has defined eight dimensions that can be used at a strategic level to analyse quality characteristics. Some of the dimensions are mutually reinforcing, whereas others are not improvement in one may be at the expense of others. Understanding trade-offs desired by customers among these dimensions can help build a competitive advantage. 64

65 1. Performance: The product s primary operating characteristic. For example, a performance of an automobile includes traits such as acceleration, handling, cruising speed, and comfort. Performance of an airline includes on-time arrival.. Features: Secondary aspects of performance. These are the bells and whistles that supplement the basic functions. Examples include free drinks on planes and sunroofs on cars. The line separating primary characteristics from secondary features is often difficult to draw. Further, customers define value in terms of flexibility and their ability to select among available features, as well as the quality of those features. 3. Reliability: Probability of successfully performing a specified function for a specified period of time under specified conditions. Reliability of durable goods is often measured as the mean time to first failure or mean time between failures. These measures, however, require a product to be in use for a specified period of time and are not relevant in the case of products and services that are consumed instantly. 4. Conformance: Degree to which a product s design and operating characteristics meet established standards. Although this is sometimes defined as conformance to requirements, a sounder analysis will be obtained by examining each characteristic s divergence from its target value. This more robust measure of conformance is built on the teachings of a Japanese statistician, Taguchi. 5. Durability: A measure of product life. Durability can be defined as the amount of use obtained from a product before it deteriorates to the point that replacement is preferred over repair. Durability is closely linked to both reliability and serviceability. Consumers weigh the expected costs of future repairs against the investment in and operating expenses of a newer, more reliable model. 6. Serviceability: The speed, courtesy, competence, and ease of repair. The cost of repairs includes more than the simple out-of-pocket costs. Serviceability covers this full dimension by recognizing the loss and inconvenience due to downtime of equipment, the nature of dealings with service personnel, and the frequency with which repairs fail to correct the outstanding problems. 7. Aesthetics: How a product looks, feels, sounds, tastes, or smells. Aesthetics is largely a matter of personal udgement and a reflection of individual preference; it is highly a subective dimension. 8. Perceived quality: Reputation. Consumers do not always have complete information about a product s or service s attributes; indirect measures or perceived quality may be their only basis for comparing brands. Ten determinants of service quality Research by Berry, Parasuraman and Zeithaml in the early 1980s provides a strong foundation for understanding the attributes of service quality. Through interviews with business executives and customer focus groups, Berry et al identified ten determinants of service quality: 1. Reliability: Consistency of performance and dependability; performing the right service right the first time; honouring promises; accuracy.. Responsiveness: Willingness or readiness of employees to provide service; timeliness. 65

66 3. Competence: Possession of the skills and knowledge required to perform the service. 4. Access: Approachability and ease of access; waiting time; hours of operation. 5. Courtesy: Politeness, respect, consideration, and friendliness of contact personnel. 6. Communication: Keeping customers informed in language they can understand; listening to customers; adusting language to different needs of different customers; explaining the service itself, how much it will cost, and how problems can be handled. 7. Credibility: Trustworthiness, believability, honesty; company reputation; personal characteristics of personnel. 8. Security: Freedom from danger, risk, or doubt; physical safety; financial security; confidentiality. 9. Understanding the customer: Making the effort to understand the customers needs; learning the customer s specific requirements; providing individualized attention; recognizing the regular customer. 10. Tangibles: Physical evidence of the service; physical facilities; appearance of personnel; tools or equipment used to provide service; physical representation of the service; other customers in the service facility. Five rater criteria As an outgrowth of their work in developing ten determinants of service quality, Berry et al distilled their list to five broader categories: 1. Reliability: Ability to perform the promised service dependably and accurately.. Assurance: Knowledge and courtesy of employees and their ability to inspire trust and confidence. 3. Tangibles: Physical facilities, equipment, and appearance of the personnel. 4. Empathy: Caring, individualized attention the firm provides its customers. 5. Responsiveness: Willingness to help customers and provide prompt service. IV.3.d. Compendium of quality characteristics By synthetizing the four previously described models, we are able to build a single comprehensive set of quality characteristics. This compendium integrates Garvin s eight dimensions and Berry et al. s ten determinants into the macroset of faster, better, cheaper. However, rather than distinguishing between elements of product and service quality, this set reclassifies quality into two components: deliverables and interactions. The deliverables define what attributes are provided to customers. Their interactions characterize how behaviours and styles impact on customers while they are experiencing the service process. Both elements apply to all products and services, and this description can be used to confirm that all maor characteristics have been considered. 66

67 Faster Better Cheaper VI. Table Deliverables Availability Convenience Performance Features Reliability Conformance Serviceability Aesthetics Perceived quality Price Interactions Responsiveness Accessibility Reliability Security Competence Credibility Empathy Communications Style IV.3.e. What performance level is needed to satisfy expectations? Service quality features are often measured in subective, qualitative terms based on observations and comparisons. Product quality is usually measured in absolute, quantitative terms based on physical or chemical properties. VII. Table Attributes Product quality Obective Tangible Measures in absolute terms such as physical or chemical properties Examples Size, weight, volume, delivery time, material, count, colour Service quality Subective Intangible Observed in comparative terms relative to expectations or prior experience Attitude, cooperation, reputation, friendliness courtesy, attentiveness, dependability, In order to measure quality, we should begin by defining the characteristics that are important to customers. The next step is to determine how to obtain meaningful data. This approach is the same whether the feature is an element of product quality or service quality. 67

68 IV.3.f. Implicit, explicit and latent requirements The characteristics of products and services expected by customers can be viewed as a progressive hierarchy of three levels: base expectations, specifications/requirements, and delight. These three levels are often referred to implicit, explicit and latent. Delineation in this fashion provides a model for understanding the level of performance needed to satisfy customers expectations. Level 3 Level Value-added characteristics and features that customers did not expect Options and tradeoffs available for selection by customers Delight! Specifications and requirements Level 1 Minimum performance levels always assumed present Base expectations 6. Figure The customers base expectations form the lowest level. These are the characteristics (or levels of performance) that are always assumed to be present, and if they are missing, customers will always be dissatisfied. When buying a car, customers assume certain basic attributes will be included without any discussion with the dealer, e.g. cornering stability, collision protection, motor vehicle inspection, rustproofing. The next level is presented by the specifications and requirements that are visible to customers and are actively involved in their selection process. This is the level at which explicit trade-offs are made and terms negotiated. Staying with cars as the example, this level of performance covers the features that are advertised and those that are discussed with dealers. They include such characteristics as fuel economy, horsepower, acceleration, color, number of seats, body style, interior décor, price, delivery, and warranty protection. The highest level of performance is represented by the value-added features that the customer did not even know about but is delighted to receive. Performance at this level can be described as delivering all of the explicit requirements as well as latent ones. Latent requirements are real but are not visible or obvious to the customer. Some people define quality as simply conformance to requirements, whereas others argue that this is an incomplete definition. The latter view is particularly evident when your competition is able to ascertain and provide features that exceed expectations and delight customers. Note, however, that performing at the level of delight is not simply a matter of delivering more than specified. 68

69 Application of the three expectation levels Recognizing the performance level of any particular product/service characteristic offers three advantaged. First, it clarifies which characteristics should be discussed in detail with current or prospective customers. Second, it helps to predict how the level of customer satisfaction might respond to a change in characteristics. Third, it indicates possible future trends in expectations. This knowledge can be applied as follows: Focus discussions with customers around characteristics that represent the conspicuous specifications and requirements (level ). Customers take the basic expectations (level 1) for granted and assume that knowledgeable suppliers know this. Similarly, customers cannot be expected to appreciate level 3 features until they are experienced. Meeting level 1 expectations is a defensive requirement that at best will help void creating dissatisfied customers. High levels of customer satisfaction can be expected by consistently delivering the implicit base expectations (level 1) and every explicit specification (level ), as well as including the value-added features (level 3) that delight customers. Customers expectations will escalate, and performance levels will migrate down through the hierarchy over time. Responding to complaints: the hidden level There is one more level of performance a hidden one that is only discovered after unhappy customers bring problems back to their suppliers. One course of action available to a dissatisfied customer is to accept the problem begrudgingly and not complain to the supplier. Unfortunately, this customer is likely to complain to friends and influence them to avoid buying from this supplier. The other course of action for a dissatisfied customer is to seek compensation from the supplier (replacement, refund, repair, etc.). By providing convenient avenues for complaining, dissatisfaction will be reduced. Depending on how the problem is handled, three outcomes are possible: 1. If the supplier s corrective action does not meet the customer s expectations, the original feeling of dissatisfaction will be exacerbated.. If the supplier s corrective action meets the customer s expectations, the original feeling of dissatisfaction will likely be neutralized. 3. If the supplier s prompt, effective, complete, and courteous action exceeds the customer s expectations, the original feeling of dissatisfaction can potentially be converted into delight. What is the relative importance of each characteristic? The relative importance of each quality characteristic varies in relation to the specific expectations of the customer at any particular time. 69

70 As evidence of variability in the relative importance of characteristics, consider something as simple as your own purchase of groceries. Under one set of circumstances, the quality of the food might be paramount to you, and at another time, the ambiance of the supermarket could be of maor importance. At another time, convenience might be the critical factor, as in the case of running out of soda for a party you are giving late at night on a holiday weekend. Finally, there may be situations when price alone dictates your shopping decisions. Rather than attempting to prioritize customers needs on a global basis, we suggest that the relative importance of performance characteristics should be determined with customers for each product and service, and then updated frequently. The relative rating of importance can be used to guide the allocation of resources for quality improvement efforts. Unfortunately, building an understanding of how customers rank the relative importance of quality characteristics is not a simple task and can be confounded by customers lack of perspective. For example, how important is safety in selecting a flight on a commercial airline? Although safety might be the single most important characteristic, the maority of passengers take safety for granted. Suppliers who understand the three levels of quality explained earlier can avoid contaminating their findings by recognizing that characteristics like flight safety are among customers base expectations. As long as base expectations are satisfied, they will not command the attention or interest of customers. An airline with a pattern of safety incidents will be shunned by passengers, since it has violated their implicit expectations. Another degree of complexity occurs because customers will seemingly change their priorities overnight. Whereas the types of features and characteristics expected by customers might remain stable for long periods of time, their relative importance and level of expectation will appear to change with headline news, weather, competitor s advertising, and technology advances. The word appear is emphasized to draw attention to the need to differentiate between fads and fundamental beliefs, or between topical headlines and underlying values. When asking customers about the importance of a characteristic, their reply might be confused with their level of satisfaction. IV.3.g. Quality Function Deployment We have offered a process for listening to the voice of the customer in order to learn expectations. Once those expectations are understood, the next step is to translate them into product and service specifications. The concept of this translation was introduced in the previous chapter through the establishment of two sets of measures at the output level: customer requirements and process capability. However, the translation between this pair of measures is sometimes complicated. Help is available through a technique known as quality function deployment (QFD). QFD can be used to translate customer requirements into appropriate technical specifications. The technique helps in defining units of measurement, and it provides a framework for evaluating trade-offs among various combinations of design features. 70

71 QFD was formalized in 197 at Mitsubishi s shipyard in Kobe, Japan. It has been used to reduce development time for introducing new products and to reduce disruptive and expensive engineering changes. The heart of QFD is a large matrix that relates what customers require with how products and services will be designed and produced in order to satisfy those requirements. IV.3.h. Summary Customers tend to perceive the quality of services by comparing the actual level experienced to their expectations. The process of satisfying customers therefore begins by fully understanding their expectations. This process can be referred to as listening to the voice of the customer and requires learning what features and characteristics customers want, the performance level they expect, the importance they attach to each characteristic, and how satisfied they are with performance at the current level. Faster, better, cheaper is the most fundamental set of characteristics to describe how customers define value and provides a springboard for understanding other, more complex models. The array of quality features can be subdivided into the categories of product and service or of deliverables and interactions. All products and services contain some elements of service quality. For example, an automobile includes elements of service quality in both the purchasing process and the inevitable maintenance and repair. The value added by the dealer in the purchasing process includes attributes as simple as availability of the desired vehicle in inventory and convenience of the location. The overall quality perceived by the customer is also impacted by interactions with the sales staff. The level of performance expected by customers can be measured, although some data will be obtained through subective comparisons instead of obective measures against absolute standards. The three-level hierarchy helps in the analysis of performance measures. At the base level are the implicit expectations, which must never be violated. The intermediate level contains the explicit specifications, which can be discussed and negotiated. The highest level addresses latent requirements, which might not be evident to customers but which, when provided, will lead to their delight. Furthermore, customers expectations escalate: features that currently delight customers may eventually become embedded in their base expectations. Building an understanding of customer satisfaction and the relative importance customers attach to each quality characteristic is a complex and difficult task. For example, a customer survey might tend to underrate the importance of a vital performance characteristic because customers are satisfied with its current delivery and distracted by other irritants. Knowledgeable suppliers are able to discern customers true underlying values. Once understood, customers expectations must then be translated into product and service specifications. QFD is a technique for accomplishing this translation. 71

72 IV.4. Mechanisms for understanding customers Understanding customers expectations is a prerequisite for improving quality and achieving full customer satisfaction. Significant research has been conducted in understanding customers expectations of products, but what has been done for service? This chapter provides a two-dimensional framework for defining the mechanisms available for understanding customers expectations. The first dimension classifies the approach initiated by the supplier, moving from reactive to proactive. The second dimension indicates the level of understanding likely to be achieved by each mechanism. Although this approach was developed for service groups within large organizations, it can be used to understand the requirements of any type of customer. The reactive mode of waiting to hear complaint reveals only a minimal understanding of the customers expectations. A better understanding will be gained by initiating more active approaches to listening to customers. Examples include help desks, hot lines, analysis of sales data, obtaining feedback from customer representatives, and conducting informal surveys. Full understanding, however, can best be achieved through mechanisms specifically initiated to listen to customers. These approaches include personal interviews, focus groups, structured surveys, and benchmarking. The key is to build a profound understanding of your customers needs and expectations. What are the distinguishing characteristics of the products and services that are important to the customers? How satisfied are the customers with the product and service characteristics delivered by the organization? By the competitors? By the best available technology? This chapter concludes by showing how one organization progressed through the framework to build its level of understanding of its customers needs. IV.4.a. Development of the framework As shown by the work process model introduced in a previous chapter, two sets of information flow from customers to suppliers: (1) requirements a description of the product or service expected by customers before it is produced or delivered; () satisfaction feedback on what the customers liked or did not like about the work that was performed form them. The figure shows a two-dimensional framework that has been developed for categorizing mechanisms commonly used to gather information from customers. The first dimension displays the degree of activity initiated by the supplier. The second dimension maps the level of understanding that might be attained. Although actually following a continuum, the framework is shown with three discrete levels to simplify its explanation. Level 1 The reactive mode is likely to reveal only a minimal understanding of the customers expectations. Approaches here are primarily ones of gathering complaints. Examples include 7

73 logging irate phone calls from dissatisfied customers and responding to letters to employees bosses. The effectiveness of learning customers expectations and satisfaction through these reactive approaches is impeded by four factors. First, the data are obtained from a biased, nonrepresentative set of customers who are sufficiently unhappy to initiate complaints, frequently in the form of a solution. Second, only limited samplings of these unhappy customers actually volunteer information to suppliers through these mechanisms. Third, systems to synthetize and analyse these data are often lacking. Fourth, employees receiving information through these channels are often distracted by the need to fix their customers problems or defend themselves against the accusations. However, actions can be taken to minimize these four shortcomings and maximize the value of level 1 mechanisms. Customers failure to complain is often the result of their not knowing how to complain or where to direct a complaint, or feeling that complaining ust is nor worth their efforts. Organizations can reduce these barriers through several approaches. They can display clearly the avenues available to customers, including toll-free phone numbers. They can make complaining worthwhile to their customers by responding to their problems and trying to create advantage out of adversity. In addition, employees receiving complaints can be supported and rewarded for their role to offset the distaste usually inherent in this task. Training programs and user-friendly systems are examples of supporting systems. Level A higher level of understanding will be gained by initiating active approaches to listen to customers. Mechanisms at level are defined as those approaches that communicate with customers but have listening to customer expectations as their secondary obective. Their primary obective often includes answering customers questions or selling more/new products. Although they are more effective than the previously discussed reactive approaches, the ability of level mechanisms to capture customers views is compromised because these mechanisms are designed mainly to satisfy their primary obectives. Level examples include help desks or hot lines, analysis of sales data, feedback from customer representatives, and unstructured surveys. As with level 1, actions can be taken to maximize the effectiveness of level mechanisms. Additional resources can be allocated to gather and analyse information obtained, and systems can be installed to minimize the effort required to satisfy the secondary obective. Employees involved in these types of functions can be coached and trained to gather customer data in addition to fulfilling their primary obectives, and they can be rewarded for doing so. Level 3 Full understanding of customers expectations can only be attained through the use of mechanisms specifically designed to extract this information. Approaches at this third level include personal interviews, focus groups, and designed surveys. Another mechanism at this 73

74 highest level is the mystery hopper, which enables suppliers to take the viewpoint of their customers by planting employees in positions to use their own services. Focus groups can offer insights that are impossible to capture through surveys or even through interviews with individual customers. Interviewing and surveying former customers is a frequently neglected level 3 mechanism. Unlike communications with prospective or current customers, communications with former customers can provide specific, obective data on the shortcomings of the products and services. High Maximum level of understanding Low Level 3 Personal interviews, focus groups, designed surveys, benchmarking, mystery shopper Level Service desk, networks, hotline, sales data analysis, unstructured surveys, customer reps Level 1 Unsolicited complaints Full customer understanding Reactive Approach 7. Figure Proactive IV.4.b. Benchmarking to understand expectations Benchmarking is the process of continually researching for new ideas and methods, practices and processes, and either adopting the practices or adapting the good features, and implementing them to obtain the best of the best. Customers expectations are influenced by what they learn through comparing shopping. Although they might settle for your particular product or service now, what is to stop them from finding something better next time? Benchmarking enables suppliers to establish performance targets based on best possible practices and continuously improve toward those targets. Benchmarking is the search for best practices. It is the search for what is best and for an understanding hoe the best is achieved. There are four distinct types of benchmarking: 1. Internal: One of the easiest benchmarking investigations is to compare operations among functions within your own organization. This type of investigation is applicable to multidivisional or international firms. Data should be readily available and reportable on a consistent basis.. Competitive: Direct product or service competitors are the most obvious to benchmark against. Although this information may be difficult to obtain, its value is high. 74

75 3. Functional: It is not necessary to limit comparisons to direct competitors. In fact, a narrow focus may risk missing potential breakthroughs. Functional benchmarking investigates leaders in dissimilar industries. The relevance of comparison is maintained by defining the performance characteristics that must be similar to your own functions. 4. Generic: Some business functions and processes are the same regardless of dissimilarities across industries. Generic benchmarking is the purest form of benchmarking, in that it may uncover methods that are not implemented in the investigator s own industry. It extends functional benchmarking by removing the constraints imposed by limiting the investigation to practices with similar characteristics. It holds the potential for revealing the best of the best. It requires broad conceptualization. Although it is the most difficult type of benchmarking to use, it probably provides the highest potential payoff. IV.5. Managing key processes One key to breaking the view of fire fighting and building a long-term perspective is to strengthen the underlying business processes instead of addressing each specific output and deviation. IV.5.a. What is a work process? All products and services are delivered through work or business processes. Before explaining how to manage and improve processes, it might be useful to review the basic terms and concepts that define work processes. A process can be defined as the sequential integration of people, materials, methods, and machines in an environment to produce value-added outputs for customers. A process converts measurable inputs into measurable outputs through an organized sequence of steps. Four groups of people are involved in the operation and improvement of processes: 1. Customers: the people (or person) for whom the output (product or service) is being produced. Customers are the people who will use the output directly or who will take it as input into their work process.. Work group: the people (or person) who work in the process to produce and deliver the desired output. 3. Supplier: the people (or person) who provide input to the work process. The people in the process are in fact the customers of the supplier. 4. Owner: the person who is responsible for the operation of the process and for its improvement. Customers are the ones, who define the output desired from the process. This is accomplished through two broad categories of information that flow from customers to the work group. The first category consists of the requirements: a description of what the customers need, want and expect. The second category of communication is feedback: an explanation how well the output was delivered in comparison to the customers expectations. This feedback signal is 75

76 vital for the improvement of the process for future operations. The flow of information and products with suppliers appears as a mirror image of the process used to connect the work group to its customers. IV.5.b. Six ingredients of process management There are six ingredients that are essential for process management: 1. Ownership: assign responsibility for the design, operation, and improvement of the process.. Planning: establish a structured and disciplined approach to understand, define, and document all maor components in the process and their interrelationships. 3. Control: assure effectiveness: all outputs are predictable and consistent with customers expectations. 4. Measurement: map performance attributes to customers requirements and establish criteria for the accuracy, precision, and frequency of data acquisition. 5. Improvement: increase effectiveness of the process by permanently embedding identified improvements. 6. Optimization: increase efficiency and productivity by permanently embedding identified improvements. These six ingredients are fundamental to the successful management of any type of process. These ingredients are needed for the work processes that produce and deliver products and services to customers, for the processes that clarify requirements and satisfaction along the customer-supplier chain, and for the processes that support employees in their obs. IV.5.c. Defining key processes Every organization can identify the key processes on which its success depend. The way in which businesses define their organizational structure provides an inherent advantage for the ownership and planning of processes that exactly align within formal organizational boundaries. This generalization contains both good news and bad news. The good news is that the processes that are complemented by the organizational structure tend to have a history of successful control, measurement, improvement and optimization. The bad news is that the processes that cross organizational boundaries are likely to be in poor conditions. For those interested in more bad news, the performance of the processes that cross organizational boundaries and are in poor condition may be more important to customers than the performance of some processes that fit neatly within administrative boundaries. Furthermore, this importance to the customer may not ust involve the outputs from these processes, particularly in service functions, where the customers are active participants in the delivery process. The good news behind all this is that the greatest short-term improvement opportunities are likely to be found among these cross-functional processes. Immunization against unhealthy cross-functional processes can be gained by systematically searching for key business processes, regardless of organizational boundaries. 76

77 There are six questions to help in the identification of key processes that have the greatest impact on customers: 1. Which products and services are most important to customers?. What are the processes that produce these products and services? 3. What are the key ingredients that stimulate action in the organization, and what are the processes that convert these stimuli to outputs? 4. Which processes have the highest visibility with customers? 5. Which processes have the greatest impact on customer-driven performance standards? 6. Which processes do performance data or common sense suggest have the greatest potential for improvement? Answers to these questions will differ for different types of organizations. No universal formula or prescription has been found. Key processes at a hospital will be different from those at a nationwide fast-food franchise or at an aircraft manufacturer. Nonetheless, all will share some elements of commonality. Once the key processes have been identified, their systematic and continuous improvement can begin. First, assign ownership-responsibility for the design, operation, and improvement of overall system. Next, plan a structured and disciplined approach to understand, define, and document all maor components and their interrelationships. Then, follow the processimprovement road map presented in the next chapter. IV.5.d. Summary The importance of ownership should be reemphasized, particularly the ownership of processes that cross functional boundaries. Process ownership can be identified through two characteristics authority and responsibility. The owner of any process is the person at the lowest level who has the authority to implement changes and who is responsible for the consequences of these changes. Many organizations have traditionally focused managers attention on functional goals and fixing, maintaining, and running their respective operations. Instead, process owners should focus on the overall system that delivers products and services to customers and on improving the performance of that system. Ideally, compensation and promotion of process owners will include consideration of their ability to continuously improve performance of their processes. TQM achieves quality improvement through prevention and the systematic improvement of key processes rather than through fire fighting and focusing on near-term results. Key processes can be defined as the ones that have a material effect on the success of the organization. Since organizations are often structured around their areas of specialization, key processes sometimes cut across functional boundaries and receive less attention than they deserve. There are three typical failings that most frequently prevent organizations from systematically improving business processes: (1) the failure to identify key processes and assign ownership; () the failure to apply a robust improvement approach that builds an understanding of fundamental root causes of problems, and (3) the failure to measure the right things. 77

78 IV.6. Six steps to process improvements The six-step process improvement model was prepared specifically to bridge the gap between manufacturing and nonmanufacturing applications of quality methods. It can serve as a universal road map for all applications. It begins by identifying the outputs, customers and the work processes that produce these outputs. The methodology continues by closely examining customer requirements and defining gaps between them and the capabilities of the work processes involved. It next stimulates exploring and analysing these processes to understand the root causes underlying the gaps. It sparks development of new outputs and processes and requires that these new ideas be tested with data. Appropriate changes are implemented, their effects are evaluated, and the cycle is repeated to secure continuous improvement. The six-step process offered in this chapter was developed through years of frustration and failure, and its application requires discipline. Simpler approaches have been tried and have been successful in some applications, but they have invariably encountered difficulties when generalized to different types of problems. The six-step process has been designed as a universal approach. 1. Define problem. Identify and document process 3. Measure performance 4. Understand why 5. Develop and test ideas 6. Implement solutions and evaluate 8. Figure The six-step improvement model introduces a systematic approach for applying quality management to any type of process. It can be applied to any operation: information systems, marketing, finance, administration, R&D, engineering, service, or manufacturing. It can be applied to any system: those that exchange information with the customer, those used to produce and deliver products and services, and those that create the work environment. Universal application is possible because the approach helps build a fundamental understanding of the business process before attempting to improve them. Continuous 78

79 improvement requires knowing what these processes are, why they are performing the way they are. This profound knowledge is built through application of the first four steps. In some applications, one or more of these first four steps can be shortcut because they have already been accomplished through the normal course of business. This does not mean that these steps can be skipped, merely that their requirements are satisfied through existing systems. Step 1: Define the problem in the context of the process Unlike manufacturing, service providers or knowledge workers sometimes do not recognize that they are performing within a business system. The process improvement model begins by clarifying which systems are involved, so that efforts can focus on processes, not outputs. Specific activities within the first step are: 1. Identify the output.. Identify the customers. 3. Define the customers requirements. 4. Identify the processes producing these outputs. 5. Identify the owner(s) of the processes. When customers needs and expectations are not well understood, special efforts must be made to define them clearly and obectively. This lack of understanding is often encountered when output properties are difficult to measure, as is commonly the case when the outputs are intangible. The owners of the processes involved should also be identified in this first step. Again, this task is often taken for granted because of how clear ownership is in most manufacturing operations. For example, in manufacturing, responsibility is usually defined clearly and documented for daily operations, preventive maintenance, and individual manufacturing blocks of process units. On the other hand, this clarity is often lacking in service processes. Understanding who the owners are is an essential step in assuring that appropriate resources are applied and that identified improvements can be authorized. Step : Identify and document the process In the absence of clearly defined work processes, work is needed to establish this basis. Since manufacturing processes are usually well documented, this step has historically been taken for granted. This second step in process improvement demands that the process be described in understandable terms, which is usually accomplished with a picture or model, not merely through a written or verbal description. The flowchart is a commonly used tool for describing processes. The technical literature is filled with methods for flowcharting, and several sources are listed as references at the end of this chapter. Creating a flowchart enables you to perform the following four improvement activities. 1. Identify the participants in the process, either by name, by position, or by organization. 79

80 . Provide all participants in the process with a common understanding both of all steps in the process and of their individual roles. 3. Identify inefficient, wasteful and redundant steps. 4. Offer a framework for defining process measurements. Step 3: Measure performance In the absence of documented performance standards, remedial work is needed to quantify how well the system is performing. Further, these measures must be defined and evaluated in the context of customer expectations. This step is of double importance in situations where neither the output requirements nor processes have been defined previously. One of the former chapters introduced the concept of measuring performance at three levels: process, outputs, and outcomes. Process measures define activities, variables, and operations of the work process itself. Output measures define specific features, values and attributes of each product or service and can be examined from two sides. One side represents the output characteristics desired by the customers requirements, and the other side represents the output characteristics actually delivered by the process (capability). The former is referred to as the voice of the customer and the latter as the voice of the process. Outcome measures define the ultimate impact of the process and are dependent on what the customer does with the product or service. Customer satisfaction represents the key measure of outcome. VIII. Table Outcome Output Process Customer satisfaction Characteristics desired by customer Characteristics delivered by process Performance measures In making decisions as to which brand of grade of gasoline to buy, customers are concerned with how the product itself will perform in their car, the purchase price, and the attributes of service at the point of purchase, such as convenience with regard to time and location, appearance of the facilities, and behaviour of the attendants. To focus on ust one characteristic for the purpose of this example, consider the performance of the product itself. The customer s need for a smooth, quiet, trouble-free operation of his or her car s engine has been translated into a set of technical specifications for gasoline. One of these is octane, and one measure of the voice of the process is the actual octane of the gasoline in a specific tank at a specific service station at a given time. Other related measures include the variation of octane within this tank over time, the variation of octane among tanks of various service stations, and the variation of octane for this grade of gasoline in the blending tanks of the refinery. Moving into the processes at the refinery, product specifications such as octane are translated into other performance characteristics for the various feedstreams. One example might be the temperature, pressure, and flow rate of one stream into a reactor. 80

81 So what this example shown? It began by defining the desired outcome as smooth, quiet, trouble-free operation of the customer s car engine. Yet it somehow twisted around to define one aspect of process measures as temperature, pressure, and flow rate into a reactor at a particular refinery. This translation is of fundamental importance because the process is actually controlled by measuring temperatures, pressures and flow rates in the refinery, not by measuring octane in the service station tanks. This translation was possible because all of the process involved were described, modelled and understood. One model related the customer s satisfaction to the performance of his or her car s engine. A second model related the engine s performance to product specifications of gasoline. A third model related product specifications to refinery operating conditions. 1. Customer satisfaction to engine performance. Engine performance to fuel specifications 3. Fuel specifications to refinery performance 9. Figure One snapshot of example measures shown in the following table. IX. Table Outcome Customer survey regarding engine knock 67% pleased 31% satisfied % dissatisfied Output Customer wants 89.0 octane Process delivers 89.3 octane Process Feed to reactor A flow rate: 37,000 barrels day inlet temperature: 455 degrees F inlet pressure: 75 psi This snapshot offers one specific set of values for each level of measurement. In reality, however, the picture is complicated because the variability of the data must also be understood. Even for a single driver with one car, the required octane will vary with weather, altitude, and condition of the engine (e.g. wear, deposits, and intake and cooling system performance). Likewise, the process characteristics and performance will vary simultaneously. 81

82 Although this example relates to the product and its manufacturing conditions, it is analogous to measurements required in service. As an alternative, this example could have defined the voice of the process in service dimensions rather than along one characteristic of the product itself. Could the voice of the process in the service dimension be related to measuring the employee selection, training, and compensation criteria for service station attendants? Could it be related to the employees ob satisfaction? Could it be related to process for determining the design, location and hours of operation for the station? How can the process that impact on the service characteristics be modelled and measured? Answers to these questions are provided in the next chapter. Step 4. Understand why The lack of data increases the difficulty of understanding why the system is performing the way it is. This problem is compounded when the outputs do not appear to be produced by a repetitive system, as is the case for long-term process, such as research, development, and engineering. The six-step process improvement model is designed to bridge this gap to assure identification of the specific factors limiting the system s capability. By defining problems in the context of their process (step 1), identifying all steps in the process (step ), and measuring performance in obective terms (step 3), the classic tools of statistical analysis and quality can be applied to understand the root causes of the performance gap. Four tools stand out for accomplishing this step: Pareto analysis, cause-and-effect analysis, histograms and control charts. Steps and 3 of the process improvement model were designed to provide the base of fundamental principles. This is accomplished through flowcharting, modelling and measuring. Step 4 offers methods to acquire the next level of knowledge and to gain a profound understanding of the process. The following questions need to be answered: Why are people with good intentions often frustrated in their efforts and seem to accomplish precious little that is of long-term benefit? Why do problems recur for no apparent reason? Why do things go from bad to worse? Answering three basic questions can help build a fundamental understanding of the process in order to take effective and efficient steps toward permanent improvement. Have we distinguished the vital few from the trivial many? Juran believes that a fundamental law of nature dictates that 80 per cent of the problems are the result of 0 per cent of the causes. One key improvement is to find those crucial 0 per cent and focus attention on them. Juran named this concept, which represents a maldistribution of quality losses, in honor of Vilfredo Pareto, a nineteenth century Italian economist (Pareto found that a large share of wealth was owned by a relatively few people maldistribution of wealth). The Pareto analysis is a method for categorizing and recategorizing causes until the vital few are found. 8

83 Interesting many 0 0 Vital few Causes Effect Figure Have we designed the root causes? Ishikawa suggests that the first signs of a problem are its symptoms, not its causes. Actions taken on symptoms cannot be permanently effective. It is necessary to understand and act on the underlying root causes. Cause-and-effect diagrams, fishbone charts, and Ishikawa diagrams are synonyms for a basic tool that can be used to help differentiate among symptoms, causes and root causes. X. Table Level Observation Action Outcome Symptom Car does not start Call tow truck $5 bill for ump start Cause Dead battery Recharge by driving Arrive at work (late) Cause Broken fan belt Call tow truck (again) $50 bill for ump start and replacement Root cause Inadequate preventive maintenance Implement manufacturer s recommended service Problem eliminated Do we understand the sources of variation? Deming explains that all variation is caused and the causes can be classified. Common causes are inherent within the system and yield random variation within predictable bounds. Special causes are assignable to specific reasons or events and result in sporadic variation that defies prediction. Taking correct action to control variation requires knowing its type, since appropriate actions differ according to the type of variation. Common causes can be solved by addressing the underlying system. On the other hand, special causes are addressed by eliminating their specific, identifiable source. Control charts help to distinguish between common and special causes of variation. 83

84 After the special causes are identified, the remaining inherent variation is attributable to the common causes. The capability of the process is defined by comparing this range of variation to the specifications or customer requirements. Capability can be calculated and can be visualized by plotting the data as a histogram. Understanding variation and process capability are the preferred first activities in step 4. However, when outputs are intangible and not frequently repeated, data for this analysis might not be available. In these cases, the question posed in step 4 will need to be answered in parallel or in an iterative way. Step 5. Develop and test ideas The first 4 steps built the foundation for understanding the critical dimensions of the process. They assured knowing what the processes are, measuring how well they are performing, and understanding why they are performing the way they are. These steps led to identifying the underlying causes of the principal problem. Developing ideas for improvement begins with step 5. Ideas for improvement must address the root causes of the problem. Step 5 is the point at which development for new ideas and potential solutions should be encouraged. What are new and different ways to design and operate the process to eliminate the root causes? A complementary approach for developing improvements is through experimentation. Design and conduct experiments to test the hypotheses developed in step 4. Also, design experiments to test the ideas developed in step 5 before implementing them. When tests fail to produce the desired results, determine the cause. Was the test valid? Was the improvement idea effective? Were you mistaken about the root causes of the problem? Was the process completely identified? Were the customers requirements misunderstood? Recycle back to the appropriate step on the process improvement road map. Step 6. Implement solutions and evaluate The sixth step begins by planning and implementing the improvements identified and verified in step 5. Step 6 continues to measure and evaluate the effectiveness of the improved process. But these activities represent only one side of this step. The other side is to evaluate the six-step process itself and to acknowledge and celebrate the contributions of those who participated in this increment of process improvement. It makes no difference whether this was the work of an individual or, as is more likely, a team effort. Reward the contributors for the result achieved as well as for their discipline in applying the six-step model. Finally, return to step 1 begin the next increment in the ongoing process of continuous improvement. 84

85 Is this the only approach to process improvement? The previously presented six-step process serves as a universal road map to process improvement. It provides a systematic approach to build a fundamental understanding of customers requirements, process capabilities, and the causes for gaps between them. A road map differs from a prescription. People who are unfamiliar with the route use a route map as a guide for getting from point A to point B. On the other hand, a prescription specified requirements that must always be followed. The six-step road map guides the application of a fundamental improvement strategy known by any of four names: PDCA, plan-do-check-act, the Shewart cycle or Deming cycle. The plan phase is guided by the first four steps along the road map. These steps help clarify problems and develop hypotheses as to their root causes. Step 5 covers the do and check phases by testing the previously developed hypotheses. Step 6 completes the cycle ( act ) by implementing improvements to the process. Numerous alternative guides for PDCA are available, and several organizations have customized the six-step model for their own use. Some have modified the terminology to incorporate their own argon. Some have reduced the number of steps by combining elements, and others have increased the number of steps by separating elements. IV.6.a. Summary Systematic process improvement relies on building fundamental understanding of customers requirements, process capability, and the causes for gaps between them. Hypotheses are developed and tested, and improvement gained through the continuous cycle of plan-docheck-act. This systematic approach bears striking contrast to the classis short cut of problem detection and subsequent solving, and approach resembling plan-act-plan-act. The six-step road map was developed as a universal guide for systematic improvement. It can be used to improve manufacturing as well as nonmanufacturing processes, and helps assure application of the plan-do-check-act strategy. Many quality improvement practitioners have focused on merely on the application of basic tools such as flowcharts, control charts, and cause-and-effect diagrams. The six-step road map offers a robust approach to improvement, and places the tools in their proper perspective. Although they are valuable, the basic tools are merely techniques to help make the best performance of work processes visible and by themselves are of relatively little benefit. IV.7. Measuring performance One element of TQM is to base decisions on data, not opinions. This chapter provides a framework for defining the parameters that are needed to measure performance. Clarifying which parameters can and should be measured will help in systematic and continuous improvement of all products, services and processes. First part establishes three levels of measures: this first one for controlling operations within the process, the second for measuring the outputs delivered, and the third for quantifying the 85

86 outcomes. The second part of framework defines four dimensions of results: products and services delivered to the users and customers, financial return for shareholders, ob satisfaction for employees, and social impact upon the community. IV.7.a. Measure at three levels The use of measurement is pervasive in systematic process improvements. Although measurement was introduced in the third step of the process improvement model, various measures are used in all steps. Four types of measures were introduced, and they represent three distinct levels: process, output, and outcome. The relationship between the four types of measures and the three levels is shown in the figure below. Process Performance parameters Operations that control products/services produced and delivered Output Requirements Features, values and characteristics desired by customer Capability Features, values and characteristics delivered by process Outcome Customer satisfaction Degree to which products / services are perceived to meet expectations 11. Figure Process Performance parameters: measure each step/activity in the process and the characteristics of inputs delivered by suppliers that control the desired output characteristics. Identify the behaviours that govern each step, and use these measures to control operations and to predict the outputs before they are produced or delivered. Performance parameters are analysed in steps 4 and 5. Output Requirements: Define the specific features, values, characteristics and attributes desired by customers for each product and service. These measures represent the voice of the customer and are used in step 1 along the six-step road map. Requirements and expectations must then be translated into product/service specifications. Capability: In direct correspondence to every specific feature, value, characteristic and attribute desired by customers, measure its level, value or presence in each product and service actually delivered by the process. These measures represent the voice of the process and define what the process has delivered. Process capability is measured in step 3 and tested in step 5. 86

87 Outcome Customer satisfaction: This is the highest level of measurement and represents the ultimate desired result. Measure how well each product and service satisfies the needs and expectations of the customer, and recognize that these measures are based on vague, idiosyncratic perceptions. Outcomes are beyond direct control of the supplier and rely on the customers expectations and actions. These measures often trigger process improvement initiatives (step 1) and can also be used to evaluate the results achieved (step 6). IV.7.b. Measure in four dimensions In addition to the output actually delivered to the end customer and the resultant outcome, each process generates by-products and outcomes for other customers. One is financial return for the shareholders. Another is ob satisfaction for the employees. The third byproduct is the social impact on community. The three levels of measures were described in the following figure with respect to the actual output, but they also apply to each by-product. Satisfaction and desired characteristics of these by-products are defined by their respective customers. These, in turn, represent three additional sets of specifications against which process performance can be measured. As shown in the figure, performance parameters can be defined in four dimensions. Each dimension corresponds to the output or outcome desired by its respective customer: the end user, the shareholder, the employee, or the community. XI. Table Dimensions Focus Example measures 1. actual product or services End user or customer specific features, characteristics and attributes defined by customer. Financial return Shareholder costs, profits, prices, throughput, sales volume, waste, productivity, efficiency, cost of quality, utilization, return on investment 3. Job satisfaction Employees Specific needs and values defines by employees, employee satisfaction 4. Social impact Community regulatory compliance, atmospheric emissions, liquid discharge, waste disposal and recycling, grants and contributions, presentations and publications, taxes and fines 87

88 Four dimensions of performance measured applied at three levels are offered as an antidote to the deadly disease of running a company on visible figures alone. In essence, Western companies have focused on themselves and the financial markets rather than on their customers, employees or the community. Organizations moving toward focusing on the customer can facilitate their transformation by defining and using measures indicative of customer satisfaction. Similarly, organizations striving for total involvement of their work force need to pay attention to the parameters that directly relate to improving employees satisfaction with their obs. It should be obvious that a balanced approach is needed. None of the dimensions can be ignored. To the extent that the shareholders obectives, the customers requirements, the employees needs, and the community s expectation all overlap with each other, the easier the task becomes. Long-term success can be assured by selecting and leading a balanced attack to improve systematically in all dimensions. What are your values? Where is your focus? Are they consistent? Customers Shareholders Community Employees If this is your focus Then this is who you need to please and these are key measures Customer Customers Customer satisfaction Output characteristics defined by customers Shareholder Bosses Owners Financial analysts Financial indicators (costs, sales and profits, cost of quality) Goals and obectives defined by management Employee Employees Employee satisfaction Factors contributing to ob satisfaction Community Government agents Social services Professional societies Academia The press Regulatory compliance Factors impacting on society 1. Figure IV.7.b.1 Summary Many organizations are driven by performance measurement. The absence of meaningful goals and measures can lead to useless meandering. Perhaps even worse, use of the wrong measures can drive organizations in the wrong direction. To overcome these problems, this chapter offers a two-part measurement framework. One part establishes three distinct levels of measures: the first for controlling operations within the process, the second for assessing the outputs delivered, and the third for quantifying the outcomes achieved. The second part of the 88

89 framework defines the perspectives of four basic stakeholders: customers, employees, shareholders, and the community at large. Underlying these frameworks is a performance measurement paradigm that is provided as a summary of measurement theory. 1. Every product and service can be characterized by a set of performance measures.. The ob begins by understanding your customers and identifying the set of characteristics that fully define their needs. 3. These customer-driven characteristics must be translated into process measures and learn the performance level that your process is capable of delivering for each characteristic. 4. Then it should be determined how satisfied customers are with performance at current level, and the relative importance customers place on changing the level of each characteristic. IV.8. Leadership Total involvement forms the third and final core concept of TQM. Through it, the idea of winning loyal customers is mirrored into the organization, to build loyal employees and loyal suppliers. Total involvement aligns and integrates the efforts of everyone: managers, workers and suppliers. This chapter describes the transformation of managers into leaders and their role in championing TQM. The elements and characteristics of leadership have been described and defined many times in many ways. Our intention is to discuss leadership as it applies to the understanding and implementation of TQM. In this regard, leadership has specific implications for the senior managers embarking on a quality ourney. This chapter describes these implications and suggests actions that senior managers can take to enhance the likelihood of success. IV.8.a.1 What is leadership for TQM? When implementing TQM, several fundamental activities must be accomplished by the senior management team. These senior managers bear ultimate responsibility for the success of the organization and, by virtue of their positions, have the authority to set direction, establish policy, allocate resources, and select the markets in which the firm will participate. These individuals are responsible to their customers, their employees, and ultimately their shareholders for the success of the enterprise. TQM requires skills in both leadership and management. The difference between these two skill sets: leaders are people who do the right thing; managers are people who do things right. The next figure offers a further clarification of the distinction between management and leadership. To implement TQM, managing to get results will share center stage with leading to improve systems. The old roles of planning, organizing, directing, coordinating, and controlling will be diminished. In their place, we will find leaders who vision, align, empower, coach and care. 89

90 Managing Plan Organize Direct Coordinate Control Leading Vision Align Empower Coach Care GETTING RESULTS IMPROVING SYSTEMS 13. Figure Continuous improvement of all products, services and processes is accelerated if everyone challenges the status quo every day. Leaders can set the stage for this challenge by developing answers to six fundamental questions: 1. Why do we exist; what is our purpose? (mission). What will we look like in the future? What do we want to become? (vision) 3. What do we believe in, and what do we want everyone to abide by? (values) 4. What guidance will we provide to the many individuals in our organization as to how they should provide products and services to our customers? (policy) 5. What are the long- and short-term accomplishments that will enable us to fulfil our mission and attain our vision? (goals and obectives) 6. How are we going to move toward our vision and accomplish our goals and obectives? (methodology) Answers to the first three of these questions form the cornerstones of the leadership framework. The mission defines why the organization exists. The vision show what the organization wants to create, and the core value explain how we want to act. Answers to the remaining three questions fill in the details and build on these cornerstones. 90

91 Implement (actions) Commitment Allocate resources consistent with leadership framework Vision What we will look like in successful future state Stlye Role model How we will behave Specific intent Obectives What is to be achieved in scheduled period of time Methodology How obectives will be achieved through strategy and plans Broad intent Goal What is to be achieved through sustained effort over long time period Policy Statement of principles that guide how all business activities are to function Foundation (direction) Mission The purpose of reason for the existence of the business Doing the right things (What to do) Values Define how we will behave by Delineating shared beliefs Doing things right (How to do it) 14. Figure These questions, while seeming simple and obvious, are enormously complex and difficult to answer, especially when an organization s traditional products and services are buffeted by new technologies, by competent and aggressive competitors, or by changes in customers expectations. Nonetheless, failure to respond to each of these questions renders and organization incapable of understanding and meeting its customers demands, unable to allocate its resources effectively and efficiently, and incapable of capitalizing the talents of its people. IV.8.b. Why do we exist? The question of why we exist is answered in the organization s mission statement. This statement is comprehensive, easy to understand, usually one paragraph in length, and describes why an organization was created and its primary function. The mission provides a clear statement of purpose to all employees as they perform their daily tasks. The following mission statements are offered as examples to illustrate answers to question Why do we exist? These statements, when published and disseminated to employees throughout the organization and widely communicated to customers and the public at large, commit an organization 91

92 internally and externally to a specific purpose. A well-articulated mission statement also guides employees as to how they should manage their activities and sets a level of expectation among customers as to what they should get from this organization. IV.8.c. What type of organization do we want to become? The vision provides a description of what the organization will evolve into in the future and, like the mission statement, provides continuous guidance to employees at every level as to how they should manage their respective responsibilities. Usually, the vision is drafted as soon as the mission statement is completed and is developed through the same collaborative process. The vision need not be described in precise financial or market terms. Rather, the intent is to provide a broad description of what an organization can become if everyone s efforts are focused and successful. The vision is less a dream, less a soft specification of what is desirable, than a realistic picture of what is possible. The vision states what the organization wants to become and where it wants to go. This direction may be based on data regarding what is being achieved by others and what is possible for this organization. Any of three flaws can be fatal to the process for building an inspirational vision: (1) failure to have a genuine vision one that is important, challenging and at the same time realistic; () failure to communicate the vision; (3) failure to rally everyone s genuine support. Develop an enabling and empowering vision 1. Effective visions are inspiring.. Effective visions are clear and challenging and involve excellence. 3. Effective visions make sense in the marketplace and, by stressing flexibility and execution, stand test of time. 4. Effective visions must be stable but constantly challenged and changed at the margin. 5. Effective visions are beacons and controls when all else is up for grabs. 6. Effective visions are aimed at empowering our own people first, customers second. 7. Effective visions prepare for the future but honour the past. 8. Effective visions are lived in details, not broad strokes. IV.8.d. What is important to us? Every organization operates with some behaviours being understood without ever being explicitly stated. But why leave these things unstated if they are important to running the business? Why not provide clear definitions of those important things, so everyone has a clear sense of the values of the organization? IV.8.e. How do we want everyone to work? Organizations consist of many different individuals with many different opinions. The quality level for the production and delivery of goods and services should not be left for 9

93 everyone to determine individually. Instead provide a written quality policy and consider the elements listed in the following table. Element Importance of quality Quality competitiveness Customer relations Internal customers XII. Table Example statement Quality is first among equals Best in class Meet customer s needs Quality extends to all segments Work force involvement Our policy is to provide 40 hours of quality training annually to every employee Quality improvement IV.8.f. What do we have to do? We will continuously improve our services Leaders must next move from the broad direction provided by mission, vision, value and policy statements to declare what must actually be accomplished. Unless these statements are translated into measurable performance targets, they will be little more than nice sets of words. Goals cover the organization s broad intent by defining what is to be attained through sustained effort over the long term. By contrast, obectives define what is to be achieved in a specified period of time. Clear-cut, measurable obectives help spur everyone in the right direction and serve as mileposts against which progress can be gauged. Five criteria help in framing effective obectives: 1. Obectives should be definitive and specific.. Obectives should describe accomplishments or results, not activities or behaviours. 3. Obectives should be measurable (quantifiable). 4. Obectives should delineate a time frame or deadline. 5. Obectives should be challenging yet achievable (they should be within the control of the responsible business unit and not rely on what might be accomplished by other organizations. IV.8.g. How are we going to accomplish these obectives? Having established what is to be achieved through a sustained effort over a specific period of time, the next step is to determine how these obectives are to be achieved. This is accomplished through development of strategies and plans for specific courses of action: who does what by when. One element of the strategy is the adoption of TQM as a comprehensive management system to achieve obectives by assuring that all work is performed in a systematic, integrated, 93

94 consistent manner. The installation of TQM becomes the work of senior management, and its pursuit of TQM is a prime example of its leadership. IV.8.h. How do we develop these statements? Statements of mission, vision, values, policy, and obectives developed behind closed doors by a single executive or by ust a few top-level managers in isolation serve a useful purpose but are limited in impact. In this case, the key advantage of developing these statements is quickly lost once the task is completed. The advantage of securing ownership, commitment, and involvement will accrue to organizations that use a collaborative process to forge these key statements. Furthermore, the benefits will last for an extended period of time. A senior manager who either establishes a new vision or reaffirms an existing mission by involving others in a participative manner garners ownership and commitment from all those who were involved in the process. Three notes of caution: First, managers are often more preoccupied with complementing the task of formulating these statements and moving on to new activity that they are with using this exercise as a means of securing support from key managers, employees, and impacted departments. Be aware of the task-oriented mentality, and recognize the value of the process itself. Second, using a participative decision-making process to develop fundamental statements does not imply democracy or a need for total consensus. When making a maor change in the organization s culture, some individuals invariably disagree with the new direction. Determine which views represent added value and which represent anchors holding the organization to past practices that are inconsistent with the new vision. Third, having stated that participation is important, be careful not to cross the fine line that separates participation from abdication. Senior management is responsible for the direction of the organization and for developing the enabling strategies. Participation is one way to accomplish this task, but ultimate responsibility for deciding which course to take cannot be delegated. IV.8.i. Other elements of leadership The last two elements of leadership framework cover management s actions in day-to-day activities. Commitment relates to what is done by managers themselves and in the allocation of resources. Style relates to how they do things. Commitment Although there are many additional examples that illustrate leadership in the installation of TQM, one that remains as significant as the process of forgoing key statements is the demonstration of commitment by senior managers. The demonstration of commitment takes many forms and is evidenced by what managers do as well as by what they say. The first opportunity to demonstrate commitment is often found in reallocating resources and funding additional staff to implement the extensive training that is required to equip all 94

95 employees with the skills and knowledge they need to pursue TQM. Will the resources be provided? Will funds be made available from the redirection of traditional training programs? Equally important are senior managers willing to demonstrate commitment by participating in the same training that everyone else will undertake. Will senior managers use the tolls, argon and processes to pursue continuous improvement? Or will they merely instruct their subordinates to do it? Are senior managers willing to create structures to support TQM, such as boards and steering committees? And are they willing to participate in these structures? Or are they going to delegate this responsibility? Are senior managers capable of demonstrating long-term commitment to implement continuous improvement, even when improvement may be viewed as having high start-up costs? Commitment means more than new procedures, policies, directives, letters and speeches. Employees look for commitment from top management by observing their behaviours and their style. You should expect that employees will question and test the beliefs you now espouse. Leaders must be committed to the following: maintaining a long-term perspective in the face of short-term pressures providing a focus on meeting customers needs and being sensitive to the markets in which the organization participates realizing that some returns are immediate but others will take three to five years of dedication and hard work stressing that improving the process is as important as attaining results and that both must be accomplished allocating resources to support TQM training and implementation Instituting compensation, recognition, reward, and promotions based on the criteria that support TQM Fostering employee involvement and better relationships across functions and between unions and management Building partnerships with suppliers for mutual benefits Acting as a role model; walking the talk using quality improvement tools and argon. Style In addition to what leaders do, how they do it is equally important. Leaders base decisions on data. Opinions are interesting, but decisions should be based on what you know, not what you think. Leaders are resources, coaches, facilitators for the individuals with whom they work. Employees who are looking to their leaders for direction, decisions, or approvals don t understand the leader s new role. Have they explained the difference? Is the leader s style consistent with the leader s intent? 95

96 Leaders are actively involved. They learn new skills along with their employees. Being proficient, leaders are then able to assist others in their pursuit of continuous improvement. Leaders build commitment. They assure that everyone understands the organization s mission, vision, values and goals. Furthermore, leaders assure that everyone knows his or her individual role and is eager to contribute. Leaders inspire confidence. They extract the best from everyone and encourage personal development. Leaders say thanks. They do so with every imaginable form of monetary and nonmonetary incentive. IV.8.. Summary TQM is a strategic process that is launched from the executive suite and requires both management and leadership skills. This principle, when accepted, sets in motion a series of activities by senior managers that shape the organization s future. These activities cannot be delegated, and the process of how things are done is as important as what is actually accomplished. That is to say, the right things must be done right. Leaders will begin by addressing the basic questions about what the organization is and what it is to become. Through a collaborative process, they will secure the cornerstones of the organization s foundation: the mission, vision and core values. They will galvanize commitment by all to this foundation and provide an environment within which the details can be built. The degree to which these issues are resolved is the extent to which the organization can meet its stated obectives in the short run and its vision in the longer run. Leaders will be unrelenting in their pursuit of the organization s goals. Finally, leaders must use their own style to reinforce the crusade toward the vision. 96

97 IV.9. Empowered workforce The work force has changed dramatically since the end of World War II. Work used to be a series of simple, manual tasks performed by unskilled labourers, and those tasks could be observed easily. For every ten to twelve workers driving spikes to lay railroad track, one crew chief assured that work progressed smoothly and that everyone contributed his fair share. For every ten crew chiefs, one foreman monitored the performance of the crew chiefs; and for every ten foremen, one superintendent supervised the entire operation. This hierarchical model, fashioned after the military organization, was necessary because uneducated, unskilled labor toiled with mechanical tools on tasks that had to be observed to assure that work was done, and done correctly. The boss dispatched and allocated tasks and then monitored the work by being physically prevent to observe directly that all was progressing as required. Contrast that scene to one in a modern office where an individual sits in front of a computer screen. It s not possible to know, simply through observation, whether the individual is daydreaming or developing answers to the organization s most critical problems. In addition, it is no longer possible to tell who is in charge since the old hallmarks of physical appearance no longer apply. An administrative aide may dress as well as the CEO, drive a similar car, dine at the same restaurants, and live in the same town. The distinction between blue collar and white collar as well as other signs of social standing are passé. More important, the education of subordinates may be similar if not superior to that of the manager, especially in highly skilled areas of technology. Organizational distinctions are blurred, and the simple mantle of authority may be insufficient to manage young, competent, highly motivated employees. Most important, those people who are actually doing the work are likely to know far more about it than managers. So how does one supervise a large number of educated, competent, motivated employees? Many US managers learned, during the 1970s and 80s, about participative practices by visiting Japan or by reading about Japanese management techniques. Many grew certain that employee participation was the answer they needed for improved productivity. There was a great rush to establish quality circles in US plants, with some encouraging initial results. Today, however, there is little to show for all this effort. If participation is the answer, and we tried it without achieving the tremendous benefits we were seeking, where do we go from here? IV.9.a. What is empowerment? The first dimension of empowerment is alignment. All employees need to know the organization s mission, vision, values, policies, obectives and methodologies. Furthermore, the broad direction of the overall organization must be sharpened as the message cascades to define the role of work groups and individuals. Fully aligned employees not only know their role, they are dedicated to supporting it. Leaders have inspired them to contribute to the mutual benefit of the organization and themselves. 97

98 This dedication is synonymous with commitment, and commitment can neither be brought nor sold. It is earned. The traditional hierarchical organization did not require commitment, its authoritarian style merely needed compliance. Gaining true alignment requires bridging the subtle gap between compliance and commitment. In essence, the difference between compliance and commitment is that people who are committed the organization s direction truly want it. Although empowerment requires commitment, in the near term, leaders may need to settle for compliance and work to move people up the compliance ladder. Commitment cannot be enforced; efforts to do so will, at best, build compliance. All that can be done is to establish an environment that is favourable to the growth of commitment. The second dimension is capability. Employees must have the ability, skills, and knowledge needed to their obs. They must also have the resources needed from the organization: materials, methods, and machines. In our experience, many organizations have overachieved in this dimension. Selection criteria bring in new hires who are overqualified, and investments in facilities, equipment, and training have far outstripped growth in the other two dimensions trust and alignment. Mutual trust is the third dimension of empowerment. Once we have developed alignment and capabilities, we are in a position to unleash the power, creativity and resourcefulness of the work force. This will not happen, however, unless we have provided this third dimension. Employees need to trust management and feel that management trusts them. Mutual trust therefore completes the picture required to build an empowered work force. Empowerment can therefore be seen as the natural consequence of effective leadership. If empowerment relies on alignment, capability, and trust, it is reasonable to expect employees to feel empowered if they don t feel trusted. It is so simple: treat people with the same respect with which you would like to be treated. IV.9.b. What is teamwork? Teamwork is a group of individuals working together to reach a common goal. The goal may be to increase market share, to boost customer satisfaction, or to improve overall performance through cooperation and collaboration. Teamwork is at first sharing of responsibility and eventually sharing of decision making that impacts the entire organization. Collaboration and teamwork build a new level of capability, a new strength that the organization can harness to increase its customers satisfaction. Like empowerment, teamwork and team building have become popular terms in business circles. Before the industrial revolution, a team usually consisted of horses, mules or oxen drawing carts or plows. During the past century, a work team might more aptly have been described as a gang or crew. It is relatively recently that teams have become recognized in industry and discussed in the context of collaboration, cooperation, competition and sporting events. But even now, the terms are frequently misused. In traditional, hierarchical, autocratic organizations, teamwork 98

99 has often meant compliance, and a team player is often a euphemism for a conformist who suppresses his or her own goals for the sake of the organization. Some managers mistakenly assume that a group cannot be called a high-performing team if is doesn t fit some idealized model of total autonomy and equality. The optimum degree of autonomy, control, interdependence and collaboration varies for different types of teams. Different types of tasks demand different types of teams. For instance, consider the differences in decision making, interdependence and collaboration required to perform the following tasks: driving spikes into miles of railroad tracks, deciding on the guilt or innocence of an alleged criminal, winning a football game, or improving the design of a data-processing system. Constructing miles of railroad track is an example of additive task, and the accomplishment of the team equals the summation of individual contributions. In this type of task, the team will accomplish more than any individual; however, the contribution of each member is frequently below that of the same individual performing on his or her own. The maximum potential capability of teams performing additive tasks is proportionate to the size of the team, but the actual work accomplished will be less. The key to leading this type of team is to minimize this gap. A ury deciding on guilt or innocence represents a team performing a disunctive task. Solving problems and making decisions are common disunctive tasks. In this type of task, the performance of the team will approximate that of the most knowledgeable member. Teams cannot perform beyond their available resources, and optimal performance will be realized through identification and support of the most knowledgeable member. Popular corporate team-building exercises such as desert survival or lunar landing games provide practice for teams performing disunctive tasks. By contrast, playing a football game is a conunctive task. Here, each member must perform his or her own task, and the team s capability may be limited by the ability of the weakest member, not the strongest. A quality improvement team working to improve the design of a data-processing system is an example of an optimizing task. Here, the goal is to produce some specific, most-preferred outcome. Collaboration and cooperation among members result in the team s performance potentially exceeding the summation of abilities of all the individuals. Furthermore, this quality improvement team will likely address subtasks that are additive, disunctive or conunctive. The mode of operation of the team should adust for each subtask. When forming teams, examine how the type of task will determine the teams needs for autonomy and collaboration. Also evaluate the stake the members have in the outcome. IV.9.c. Guidelines for building teamwork Teamwork should be beneficial to all: employees, customers, and the shareholders. The following guidelines are offered to help maximize likelihood of building successful teams. 99

100 Management support Managers are delegating responsibility, not abdicating, when embarking on a participative management process. This requires managers to clarify expectations and boundaries. Team members need to understand what is important to the manager and to the company. Members need to know the issues that the organization is facing. In short, they need information. A sure way for a nonsupportive manager to kill a team is to withhold information. Another surefire killer is to withhold resources. Clear charter Information needed to clarify the direction and scope of a team s effort can be included in its charter. Three essential elements are: (1) a general description of the problem or opportunity to be addressed; () the expected outcome; and (3) the boundaries. Teams are formed to resolve specific issues will benefit from clear and comprehensive charters. Teams that are natural work groups will likely find that their charters are embedded in their mission (description): vision, goals, and obectives (expected outcomes): and core values, policies, strategies and plans (boundaries). Benefits for team members Organizations introduce teamwork to improve work processes and increase customer satisfaction, and to do both of these at substantially lower cost. These benefits are part of the organization s stated obectives. Yet where are the benefits of teamwork for employees stated? Our employees are our most valuable resource. If this claim is to ring true, then clear statements of how teamwork will benefit employees are needed. Benefits for employees can include improves quality of work life, development of personal skills, rotational ob opportunities, and increased ability to make decisions that will influence the direction of the organization. Benefits also include direct, specific and timely rewards and recognition. Bias for action Recommendations must be acted on in a timely manner. Managers should expect that teams are aligned with the organization and that their recommendations will be of value. If recommendations are not accepted, team members are owed an explanation from the highestlevel executive. The process of any successful team should be studied so that future teams will not stumble. What were the root causes of the problem that derailed the team? Skills Although participation on most teams is voluntary, individuals may occasionally be drafted because they possess unique skills that are required by the team. Such individuals might be full-fledged team members, or they might merely participate during one phase of the work. Regardless of their role, if expertise is required, get it. 100

101 Training Members need to be skilled in one or several of the disciplined processes used by teams. Problem solving, benchmarking, or process improvement are examples of road maps that teams may follow to reach their destination. Training in these techniques equips everyone on the team with a common core of skills, regardless of dissimilarities in their work experience or education. Training in group dynamics and meeting skills may also be required if members have difficulty working in a group environment. Facilitators Even when individuals are trained in team dynamics, support may still be needed, particularly in the early stages. Such assistance can be provided by an outsider who acts as a process expert someone who will focus on the team s process and not the content. Start small Teams should begin with simple opportunities within their own function and, when successful, branch out more effort, expertise, and resources to solve. They also often yield bigger payoffs, but teams should develop some success, expertise, and confidence before tackling more complex issues. Life cycle Teams are born, live, and die like natural organisms, and this cycle should be recognized. When the problem is solved, then perhaps the team created to solve that problem should disband. If, however, related opportunities are found, then the team may move from issue to issue. If members are from the same group and are working to improve their normal work processes, then that team should be encouraged to continue. Celebrate True teams succeed or fail together. Members of successful teams should be recognized and rewarded. One form of recognition is to present their story to others during publicized events. This not only serves to recognize the contributions of the team but also provides a learning experience for other employees who may be struggling with similar issues. Communication of these successes reinforces the drive to empower individuals. The value of appropriate communications cannot be overstated with regard to fostering teamwork, empowerment, and process improvement. IV.9.d. Criteria for team performance Teams are advocated to get results. However, the process of building teams provides value by itself. To keep sight of this, six criteria, each of equal weight, are offered for evaluating the performance of teams. Alternatively, results may become paramount, and other criteria may diminish in perceived value. When this occurs, the organization has reverted the old way of 101

102 doing things and is not following an approach that has been validated as essential to TQM. Evaluate a team s success by considering how well it performs in each of the following areas: Goals Are goals understood and supported by all team members and by management? Are goals realistic ambitious but achievable? Are goals within the scope of ownership of the team members or the sponsoring manager? Do the goals contribute to the mission, vision, and obectives of the organization? Roles Do all members know and fulfil their responsibilities? Members responsibilities may include allocating time to team meetings and performing specific roles as well as working outside of meetings. If one member is not supporting the team, others have the responsibility of asking that individual to leave the team. Process Has a structured, defined, disciplined process been followed by the team in pursuit of its goal? Examples of such processes include problem solving, benchmarking, and the six-step road map for process improvement. Development As a result of involvement on the team, have all members learned new concepts, tools, or techniques that they can apply to their own work outside of the team? Innovation Did the team question conventional wisdom and traditional approaches? Did the team find new ways to attack old problems? Did the team recommend an entirely new approach that both improves customer satisfaction and reduces costs? Results Did the team achieve its goal? Does the solution permanently improve the underlying work process? IV.9.e. Stages of team development Stage 1: Forming This is the initial stage in which new relationships are formed among team members. A great deal of testing is going on as individuals seek to understand the reason for the team being formed, the scope of responsibility, the legitimacy of the team, why various members were 10

103 selected, the doability of the task, and what the task really is. This initial activity helps the group to orient itself to the work to be done. In this initial stage, the role of the sponsoring manager is critical to get the team started on a positive and constructive manner. The sponsor must provide complete information, establish a level of trust that enables the group to be open and honest, and explain roles and expectations. Stage : Storming This second stage is characterized by the interpersonal conflicts that arise among team members while clarifying their tasks, roles, responsibilities, expectations, and organizational issues. This phase is characterized by debate, arguments, conflict, and perhaps hostility and open warfare. Care must be taken by the sponsoring manager to assure that discussion remains constructive and is directed at the issued and not at individuals, and that all issues are surfaced, debated, and clarified. Managers at this stage will draw on their skills as facilitators. Obective information gathering and decision making will be aided by techniques such as brainstorming, weighted voting, and consensus building. During this second stage, little progress will be made toward the team s goals; instead, energy will be directed toward the team s processes. This stage, however, must be successfully completed before real work can begin. Stage 3: Norming This third stage is marked by a coming together of team members. Cohesion is exhibited through a healthy flow of opinions, sharing of personal experience and data, cooperation, and overall good work. At this point, members feel good about their involvement on the team and feel that their work has purpose and value. The group of individuals starts to evolve into an effective team and begins to work together toward achieving its goal. The sponsoring manager at this point may feel that the team is self-sufficient, fully capable of managing its activities. The role of the manager changes from being a director or referee to being a participant or resource. The manager may even withdraw and allow the team to define its own leaders and administrators. Stage 4: Performing This stage is the payoff. Team members are performing their roles. Work is characterized as cooperative and collaborative, and the team s obectives are being achieved. Teams at this stage are mature and self-directing and exhibit a natural sharing of roles and responsibilities. They effectively use systematic processes. They achieve results and bring innovative solutions to the organization. The sponsoring manager in this phase may well see the makings of an autonomous work team in which recommendations are developed and implemented without prior approval. Here, the manager s role is that of a coach. The manager helps the team improve the process by which it works. The manager no longer needs to allocate and monitor work, since the team is fully 103

104 capable of doing this itself. The manager works with the team on its process; the team works on providing the results. IV.9.f. Stages of application The following table lists how employees can progress through phases of development as a team addresses the need to improve quality or productivity. The task for managers is to foster this progression among all employees. XIII. Table Process improvement step Low involvement 1. Define problem Managers defines customers, requirements, and necessary processes. Identify and document process 3. Measure performance Manager or staff documents processes Managers collect data but shares little 4. Understand why Manager analyses data and draw conclusions 5. Develop and test ideas 6. Implement solutions and evaluate Manager defines improvements Manager drives implementation and evaluates outcome Degree of involvement Group proposes opportunity, manager approves Group works with staff to document processes Manager defines data for group to collect Manager suggests casual factors and group confirms Manager suggests alternatives for group to evaluate Group implements solution and shares evaluation with manager High involvement Team defines opportunity as next stage in continuous improvement Team documents and flowcharts processes Team determines what data to collect and gets them Team analyses data, determines root causes, and understands variation Team identifies and tests alternative solutions Team has responsibility and authority to implement solution and track results When employees throughout the organization are trained in the tools and techniques of process improvement, and when managers provide information, coaching, and resources, then an organization begins to take on a different set of characteristics. These are displayed on the following table for the entire organization and represent the macromodel of an organization whose employees become empowered. 104

105 XIV. Table Organizational phase Traditional Transitional High-involvement Structure Focus Authority Idea sources Stake hierarchical precise ob descriptions functional units internal targets preservation costs problem solving find-and-fix Top-down command inflexible controlling rank and title work measurement suggestion systems apathetic no ownership less hierarchical loose ob descriptions matrix management competition quality and productivity adaptation product service improvement detection special assignments open to challenge sharing committee staff studies quality circles compliant some ownership flat no ob descriptions self-directed teams Customers needs Flexible, responsive customer satisfaction process improvement prevention consensus seeks challenge trusting knowledge work teams customers and employees committed full ownership IV.9.g. Managers responsibilities Managers should recognize that not every group will become a high-performing, autonomous work team. Yet teams may still make valuable contributions as they progress through the stages of forming, storming and norming. Team progress is attributable in part to the sponsoring manager s effectiveness in guiding and supporting his or her teams. Managers must also understand that empowering others and delegating authority requires a great deal of planning and hard work on their part. Building an environment to support empowerment requires teamwork as well as leadership. However, providing both of these ingredients may still not free teams and individuals to make decisions for themselves. If the organization still does not respond, if middle managers are not being supportive, or if teams are not confronting and resolving issues in their areas, then an additional stimulant is needed. One form of stimulant can be provided by senior managers fabricating a minicrisis, if a legitimate one does not already exist. This is simply done by increasing the span of control and the responsibilities of managers at all levels. It soon becomes apparent, that the traditional way of managing have to change. Another stimulant to support change is to drive the organization to meet a new challenge one that is specific, measurable, and of recognized importance. For example, achieving 100 per cent performance on every target should be the new goal. No defects, no errors, 100 per cent accurate invoices, every product and service delivered on time, every time, 100 per cent customer satisfaction are all targets that will at first appear to be impossible. But setting 105

106 targets less than 100 per cent may artificially cap performance. More important, stretch targets will signal to everyone that we have to change the way we do business because we cannot achieve these stretch targets by doing the same old thing. IV.9.h. What opportunities will teams address? The generation of ideas on which teams would act has historically been in the domain of managers and their supporting staffs and specialists. But what about tapping the reservoir of ideas stored in the minds of those who are closest to the work and closest to the customers? What about tapping the resources who are best positioned to see the reality of day-to-day problems and who will be responsible for implementing improvements? The empowered work force captures the ideas within the work group. Systems are designed to stimulate the generation, development, and implementation of these ideas. Toyota serves as a benchmark for employee suggestion systems. By the early 1980s, Toyota was getting about 1.6 million ideas from employees annually. This represented an average of about 30 ideas from every employee. Furthermore, about 95 per cent were implemented. There are four simple guidelines for building effective suggestion programs: 1. Align: Employees who are committed to support the organization s goals are more likely to contribute good ideas. Implementing these suggestions not only benefits the organization but reinforces the suggestion-system process.. Implement: Don t analyze suggestions to death. Accepting people s ideas will encourage them to offer more. Don t expect that all ideas will be perfect; most will require work to fine-tune and develop. 3. Reward: Contributors can be rewarded in two ways. The first is for people to see that their own ideas are valued; they are acted on and implemented. The second is to receive recognition, awards, and increased compensation. But act quickly, the value of positive feedback diminishes as time increases. 4. Count: measure the performance of your suggestion system. How many ideas are submitted? How many are implemented? Work to improve the performance of this business process in the same way as you would improve any other. Post performance numbers to bring the system to everyone s attention. IV.9.i. Summary In the empowered work force, we will tap the discretionary effort of all employees. Everyone will understand what the organization is trying to accomplish and what his or her role is in this endeavour. Moreover, everyone will be committed to contribute toward these goals. Aligned employees will have the requisite capabilities in an environment built on mutual trust. Teamwork will be supported. Team members, leaders and facilitators will understand the factors that contribute to success and will recognize the natural evolution of teams through four basic stages: forming, storming, norming and performing. The organization s senior 106

107 members will shift from being commanders and controllers toward becoming coaches and teachers. Ideas will be generated from all levels and particularly from those who are closest to the work processes and customers. These ideas will be consistent with the organization s needs, and an overwhelming maority will find rapid implementation. Success will feed on success and stimulate a chain reaction benefiting everyone. 107

108 V. EFQM Excellence Model EFQM (the European Foundation for Quality Management) is a not-for-profit membership foundation in Brussels, established in 1989 to increase the competitiveness of the European economy. The initial impetus for forming EFQM was a response to the work of W. Edwards Deming and the development of the concepts of Total Quality Management. What does business excellence mean? Business excellence focuses on the client and requires inspirational leadership as well as systems, processes and facts to manage an organisation. But for everything to happen there is a need to maximise employee s contribution by helping them to develop and grow. Business excellence means that excellence is achieved in everything an organisation does including its: customers, strategy, leadership, people, partners, society, processes, products and services. V.1. Fundamental concepts How does EFQM define excellence? Excellent Organisations achieve and sustain outstanding levels of performance that meet or exceed the expectations of all their stakeholders. We can all think of organisations that we would recognise as being excellent. They may well operate in different environments, with different stakeholder constituencies, and come in all shapes and sizes but what they share a common mindset that is based on a number of attributes and ways of working that separate them from the crowd. The Fundamental Concepts of Excellence outline the foundation for achieving sustainable excellence in any organisation. They can be used as the basis to describe the attributes of an excellent organisational culture. They also serve as a common language for top management. These eight Concepts have been identified through a rigorous process that included benchmarking globally, searching extensively for emerging management trends and, last but not least, a series of interviews with senior executives from a cross-section of industries operating across Europe. Each of the Concepts is important in its own right but maximum benefit is achieved when an organisation can integrate them all into its culture. 1. Adding Value for Customers Excellent organisations consistently add value for customers by understanding, anticipating and fulfilling needs, expectations and opportunities.. Creating a Sustainable Future Excellent organisations have a positive impact on the world around them by enhancing their performance whilst simultaneously advancing the economic, environmental and social conditions within the communities they touch. 3. Developing Organisational Capability 108

109 Excellent organisations enhance their capabilities by effectively managing change within and beyond the organisational boundaries. 4. Harnessing Creativity & Innovation Excellent organisations generate increased value and levels of performance through continual improvement and systematic innovation by harnessing the creativity of their stakeholders. 5. Leading with Vision, Inspiration & Integrity Excellent organisations have leaders who shape the future and make it happen, acting as role models for its values and ethics. 6. Managing with Agility Excellent organisations are widely recognised for their ability to identify and respond effectively and efficiently to opportunities and threats. 7. Succeeding through the Talent of People Excellent organisations value their people and create a culture of empowerment for the achievement of both organisational and personal goals. 8. Sustaining Outstanding Results Excellent organisations achieve sustained outstanding results that meet both the short and long term needs of all their stakeholders, within the context of their operating environment. The Fundamental Concepts of Excellence form the basis for the criteria of the EFQM Excellence Model. V.1.a. Model Criteria 15. Figure The beauty of the Model is that it can be applied to any organisation, regardless of size, sector or maturity. It is non-prescriptive and it takes into account a number of different concepts. It provides a common language that enables our members to effectively share their knowledge and experience, both inside and outside their own organisation. It ensures that all the management practices used by an organisation form a coherent system that is continually improved and delivers the intended strategy for the organisation. 109

110 The EFQM Excellence Model is based on nine criteria. Five of these are "Enablers" and four are "Results". The "Enabler" criteria cover what an organisation does and how it does it. The "Results" criteria cover what an organisation achieves. To achieve sustained success, an organisation needs strong leadership and clear strategic direction. They need to develop and improve their people, partnerships and processes to deliver value-adding products and services to their customers. In the EFQM Excellence Model, these are called the Enablers. If the right Enablers are effectively implemented, an organisation will achieve the Results they, and their stakeholders, expect. The arrows emphasise the dynamic nature of the Model, showing learning, creativity and innovation helping to improve the Enablers that in turn lead to improved Results. Each of the nine criteria has a definition, which explains the high level meaning of that criterion. To develop the high level meaning further each criterion is supported by a number of criterion parts. Criterion parts are statements that describe in further examples of what, typically, can be seen in excellent organisations and should be considered in the course of an assessment. Finally, below each criterion part are guidance points. Many of these guidance points are directly linked to the Fundamental Concepts. Use of these guidance points is not mandatory. They are intended to give examples to aid interpretation of the criterion part. 16. Figure V.. Enablers What an organisation does and how it does it There are five Enablers, pictured on the left-hand side of the Model. These are the things an organisation needs to do to develop and implement its strategy. V..a. Leadership Excellent organisations have leaders who shape the future and make it happen, acting as role models for its values and ethics and inspiring trust at all times. They are flexible, enabling the 110

111 organisation to anticipate and reach in a timely manner to ensure the on-going success of the organisation. V..b. Strategy Excellent organisations implement their Mission and Vision by developing a stakeholder focused strategy. Policies, plans, obectives and processes are developed and deployed to deliver the strategy. V..c. People Excellent organisations value their people and create a culture that allows the mutually beneficial achievement of organisational and personal goals. They develop the capabilities of their people and promote fairness and equality. They care for, communicate, reward and recognise, in a way that motivates people, builds commitment and enables them to use their skills and knowledge for the benefit of the organisation. V..d. Partnerships & Resources Excellent organisations plan and manage external partnerships, suppliers and internal resources in order to support their strategy, policies and the effective operation of processes. They ensure that they effectively manage their environmental and societal impact. V..e. Processes, Products & Services Excellent organisations design, manage and improve processes, products and services to generate increasing value for customers and other stakeholders. V.3. Results What an organisation achieves There are four Results areas, shown on the right-hand side of the Model. These are the results an organisation achieves, in line with their strategic goals. In all four results areas, we find that excellent organisations: Develop a set of key performance indicators and related outcomes to determine the successful deployment of their strategy, based on the needs and expectations of the relevant stakeholder groups Set clear targets for key results, based on the needs and expectations of their business stakeholders, in line with their chosen strategy Segment results to understand the performance of specific areas of the organisation and the experience, needs and expectations of their stakeholders Demonstrate positive or sustained good business results over at least 3 years Clearly understand the underlying reasons and drivers of observed trends and the impact these results will have on other performance indicators and related outcomes 111

112 Have confidence in their future performance and results based on their understanding of the cause and effect relationships established Understand how their key results compare to similar organisations and use this data, where relevant, for target setting V.3.a. Customer Results Excellent organisations achieve and sustain outstanding results that meet or exceed the need and expectations of their customers. V.3.b. People Results Excellent organisations achieve and sustain outstanding results that meet or exceed the need and expectations of their people. V.3.c. Society Results Excellent organisations achieve and sustain outstanding results that meet or exceed the need and expectations of relevant stakeholders within society. V.3.d. Business Results Excellent organisations achieve and sustain outstanding results that meet or exceed the need and expectations of their business stakeholders. V.4. Radar logic 17. Figure Structured approach to questioning the performance on an organisation. The RADAR logic is a dynamic assessment framework and powerful management tool that provides a structured approach to questioning the performance of an organisation. At the highest level, RADAR logic states that an organisation needs to: Determine the Results it is aiming to achieve as part of its strategy 11

113 What are we trying to achieve? Plan and develop an integrated set of sound Approaches to deliver the required results both now and in the future How do we try to achieve this? Deploy the approaches in a systematic way to ensure implementation How / Where / When was this implemented? Assess and Refine the deployed approached based on monitoring and analysis of the results achieved and ongoing learning activities. How do we measure whether it is working? What have we learning and what improvements can be made? RADAR is a simple but powerful management tool that can be applied in different ways to help drive continuous improvement: Assessing the maturity of the approaches you have implemented Assessing the excellence of the results achieved Helping to structure improvement proects To help support robust analysis, the RADAR elements can be broken down into a series of attributes which contain guidance on what we expect the organisation to demonstrate. V.4.a. Assessing Enablers 18. Figure When assessing Enablers, we look at the approaches adopted, how they have been deployed and how the organisation assesses and refines their efficiency & effectiveness over time. In practical terms, this means we look for: Sound and integrated approaches that support the achievement of the organisation's strategy 113

114 Structured deployment within all relevant areas of the organisation that enables refinements to be implemented within appropriate timescales Measurement being carried out so the organisation can understand how well the approach is working and how effectively it has been implemented Learning activities being undertaken to identify alternative or new ways of working Improvements being implemented as a result of measurement and learning (closing the loop) V.4.b. Assessing Results When assessing Results, we first look at their Relevance to the organisation's strategy and how useful they are in reviewing progress against these key obectives. When then look at the performance of the results themselves. In practical terms you should be looking for: Results which clearly show how the organisation is progressing against its key strategies for the criterion Reliable data that can be segmented to give a clear understanding of what's happening in relation to different stakeholder groups, products or processes. Positive trends over a 3 year period Targets, which are appropriate for the strategic obectives, being achieved Appropriate comparisons and benchmarks to put the results in context within the organisation's operating environment Evidence to show that the organisation understands the underlying drivers of the results and effectively managing them to ensure that performance levels will be sustained into the future 19. Figure 114

115 VI. Six Sigma The Six Sigma is a statistically based approach which reduces variability, removes defects and prevents products to be waste. Hence, the statistical methods have a big role to play in this. It uses a set of quality management method including statistical methods to find a group of experts which are familiar with these very complex methods. Each Six Sigma proect has a sequence of steps defined and all experts should follow the sequence to achieve the financial target of the proect. This target can be different like reducing cost or increasing profit. Actually, Six Sigma originally defined to reduce defects in output products, but nowadays it extended to other businesses and defects can be defined as any process output which does not meet customer specifications. In Six Sigma approach, the goal is to have % of the products free of defects (3.4 defects per million). Six sigma proects typically last 4-6 months and during this time, it uses a specific five-step methodology known as DMAIC. DMAIC process breaks down the proect into five phases: Define opportunity Measure performance Analyze opportunity Improve performance Control performance DMAIC is coming from the first letters of these five steps. This methodology guides the whole team from the beginning to the end result. The important point is that Six Sigma is applied on a specific part or characteristic and improves the performance of that special part. So the total product is not the goal of Six Sigma proect. VI.1. DMAIC Process DMAIC is an acronym for Define, Measure, Analyze, Improve and Control. DMAIC is a separate concept and can be apply in any different approaches, meaning that it is not ust for Six Sigma proects. However, almost all of Six Sigma proects use DMAIC for improvement. This is a general approach to help people to gather all tools and get the quality improved. VI.1.a. Define In Define step the proect opportunity should be identified. In this step the team needs to provide a proect charter and an action plan. The proect charter is a short document up to two pages which contains a description of the proect and its scope, the start and the anticipated completion dates, an initial description of both primary and secondary metrics that will be used to measure success and how those metrics align with business unit and corporate goals, the potential benefits to the customer, the potential financial benefits to the organization, milestones that should be accomplished during the proect, the team members and their roles, and any additional resources that are likely to be needed to complete the proect. To prepare 115

116 the proect charter the proect sponsor has a significant role. Generally, it takes -4 days to prepare this document. As mentioned before, the team should also prepare an action plan for moving forward to the other DMAIC steps specially Measure step which is going to be done in the next step. It contains individual work assignments and tentative completion dates. Finally, in Define step the team should focus on the following: Does the problem statement focus on symptoms, and not on possible causes or solutions? Are all the key stakeholders identified? Does the potential financial impact make the proect worth doing? Has the scope of the proect been verified to ensure that it is neither too small nor too large? Has a high-level process map been completed? Have any obvious barriers or obstacles to successful completion of the proect been ignored? Is the team s action plan for the Measure step of DMAIC reasonable? VI.1.b. Measure The goal of this step is to evaluate and find the current state of the process, so this involves collecting data. KPIV and KPOV concepts are abbreviations for key process input variables and key process output variables, respectively and this is important to develop a list of all of them. They may have been identified in Define step but must be carefully define and measure during Measure step. For understanding and analysis of current process performance, we need to have sufficient data and deciding what and how much data to collect is an important task. Data may be collected by examining historical records, but this may not always be satisfactory, as the history may be incomplete, the methods of record keeping may have changed over time, and, in many cases, the desired information may not have been retained. Consequently, it is often necessary to collect current data through an observational study, or it may be done by sampling from the relevant data streams. The data that are collected are used as the basis for determining the current state or baseline performance of the process. Additionally, the capability of the measurement system should be evaluated to make sure the team is not trying to solve an imaginary problem in which the process performance is fine, but the measurement system is faulty. At the end of the Measure step, the team should update the proect charter (if necessary), reexamine the proect goals and scope, and re-evaluate team makeup. The team may consider expanding to include members of downstream or upstream business units if the Measure activities indicate that these individuals will be valuable in subsequent DMAIC steps. Any issues or concerns that may impact proect success need to be documented and shared with the process owner or proect sponsor. Finally, it is necessary to prepare the followings at the end of Measure step: 116

117 There must be a comprehensive process flow chart or value stream map. All maor process steps and activities must be identified, along with suppliers and customers. If appropriate, areas where queues grow and work-in-process accumulates should be identified and queue lengths, waiting times, and work-in-process levels reported. A list of KPIVs and KPOVs must be provided, along with identification of how the KPOVs related to customer satisfaction or the customers CTQs (CTQ= Critical to Quality). Measurement systems capability must be documented. Any assumptions that were made during data collection must be noted. The team should be able to respond to requests such as, Explain where that data came from and questions such as, How did you decide what data to collect?, How valid is your measurement system?, and Did you collect enough data to provide a reasonable estimate of process performance? VI.1.c. Analyze In the Analyze step, the obective is to use the data from the Measure step to begin to determine the cause-and-effect relationships in the process and to understand the different sources of variability. That is, in the Analyze step the obective is to determine the potential causes of the defects, quality problems, customer issues, cycle time and throughput problems, or waste and inefficiency that motivated the proect. It is important to separate the sources of variability into common causes and assignable causes. Removing a common cause of variability usually means changing the process, while removing an assignable cause usually involves eliminating a specific problem. A common cause of variability might be inadequate training of personnel processing insurance claims, while an assignable cause might be a tool failure on a machine. There are many statistical tools that are potentially useful in the Analyze step. Among these are graphical data displays, control charts, hypothesis testing and confidence interval estimation, regression analysis, designed experiments, and failure modes and effects analysis. The Analyze tools are used with historical data or data that was collected in the Measure step. These data are often very useful in providing clues about potential causes of the problems that the process is experiencing. Sometimes these clues can lead to breakthroughs and actually identify specific improvements. In most cases, however, the purpose of the Analyze step is to explore and understand tentative relationships between and among process variables and to develop insight about potential process improvements. A list of specific opportunities and root causes that are targeted for action in the Improve step should be developed here. Improvement strategies will be further developed and actually tested in the Improve step. To finish the Analyze step, the team should consider the following issues and potential questions: What opportunities are going to be targeted for investigation in the Improve step? What data and analysis supports that investigating the targeted opportunities and improving/eliminating them will have the desired outcome on the KPOVs and customer CTQs that were the original focus of the proect? Are there other opportunities that are not going to be further evaluated? If so, why? 117

118 Is the proect still on track with respect to time and anticipated outcomes? Are any additional resources required? VI.1.d. Improve In the Measure and Analyze steps, the team decides which KPIVs and KPOVs to study, determines what data to collect and how to display and analyze them, identifies potential sources of variability, and determines how to interpret the data they obtained. In the Improve step, they turn to creative thinking about the specific changes that can be made in the process and other things that can be done to have the desired impact on process performance. A broad range of tools can be used in the Improve step. Designed experiments are probably the most important statistical tool in the Improve step. Designed experiments can be applied either to an actual physical process or to a computer simulation model of that process, and can be used both for determining which factors influence the outcome of a process and for determining the optimal combination of factor settings. The obectives of the Improve step are to develop a solution to the problem and to pilot test the solution. The pilot test is a form of confirmation experiment; it evaluates and documents the solution and confirms the solution attains the proect goals. This may be an iterative activity, with the original solution being refined, revised, and improved several times as a result of the pilot test s outcome. The Improve step should provide the following: Adequate documentation of how the problem solution was obtained. Documentation on alternative solutions that were considered. Complete analysis results for the pilot test. Plans to implement the pilot test results on a full-scale basis. This should include dealing with any regulatory requirements, legal issues, personnel concerns (such as additional training requirements), or the impact on other business standard practices. Analysis of any risks of implementing the solution, and appropriate risk-management plans. VI.1.e. Control The obectives of the Control step is to complete all remaining work on the proect and to deliver the improved process to the process owner along with a process control plan and other necessary procedures to ensure that the gains from the proect will be institutionalized. The process owner should receive before-and-after data on key process metrics, operations and training documents, and updated current process maps. The financial benefits of the proect should be quantified. The process control plan should be a system for monitoring the solution that has been implemented, including methods and metrics for periodic auditing. Control charts are an important statistical tool used in the control step of DMAIC; many process control plans involve control charts on critical process metrics. The transition plan for the process owner should include a validation check several months after proect completion. It is important to ensure that the original results are still in place and stable so that the positive financial impact will be sustained. It is not unusual to find that something has gone wrong in 118

119 the transition to the improved process. The ability to respond rapidly to unanticipated failures should be factored into the plan. The Control step typically includes the following issues: Data illustrating that the before and after results are in line with the proect charter should be available. (Were the original obectives accomplished?) Is the process control plan complete? Are procedures to monitor the process, such as control charts, in place? Is all essential documentation for the process owner complete? A summary of lessons learned from the proect should be available. A list of opportunities that were not pursued in the proect should be prepared. This can be used to develop future proects; it is very important to maintain an inventory of good potential proects to keep the improvement process going. A list of opportunities where the results of the proect can be used in other parts of the business. 119

120 VII. Quality management tools and methods VII.1. What causes defectives? One after another, products are arriving on the conveyor. At the end of the conveyor, there is a packaging machine which continuously packs up the incoming products and sends them to the product warehouse. A closer look reveals a man standing between the conveyor and the packaging machine. He is keeping a watchful eye on the flow of the products, and occasionally picks up products and casually throws them into a basket behind him. Those products are defective. This kind of thing is commonly seen in many factories. At first, those thrown-away products seemed to be waste of goods, but soon they are accepted as a routine process. But becoming accustomed to defectives do not solve the problem but rather a steps backward from the solution. How are defectives made in the first place? What should be done to reduce their occurrence? In order to decrease the number of defectives, one need to believe that defectives can definitely be reduces. There are peculiar causes for any given defective product, and defectives can be got rid of those causes are discovered and removed. Most people feel about defectives is that because their products must satisfy very strict quality standards and have numerous defect-causing factors, defective products are unavoidable. However, regardless of types of products or kinds of production method used, causes of defectives are universal. Variation: this is the cause. What will happen if we make products using materials of the exact same quality, identical machines and work methods and inspect these products in exactly the same manner? No matter how many products are made, they must all be identical as long as the above four conditions are identical. That is, the products will be either all conforming or all non-conforming. All of them will be defective if materials, machinery method of work or that of inspection is not proper. In such a case, exactly identical defectives will invariably be produced. As long as there is no failure in the aforementioned four conditions, the resulting products must all be identically non-defective ones. As far as the products we make are concerned, it is almost impossible that every product turns out to be defective. Of the products made, some are defective while others are not. In other words, defective and non-defective products come mixed together. Why are defectives and non-defectives produced together? The cause, as we stated before, is variation. Variations in materials, in machinery conditions, in work methods and in inspections are the causes of defectives. If none of these variations existed, all products would be identical and there would be no varying of quality like the occurrence of defectives and non-defectives. Looking at the problem, we can see that in the process of making one product there are countless factors which affect the quality characteristics of that product. When we regard manufacturing process from the viewpoint of quality variation, we can think of the process as an aggregate of the causes of variation. These causes are the explanation of the changes in 10

121 quality characteristics of products, making defectives or non-defectives. A product is udged to be non-defective if its quality characteristics meet certain standard and defective if they do not. Therefore, even non-defectives have variations within their standard. This means that these are not the exactly equal products which we discussed earlier. Defectives are caused by variations. If these variations are reduced, defectives will certainly decrease. This is a simple, strong principle which holds true regardless of types of product or kinds of production method involved. VII.. Diagnosis of processes Although causes of quality variations are countless, not every cause affects quality to the same degree. Some of them actually affect quality greatly while others, although considered to be very important in theory, have very little effect on quality variation when they are properly controlled. The countless conceivable causes can be categorized into two groups first of which consists of a small number of causes which nevertheless give a great effect (the vital few), and a second group which is made up of many causes giving only minor effects (the trivial many). Usually, there are not many factors which really cause defects. This fact is called the principal of Pareto because it applies to many instances. By applying the aforementioned principle variation and this principle of Pareto, the problem of reducing defectives becomes considerably easier to tackle with. What we need to do first is to find the vital few causes of defectives, and remove these causes after they have been clearly identified. The procedure of finding the causes of defectives from among many other factors is called the diagnosis of process. In order to reduce the number of defectives, the first necessary action is to make a correct diagnosis to see what the true causes of defectives are. If this is not done correctly, defectives cannot be reduced. How do we make a correct diagnosis? There are various methods. Some employ intuition, others depend on experience. Still others involve statistical analyses on data, while one can also use experimental research. The intuitive method is often used because it can be done very quickly. In fact, there is something beyond ordinary man s ability in the intuition of a true expert. The difficulty of in the problem of reducing defectives is that it is not always clear who a real expert is. Moreover, in the time of rapid progress, it is difficult to remain an expert in all the problems whose nature is constantly and undergoing change. As problems of defectives are often found in areas where previous experience is lacking, what is needed is not so much the years of experience as a strong will to reduce defectives and an attitude of observing the real situation in an obective way. The statistical way of looking at things and use of statistical methods are a most effective means for this observation. Statistical methods provide a very effective means for the development of new technology and quality control in manufacturing processes. 11

122 VII.3. How to obtain data VII.3.a. How to collect data Have clear defined obectives Data is a guide for our actions. From data we learnt pertinent facts, and take appropriate actions based on such facts. Before collecting data, it is important to determine what we are going to do with is. In a machine factory, a sampling inspection was made in a certain type of incoming purchased part. A lot which should be reected in itself was accepted as a special exception to keep the production schedule. However, they didn t do anything special about the accepted lot. This means that both the lots which conform to specifications and those which do not went to the next process. These data were certainly being taken to determine the acceptability of lots, but they were not utilized at all. In quality control, the obectives of collecting data are: controlling and monitoring the production process analysis of non-conformance inspection. What is the purpose? Once the obect for collecting data is defined, the types of comparison which need to be made are also determined, and this in turn identifies the type of data which should be gathered. For example, suppose that there is a question involving variation on a quality characteristic of a product. If only one datum is gathered per day, it is impossible to determine the variation within a day. Or, if you want to know in what ways defectives are produced by two workers, it is necessary to take their samples separately so that the performance of each worker can be compared. If comparing one against the other reveals a clear difference, a remedial measure which will eliminate the difference between workers will also reduce the variation in the process. Dividing a group in this way into several subgroups on the basis of certain factor is called stratification. Stratification is very important, it is necessary to make it a habit to apply stratification in your thinking in all kinds of situations. Then, suppose you want to know the relationship between the amount of ingredient and the hardness of the product. In a case like this, where you want to know whether there is a relationship between the values of two characteristics, the data have to be available in pairs. Are measurements reliable? Even if the samples have been taken properly, a wrong udgement will be made if the measurement itself is unreliable. For example, inspections made by a certain inspector showed a fraction defective which was very different from the rest, and careful examination later revealed that a measuring instrument had gone wrong. 1

123 In the case of a sensory measurement such as visual inspection, differences due to individual inspectors are very common. This fact must be taken into account when collecting and analysing data. Find right ways to record data Once data is gathered, various statistical methods are used for analysing them, so that it will become a source of information. When collecting data, it is important to arrange it neatly to facilitate later processing. First of all, the origin of the data must be clearly recorded. Data whose origin is not clearly known becomes dead data. Quite often, little useful information is obtained despite the fact that a week was spent gathering data on quality characteristics, because people forgot on what days of the week the data was collected, which machines did the processing, who the workers were, which material lots were involved and so on. Secondly, data should be recorded in such a way that it can be used easily. Since data is often used later to calculate statistics such as means and ranges, it is better to write it down in a manner which facilitate computations. A set of standard recording forms should be prepared beforehand if data is to be collected on a continuous basis. VII.4. Check sheets As stated in the preceding section, if data is to collected at all, it is essential to make the purpose clear and to have data which clearly reflects the facts. In addition to these premises, in actual situations it is important that the data should be gathered in a simple way and in an easy-to-use form. A check sheet is a paper form on which items to checked have been printed already so that data can be collected easily and concisely. Its main purposes two-fold: 1. to make data gathering easy;. to arrange data automatically so that they can be used easily later on. The collecting and recording data seems easy but actually is difficult. Usually, the more people process the data, the more writing errors are likely to arise. Therefore, the check sheets, on which data can be recorded by means of check marks or simple symbols and on which data is arranged automatically without further copying by hand, becomes a powerful data recording tools. 13

124 0. Figure VII.5. What are Pareto Diagrams? Quality problems appear in the form of loss (defective items and their cost). It is extremely important to clarify the distribution pattern of the loss. Most of the loss will be due to a very few types of defect, and these defects can be attributed to a very small number of causes. Thus, if the causes of these vital few defects are identified, we can eliminate almost all the losses by concentrating on these particular causes, leaving aside the other trivial many defects for the time being. By using the Pareto diagram, we can solve this type problem efficiently. In 1897, the Italian economist V. Pareto presented a formula showing that the distribution of income is uneven. A similar theory was expressed diagrammatically by the U.S. economist M.C Lorenz in Both of these scholars pointed out that by far the largest share of income or wealth is held by a very small number of people. Meanwhile, in the field of quality control, Juran applied Lorenz s diagram method as a formula in order to classify problems of quality into the vital few and the trivial many, and named this method Pareto Analysis. He pointed out that in many cases, most defects and the costs of these arise from a relatively small number of causes. VII.6. How to make Pareto diagram? Step 1 Decide what problems are to be investigated and how to collect the data. Step Decide what kind of problems you want to investigate. (Example: defective items, losses in monetary terms, accidents occurring.) Decide what data will be necessary and how to classify them. (Example: by type of defect, location, processes, machine, worker, method.) Determine the method of collecting the data and the period during which it is to be collected. 14

125 Design a data tally sheet listing the items, with space to record their totals. XV. Table Type of defect Tally Total Crack ///// ///// 10 Scratch ///// ///// ///// // 4 Strain ///// / 6 Stain ///// ///// ///// ///// //// 104 Gap //// 4 Pinhole ///// ///// ///// ///// ///// 0 Others ///// ///// //// 14 Total 00 Step 3 Fill out the tally sheet and calculate the totals. Step 4 Make Pareto diagram data sheet listing the items, their individual totals, cumulative totals, percentages of overall total, and cumulative percentages. Type of defect Number of defects Cumulative total Percentage of overall total strain (D) scratch (B) pinhole (F) crack (A) stain (C) gap (E) others total Cumulative percentage Step 5 Arrange the items in the order of quantity and fill out the data sheet. The item others should be placed in the last line, no matter how large it is. This is because it is composed of a group of items each of which is smaller than the smallest item listed individually. Step 6 Draw two vertical axes and horizontal axis. 15

126 Vertical axes: Left-hand vertical axes: mark this axis with a scale from 0 to the overall total. Right-hand vertical axes: mark this axis with a scale from 0% to 100%. Horizontal axis: Divide this axis into the number of intervals to the number of items classified. Step 7 Construct a bar diagram. Pareto Diagram by defective items Number of defective units 1. Figure Step 8 Draw the cumulative curve (Pareto curve). Mark the cumulative values (cumulative total or cumulative percentage), above the right-hand intervals of each item, and connect the points by a solid line. Pareto Diagram by defective items Cumulative percentage 0. Figure Step 9 Write any necessary items on the diagram (items concerning the diagram and items concerning the data). 16

127 VII.6.a. Pareto Diagrams by phenomena and Pareto Diagrams by causes As already mentioned, a Pareto-diagram is a method of identifying the vital few, and there are two types. Pareto diagrams by phenomena This is a diagram concerning the following undesirable results, and is used to find out what the maor problem is. Quality: defects, faults, complaints, returned items, repairs Cost: amount of loss, expenses Delivery: stock shortages, defaults in payments, delays in delivery Safety: accidents, mistakes, breakdowns Pareto diagrams by causes This is a diagram concerning causes in the process, and is used to find out what the maor cause of the problem is. Operator: shift, group, age, experience, skill, individual person Machine: machines, equipment, tools, organizations, models, instruments. Raw material: manufacturer, plant, lot, kind. Operation method: conditions, orders, arrangements, methods. VII.6.b. Notes on Pareto diagrams Hints on making Pareto diagrams Check various classifications and construct many kinds of Pareto diagrams. You can grasp the essence of various angles, and it is necessary to try out various methods of classification until you identify the vital few, which is the aim of Pareto analysis. It is undesirable that others represent a higher percentage. If this happens, it is because the items for investigation are not classified appropriately and too many items fall under this heading. In this case, a different method of classification should be considered. If a monetary value can be assigned to the data, it is best to draw the Pareto diagrams with the vertical axis showing this. If the financial implications of a problem are not properly appreciated, the research itself may end up as ineffective. Cost is an important scale of measurement in management. Hints on using Pareto diagrams If an item is expected to be amenable to a simple solution, it should be tackled right away even if it is of relatively small importance. Since a Pareto diagram aims at efficient problem solving, it basically requires us to tackle only the vital few. However, if an item which appears to be of relatively small importance is expected to 17

128 be solved by a simple countermeasure, it will serve as an example of efficient problem-solving, and the experience, information and incentives to morale obtained through this will be of great assets for future problem-solving. Do not fail to make a Pareto diagram by causes. After identifying the problem by making a Pareto diagram by phenomena, it is necessary to identify the causes in order to solve the problem. It is therefore vital to make a Pareto diagram by causes if any improvements are to be effected. VII.7. Cause-and-effect diagram The output or result of the process can be attributed to a multitude of factors, and a cause-andeffect relation can be found among those factors. We can determine the structure or a multiple cause-and-effect relation by observing it systematically. It is difficult to solve complicated problems without considering this structure, which consists of a chain of causes and effects, and a cause-and-effect diagram is a method of expressing it simply and easily. A cause-and-effect diagram is a diagram which shows the relation between a quality characteristic and factors. The diagram is used not only for treating the quality characteristic of products, but also in other fields, and has found application worldwide. VII.7.a. How to make cause-and-effect diagrams? Making a useful cause-and-effect diagram is no easy task. It may safely be said that those who succeed in a problem-solving in quality control are those who succeed in making a useful cause-and-effect diagram. There are many ways of making the diagram, but there are two typical methods. Prior to introducing the procedures, the structure of the cause-and-effect diagram is explained with an example. Structure of cause-and-effect diagrams A cause-and-effect diagram is also called a fishbone diagram since it looks like the skeleton of a fish, as shown in the following figure. 18

129 Big bone Small bone Medium-sized bone Characteristic Characteristic (effect) Factors (causes) 3. Figure Procedures for making cause-and-effect diagrams for identifying causes Step 1 1. Procedure Determine the quality characteristic. Step Choose one quality characteristic and write it on the right-hand side of a sheet of paper, draw in the backbone from left to right, and enclose the characteristic in a square. Next, write the primary causes which affect the quality characteristic as big bones also enclosed by squares. Step 3 Write the causes (secondary causes) which affect the big bones (primary causes) as mediumsized bones, and write the causes (tertiary causes) which affect the medium-sized bones as small bones. Step 4 Assign an importance to each factor, and mark the particularly important factors that seem to have a significant effect on the quality characteristic. Step 5 Record any necessary information.. Explanation of the procedure You may often find it difficult to proceed when you practice this approach. The best method in such a case is to consider the variation. For example, consider variation in the quality characteristic when you are thinking about the big bones. If the data shows that such a 19

130 variation exists, consider why it exists. A variation in the effect must be caused by variation in the factors. This kind of switch-over of thought is extremely effective. When you are making a cause-and-effect diagram relating to a certain defect, for example, you may discover that there is a variation in the number of defects occurring on different days of a week. If you find that the defect occurred more frequently on Monday than on any other day of the week, you can change your thinking as follows: Why did the effect occur? Why did the effect occur more frequently on Monday than on any other day of the week? This will lead you to look for factors which make Monday and other days different, eventually leading to discover the cause of the effect. By adopting this method of thinking at each stage of examining the relation between the characteristic and the big bones, the big bones and the medium-sized bones, and the mediumsized bones and the small bones, it is possible to construct a useful cause-and-effect diagram on a logical basis. Having completed the cause-and-effect diagram, the next step is to assign an importance to each factor. All the factors in the diagram are not necessarily closely related to the characteristic. Mark those factors which seem to have a particularly significant effect on the characteristics. Finally, include any necessary information in the diagram, such as the title, the name of the product, process or group, a list of participants, the date, etc. Procedure for making cause-and-effect diagrams for systematically listing causes Step 1 a) Procedure Decide on quality characteristic. Step Find as many cause as possible which are considered to affect the quality characteristic. Step 3 Sort out the relations among the causes and make a cause-and-effect diagram by connecting those elements with the quality characteristic by cause-and-effect relations. Step 4 Assign an importance to each factor, and mark the particularly important factors which seem to have a significant effect on the quality characteristic. Step 5 Write any necessary information. b) Explanation of the procedure This approach is characterized by linking two different activities: picking up as many causes as possible and arranging them systematically. 130

131 For picking up causes, open and active discussion is required, and an effective method of conducting a meeting held for this purpose is brainstorming. In making the cause-and-effect diagram, the causes should be arranged systematically by proceeding from the small bones to the medium-sized bones, and then from the medium-sized bones to the big bones. VII.7.b. Notes on cause-and-effect diagrams. Hints on making cause-and-effect diagrams Identify all the relevant factors through examination and discussion by many people. The factors most strongly influencing the characteristic must be determined among those listed in the diagram. If a factor is left out in the initial discussion stage before the diagram is constructed, it will not appear at a later stage. Consequently, discussion by all the people concerned is indispensable to the preparation of a complete diagram which has no omissions. Express the characteristic as concretely as possible. Characteristic expressed in an abstract term will only result in a cause-and-effect diagram based on generalities. Although such a diagram will contain no basic mistakes from the point of view of cause-and-effect relations, it will not be very useful for solving actual problems. Make the same number of cause-and-effect diagrams as that of characteristic. Errors in the weight and the length of the same product will have different cause-and-effect structures, and these should be analysed in two separate diagrams. Trying to include everything in one diagram will result in a diagram which is unmanageably large and complicated, making problem-solving very difficult. Choose a measurable characteristic and factors. After completing a cause-and-effect diagram, it is necessary to grasp the strength of the cause-and-effect relation obectively using data. For this purpose, both the characteristic and the casual factors should be measurable. When it is impossible to measure them, you should try to make them measurable, or find substitute characteristics. Discover factors amenable to action. If the cause you have identified cannot be acted upon, the problem will not be solved. If improvements are to effected, the causes should be broken down to the level at which they can be acted upon, otherwise identifying them will become a meaningless exercise. Hints on using cause-and-effect diagrams Assign importance to each factor obectively on the basis of data. Examination of factors on the basis of your own skill and experience is important, but it is dangerous to give importance to them through subective perceptions or impressions alone. Most of the problems which can be solved by such an approach might have already been solved, and consequently, most of the problems remaining unsolved cannot be solved by this approach. Assigning importance to factors obectively using data is both more scientific and more logical. 131

132 Try to improve the cause-and-effect diagram continually while using it. Actually using a cause-and-effect diagram will help you see those parts which should be added. You should make repeated efforts to improve your diagram, and eventually a really useful diagram will be obtained. This will be useful in solving problems, and at the same time, will help improve your own skill and to increase your technological knowledge. Various methods should be applied in combination in solving problems, and the combination of a Pareto diagram and cause-and-effect diagram is particularly useful. VII.8. Distributions and histograms VII.8.a. Variation and distribution If we could collect data from a process in which all factors (man, machine, material, method, etc.) were perfectly constant, all the data would have the same values. In reality, however, it is impossible to keep all factors in a constant state all the time. Strictly speaking, even some factors which, we assume, are in a constant state cannot be perfectly constant. It is inevitable for the values in a given set of data to have a variation. The values of the data are not the same all the time, but this does not mean that they are determined in a disorderly fashion. Although the values change every time, they are governed by a certain rule, and this situation is referred to as data following a certain distribution. VII.8.b. Populations and distributions In quality control, we try to discover facts by collecting data and then take the necessary action based on those facts. The data is not collected as an end itself, but as a means of finding out the facts behind the data. For example, consider a sampling inspection. We take a sample from a lot, carry out measurements on it, and then decide whether we should accept the whole or not. Here our concern is not the sample itself, but the quality of the whole lot. As another example, consider the control of a manufacturing process using an x R determine the characteristics of the sample taken for drawing the find out what the state the process is in. control chart. Our purpose is not to x R control chart, but to The totality of items under consideration is called the population. In the first example above, the population is the lot, and in the second it is the process. Some people may feel it difficult to regard a process as a population because while a lot is indeed a group of finite individual obects, a process itself is not a product at all, but is made up of the 5M s (man, machine, material, method and measurement). When we turn our attention to product-making function, we will recognize that the process produces unmistakably a group of products. Moreover, the number of products is infinite unless the process stops producing them, and for this reason, a process is considered to be an infinite population. 13

133 One or more items taken from a population intended to provide information on the population is called a sample. Since a sample is used for estimating the characteristics of the entire population, it should be chosen in such a way as to reflect the characteristics of the population. A commonly-used sampling method is to choose any member of the population with equal probability. This method is called random sampling, and a sample taken by random sampling is called a random sample. We obtain data by measuring the characteristics of a sample. Using this data, we draw an inference about the population, and then take some remedial action. However, the measured value of a sample will vary according to the sample taken, making it difficult to decide what action is necessary. Statistical analysis will tell us how to interpret such data. VII.8.c. Histograms The data obtained from a sample serves as a basis for a decision on the population. The larger the sample size is, the more information we get about the population. But an increase of sample size also means an increase in the amount of data, and it becomes difficult to understand the population from these data even when they are arranged into tables. In such a case, we need a method which will enable us to understand the population at a glance. A histogram answers our needs. By organizing many data into a histogram, we can understand the population in an obective manner. How to make histograms Step 1: Calculate the range (R) Obtain the largest and the smallest values and calculate R. R= (the largest observed value) (the smallest observed value) Step : Determine the class interval The class interval is determined so that the range, which includes the maximum and the minimum of values, is divided into intervals of equal breadth. To obtain the interval breadth, divide R by 1, or 5 (or 10, 0, 50; 0.1, 0., 0.5, etc.) so as to obtain from 5 to 0 class intervals of equal breadth. When there are two possibilities, use the narrower interval if the number of measured values is 100 or over and the wider interval, if there are less then 100 observed values. Step 3: Prepare the frequency table form Prepare a form, on which the class, mid-point, frequency marks, frequency etc. can be recorded. Step 4: Determine the class boundaries Determine the boundaries of the intervals so that they include the smallest and the largest of values, and wrote these down on the frequency table. Step 5: Calculate the mid-point of class Step 6: Obtain the frequencies 133

134 How to read histograms VII.8.d. Types of histograms It is possible to obtain useful information about the state of a population by looking at the shape of the histogram. The followings are typical shapes, and we can use them as clues for analysing the process. General type (symmetrical or bell-shaped) The mean value of the histogram is in the middle of the range of data. The frequency is the highest in the middle and becomes gradually lower towards the end. The shape is symmetrical. This is the shape which occurs most often. a) General type Comb type (multi-modal type) 4. Figure Every other class has a lower frequency. This shape occurs when the number of units of data included in the class varies from class to class or when there is a particular tendency in the way the data is rounded off. b) Comb type 5. Figure 134

135 Positively skew type The mean value of the histogram is located to the left of the centre of the range. The frequency decreases somewhat abruptly towards the left, but gently towards the right. This shape occurs when the lower limit is controlled either theoretically or by a specification value or when values lower than a certain value do not occur. Left-hand precipice type c) Positively skew type 6. Figure The mean value of the histogram is located far to the left of the centre of the range. The frequency decreases abruptly on the left, and gently toward the right. This is a shape which frequently occurs when a 100% screening has been done because of low process capability, and also when positive skewness becomes even more extreme. d) Left-hand precipice type 7. Figure 135

136 Plateau type The frequency in each class forms a plateau because the classes have more or less the same frequency except for those at the ends. This shape occurs with a mixture of several distributions having different mean values. Twin-peak type (bimodal type) e) Plateau type 8. Figure The frequency is low near the middle of the range of data, and there is a peak on either side. This shape occurs when to distributions with widely different mean values are mixed. Isolated peak-type f) Twin-peak type 9. Figure There is a small isolated peak in addition to a general-type histogram. This is a shape which appears when there is a small inclusion of data from a different distribution, such as in the case of process abnormality, measurement error, or inclusion of data from a different process. 136

137 g) Isolated-peak type 30. Figure VII.8.e. Comparing histograms with specification limits If there is a specification, draw lines of the specification limits on the histogram to compare the distribution with the specification. Then see if the histogram is located well within the limits. The following figures represent five typical cases. When the histogram satisfies the specification: a) Maintenance of the present state is all that is needed, since the histogram amply satisfies the specification. b) The specification is satisfied, but there is no extra margin. Therefore, it is better to reduce the variation by a small degree. LSL USL a) 31. Figure 137

138 LSL USL 3. Figure When the histogram does not satisfy the specification: b) c) It is necessary to take measures to bring the mean closer to the middle of the specification. d) This requires action to reduce the variation. e) The measures described in c) and d) are required. LSL USL c) 33. Figure 138

139 LSL USL d) 34. Figure LSL USL e) 35. Figure 139

140 VIII. Application of Statistical Process Control Before going into SPC specific details there has to be a common agreement / decision whether SPC is applicable in a particular practical case or not. There is no one clear definition available supporting this decision but there are several points, which have to be considered before the decision of process control is taken. Is the actual process critical (based on: customer satisfaction and/or QP, and/or cost consequence)? Can the process be made mistake proof? If yes, then do so, and do not use SPC. Is the expected result significant? Do we have the appropriate resource - like reliable measurement system, trained personnel, enough time available? If any of these is not available then is there enough money and time to close the gap? The evaluation of the points above requires a multidisciplinary team including quality, engineering, production and CFT leader. If the team decides to use SPC then SPCS has to start and lead the process according to instructions described in the following chapters. VIII.1. What is SPC? All processes have natural variability (due to common causes) and unnatural variability (due to special causes). We use SPC to monitor and/or improve the process. Attentive use of SPC can allow us to detect special cause variation through out of control signals. Control charts cannot tell you why the process is out of control, only that it is. Control charts are the means through which process and product parameters are tracked statistically over time. Control charts incorporate upper and lower control limits that reflect the natural limits of (random) variability in the process. These limits should NOT be compared to customer specification limits. Based on statistical principles, control charts allow for the identification of unnatural (nonrandom) patterns in process variables. When the control chart signals a nonrandom pattern, we know that special cause variation has changed the process. The actions we take to attack non-random patterns in control charts are the key to successful SPC usage. VIII.. Selecting the Appropriate SPC Method This decision tree can be used as a guideline to select the most appropriate SPC method. 140

141 Start Select the process step, for which you want to select an appropriate SPC method. No Is the process characteristic a continuous variable? No Do you want to control an Input? Yes Yes Is the StDev easily computable? Yes Use Xbar and S chart. No No Do you want to weight the historical data? Yes Use Xbar and R chart. Use I-MR chart. Use EWMA chart. Can you measure more than one defect per unit? Yes Is the sample size constant? No No Yes Use U chart. No Is the sample size constant? Yes Use C chart. Use p chart. Use np chart. 36. Figure 141

142 VIII.3. VIII.3.a. Overview of Control Charts Control Charts Used to Monitor Output Variables Charts for Continuous Output (Y) Xbar & R charts, or Xbar & S: An Xbar chart measures the central tendency of Y over time. R charts measure the gain or loss of uniformity within sub-groups, which represents the variability in Y over time. R charts are based on the range of values within each sub-group. Sigma charts track the variability based on the standard deviation within sub-group, not the range. Charts for Attribute Output (Y) np charts: A simple chart used to track the number of non-conforming units (defective parts). Use when the sample size is constant. P charts: A simple chart used to track the percentage of non-conforming units (percentage of defective parts). Use when the sample size is NOT constant. C charts: A simple chart used to track the number of defects produced (not the number of defectives). Use when the sample size is constant. U charts: A simple chart used to track the number of defects per unit produced (not the % defectives). Use when the sample size is NOT constant. VIII.3.b. Control Charts Used to Monitor Input Variables Charts for Continuous Input (X) I-MR chart: Also known as an individuals and moving range chart. This is similar to the Xbar & R chart. Instead of charting the sub-group average and range over time, this chart plots each individual reading (sub-group size = 1) and a moving range. EWMA chart: (Exponentially Weighted Moving Average). Used when a process is known to be under statistical control. The user of this chart is looking to detect sustained shifts in the process mean. Exponentially weighs past data with respect to current data. It is a complicated chart and should be used only with automation. VIII.4. VIII.4.a. Defining Control Charts Charts for Continuous Output (Y) Xbar & R Chart An Xbar chart measures the central tendency of the Y over time. R charts measure the gain or loss of uniformity within sub-group, which represents the random cause variability in the Y over time. R charts are based on the range of values within each sub-group. Sigma charts track the variability based on the standard deviation within sub-group, not the range. 14

143 These types of charts are the most sensitive (powerful) charts for tracking process excursions in the mean and the variation over time. An assumption of normally distributed individuals exists. Sub-group means will tend to produce normal distribution because of the central limit theorem. Three sigma limits used are based on the sample size. The minimum number of samples recommended to establish control limits is 30. Gather the Data i. Select the subgroup size ( n ). Typical subgroup sizes are 4 to 5. The concept of rational sub-grouping should be considered. The obective is to minimize the amount of variation within a subgroup. This helps us "see" the variation in the averages chart easier. ii. Select the frequency, with which the data will be collected. Data should be collected in the order in which it is generated (in most cases). iii. Select the number of subgroups ( k ) to be collected before control limits are calculated. You can start with initial control limits after ten subgroups, but recalculate the limits each time until you get to twenty subgroups. iv. For each subgroup, record the individual, independent sample results. v. For each subgroup, calculate the subgroup average: vi. X n i 1 n X i, where n is the subgroup size, x i ' s are the observations in the subgroup. For each subgroup, calculate the subgroup range:, where is the maximum individual sample result in the R X max X min subgroup and X min X max is the minimum individual sample result in the subgroup. Plot the Data i. Select the scales for the x and y axes for both the X and R charts. ii. Plot the subgroup ranges on the R chart and connect consecutive points with a straight line. iii. Plot the subgroup averages on the X chart and connect consecutive points with a straight line. Calculate the Overall Process Averages and Control Limits i. Calculate the average range ( R ): R k i 1 k R i, where k is the number of subgroups, and subgroup. ii. Plot R on the range chart as a solid line and label. R i is the range of i th iii. Calculate the overall process average ( X ): 143

144 k X i i X 1 where k subgroup. k is the number of subgroups, and X i is the range of iv. Plot X on the Xbar chart as a solid line and label. v. Calculate the control limits for the R chart. The upper control limit is given by vi. vii. viii. The lower control limit is given by UCL R D 4 R, LCL R D 3 R LCL R, where. D 3, D 4 i UCL R th are control chart constants that depend on subgroup size (see the table below). Plot the control limits on the R chart as dashed lines and label. Calculate the control limits for the Xbar chart. The upper control limit is given by UCL X. The lower control limit is given by UCL X X A R, LCL X X A R LCL X., where depends on subgroup size (see the table below). Plot the control limits on the Xbar chart as dashed lines and label. A is a control chart constant that Interpret Both Charts for Statistical Control i. Always consider variation first. If the R chart is out of control, the control limits on the Xbar chart are not valid since you do not have a good estimate of. All tests for statistical control apply to the Xbar chart. Points beyond the limits, number of runs and length of runs tests apply to the R chart. Calculate the Process Standard Deviation, If Appropriate i. If the R chart is in statistical control, the process standard deviation, calculated as: R S, where d the table below). d S., can be is a control chart constant that depends on subgroup size (see To calculate control limits and to estimate the process standard deviation, you must use the control chart constants D 4, D 3, A, and d. These control chart constants depend on the subgroup size ( n ). These control chart constants are summarized in the table below. For example, if your subgroup is 4, then D 4 =.8, A = 0.79, and d =.059. There is no value for D 3. This simply means that the R chart has no lower control limit when the subgroup size is 4. Subgroup Size (n) XVI. Table A D 3 D 4 d

145 Xbar & S Chart The practical use of Xbar and S chart is difficult because computing of standard deviation is much more time consuming than computing the range. Therefore it is recommended to use the Xbar & R chart instead. VIII.4.b. Charts for Continuous Input (X) I-MR Chart Individual Moving range charts should be used for true X's like temperature, humidity, concentration of a solution, or gas flow per minute. Typical applications are: The production volumes are low, where only one piece is produced per day or sometimes longer. High-cost testing when the cost of testing is very expensive, time consuming, destructive, or otherwise prohibitive. Chemical processing, where only one measurement per batch is taken. Defining the Initial Settings i. Select the subgroup size ( n ). This is the length of moving range. ii. Select the frequency with which the data will be collected. iii. Select the number of subgroups ( k ) to be collected before the chart center line is calculated. iv. Collect the data of first k subgroups. (altogether nk individuals): v. Calculate the center line of Individuals as X X,..., 1, X nk vi. X nk i 1 X nk Calculate the center line for Moving Range as i k Ri i R 1, where R 1, R,..., Rk are the ranges of first k k subgroups. 145

146 Calculate the Control Limits i. The upper ( ) and lower ( D 3 n ii. The upper (, D 4, D 3 E D4, E UCL X UCL X X E R, LCL X X E R UCL MR ) and lower ( UCL MR D 4 R LCL MR D 3 R, LCL X. ) limits for individuals are calculated as: LCL MR ) limits for moving range are calculated as: are constants that vary according to subgroup size, n. XVII. Table * * * * * * There is no lower control limit for ranges for subgroup sizes below 7 EWMA Chart Unfortunately, it takes time for the patterns in the data to emerge because individual violations of the control limits do not necessarily point to a permanent shift in the process. The EWMA (exponentially weighted moving average) control chart is suited for detecting small shifts and trends. The Exponentially Weighted Moving Average (EWMA) is a statistic for monitoring the process that averages the data in a way that gives less and less weight to data as they are further removed in time from the current measurement. i. Defining the Initial Settings ii. Select the subgroup size ( n ). Typical subgroup sizes are 4 to 5. iii. Select the frequency with which the data will be collected. iv. Select the number of subgroups ( k ) to be collected before the chart center line is calculated. v. Collect the data of first k subgroups. vi. For each subgroup, calculate the subgroup average: vii. X n i t 1 i n X i, where n is the subgroup size, X i ' s the th subgroup. ( X t is the average of the t th subgroup.) Calculate the center line as are the observations in The values of D 3 and D4 in this table are same as the ones shown for Xbar and R chart. 146

147 k X i i X 1 k, where k is the number of subgroups, X i is the average of i th subgroup. Calculate the Control Limits and EWMA Data i. The upper (UCL) and lower ( LCL ) limits are computed based on historical data: ii. Initialize UCL X sk LCL X sk s, where is the standard deviation of the historical data; the function under the radical is a good approximation to the component of the standard deviation of the EWMA statistic that is a function of time; and K is the multiplicative factor, which is usually taken to be 3. : EWMA 0 EWMA 0 iii. Specify the weighting factor iv. X When =1, the chart is equivalent to the X-bar or X-individual chart (depending on the subgroup size). Higher produce smoother chart, which is less sensitive to sudden changes. The point of the chart (EWMA statistic) at time t is computed recursively: EWMA X t EWMA t1` is the t t 1 1 EWMA t, where -1th value of EWMA. X t is the average of the t values th subgroup, and The chart below is an example that demonstrates how EWMA is making the trend visible. 37. Figure 147

148 VIII.4.c. Charts for Attribute Output (Y) P Chart P chart is a simple chart used to track the percentage of non-conforming units (percentage of defective parts). Use when the sample size is NOT constant. The definition of conformance and nonconformance must exist in advance. Each component, part or item being checked is recorded as either conforming or nonconforming even if an item has several specific nonconformities, it is only tallied once as nonconforming item. Gather the Data i. Select the subgroup size ( n ). For the chart to show analyzable patterns, the subgroup size should be large enough to have several nonconforming items per subgroup (e.g., ii. n p 5 ). Note, however, that large subgroup sizes can be disadvantage if each subgroup represents a long period of process operation. Let p p p = average proportion nonconforming = n 1 p 1, n p,, n k p k and n 1,...,n k n1 p1 n p... nk p n n... n 1 k k, where are the number of nonconforming items and number of items inspected in each subgroup. denote the probability that an item has the attribute being counted. The value must be the same for all n items in any one sample. iii. Subgroup frequency should make sense in terms of production periods to aid analysis and correction of problems found. iv. Select the number of subgroups ( k ) to be collected before control limits are calculated. The data collection period should be long enough to capture all likely sources of variation affecting the process. Generally, it should also include 5 or more subgroups to give a good test for stability and, if stable, a reliable estimate of process performance. Plot the Data i. Select the scales for the x and y axes in the way the x axis refers to the subgroup identification (hour, day, etc.), the y axis refers to the proportion (or percent) nonconforming. The vertical scale should extend from zero to about 1-1/ to times the highest proportion nonconforming noted in the initial data readings. oi ii. pi, where o i is the number of nonconforming items, and n i is the total number n i of items in the be identical, so i th subgroup. 1 i k For practical reasons, n n... n 1 k n, however this is not a requirement. n i s are usually set to iii. Plot the values of p i for each subgroup. It is usually helpful to connect the points with lines to help visualize patterns and trends. 148

149 Calculate Control Limits 3 p(1 p) UCL p p n where n 3 p(1 p) LCL p p n is the constant sample size. Note: When p is low and/or n is small, the LCL p can sometimes be calculated as a negative number. In these cases there is no lower control limit, since even a value of particular period is within the limits of random variation. p 0 for a Plot process average ( dashed horizontal lines. p ) as a solid horizontal line and control limits ( UCL p, LCL p ) as Note: The control limit calculations given above are appropriate when the subgroup sizes are all equal (as they would be in a controlled sampling situation). Theoretically, whenever the sample size changes (even for a single subgroup), the control limits change, and unique limits would be calculated for each subgroup having a unique sample size. However, for practical purposes, control limits calculated with an average sample size ( n ) are acceptable when the individual subgroup size varies from the average by no more than +/- 5%. When subgroup sizes are vary by more than this amount, separate control limits are required in the way the formula has to contain n i instead of n, where If possible select constant sample size. Interpret the Chart for Statistical Control n i 149 is the sample size of the particular subgroup. i. Any point beyond the control limits is evidence of instability at that point so we presume a special cause behind. The special cause can be either unfavorable or favorable but the point has to be marked anyway. A point above the upper limit (higher proportion nonconforming) is generally sign of one or more of the following: The control limit or plot point are in error; The process performance has worsened, either at that point in time or as part of a trend; The evaluation system has changed (e.g. inspector, gage); A point bellow the lower limit (lower proportion nonconforming) is generally sign of one or more of the following:

150 The control limit or plot point are in error; The process performance has improved; The measurement system has changed; Patterns or trends within the control limits: The presence of unusual patterns or trends, even when all points are within the control limits, can be evidence of non-control or change in the level of performance during the period of the pattern or trend. This can give advance warning of conditions, which if left uncorrected, could cause points beyond the control limits. Note: When the average number of nonconforming items per subgroup ( large (9 or more), the distribution of the subgroup ii. Runs: In process under control, with p n p s is nearly normal. n p ) is moderately moderately large, approximately equal numbers of points should fall on either side of the average. Either of the following could be a sign that a process shift or trend has begun: 7 points in a raw on one side of the average 7 points in a raw that are consistently increasing or consistently decreasing Runs above the process average or runs up, generally signify one or both of the following: The process performance has worsened and may still be worsening The evaluation system has changed Runs below the process average or runs down, generally signify one or both of the following: iii. The process performance has improved The measurement system has changed Obvious nonrandom patterns: Other distinct patterns, such as trends, cycles, unusual spread of points within the control limits and relations among values within subgroups, may indicate the presence of special causes of variation, although care must be taken not to over-interpret the data. One test for unusual spread is the following: Distance of points from the process average: In case a process under statistical control, with only common-cause variation present and n p is moderately large, about /3 of the data points will be within the middle third of the region between the control limits, about 1/3 of the points will be in the outer two-thirds of the region and about 1/0 will lie relatively close to the control limits. If substantially more than /3 of the points lie close to the process average this could mean one or more of the following: The control limit or plot point are in error; 150

151 The process or the sampling method are stratified, each subgroup systematically contains measurements from two or more process streams that have very different average performance (e.g. mixed output of two parallel production lines) The data have been edited; If substantially fewer than /3 of the points lie close to the process average this could mean one or both of the following: The calculation or plotting are in error; The process or the sampling method cause successive subgroups to contain measurements from two or more process streams that have very different average performance (e.g. performance differences between shifts). If several process streams are present, they should be identified and tracked separately. np Chart The np chart is a simple chart used to track the number of non-conforming units (number of defective parts). Use when the sample size is constant. It is identical to the p chart except that the actual number of nonconforming items, rather than their proportion of the sample, is recorded. Both p and np charts are appropriate for the same basic situations, with the choice going to the np chart if (a) the actual number of nonconformities is more meaningful or simpler to report than the proportion, and (b) the sample size remains constant from period to period. The details of instructions for the np chart are virtually identical to those for the p chart; exceptions are noted bellow. Gather the Data i. The inspection sample sizes must be equal. The period of sub-grouping should make sense in terms of production intervals and feedback systems, and samples should be large enough to have several nonconforming items per subgroup. Record the sample size on the form. ii. Record and plot the number nonconforming items in each subgroup (np). Calculate Control Limits i. Calculate the process average number nonconforming ( ii. n p = n p n p... np 1 1 k k, where np 1 n p, np,, nonconforming items in each of the k subgroups. Calculate the upper and lower control limits: UCL np n p 3 n p(1 p) LCL np n p 3 n p(1 p) All other details are identical to those for the p chart. ): np k, where n is the subgroup sample size. are the number of 151

152 c Chart A c chart is a simple chart used to track the number of defects produced (not the number of defective). Use when the sample size or amount of the material inspected is constant. It is applied in two maor types of inspection situations: i. Where the nonconformities are scattered through a continuous flow of product, and where the average rate of nonconformities can be expressed (e.g. misplacements per 100 boards). ii. Where the nonconformities from many different potential sources may be found in a single inspection unit (e.g. failed boards). The details of instructions for the c chart are virtually identical to those for the p chart; exceptions are noted bellow. Gather the Data i. The inspection sample sizes must be equal. So the plotted values of c will reflect changes in quality performance. Record the sample size on the form. ii. Record and plot the number nonconforming in each subgroup (c). Calculate Control Limits i. Calculate the process average number nonconformities ( c ), which is the process capability: ii. c = c1 c... ck k, where c 1, c,, each of the k subgroups. Calculate the upper and lower control limits: UCL c c 3 LCL c c 3 c c All other details are identical to those for the p chart. c k are the number of nonconformities in u Chart A u chart is a simple chart used to track the number of defects per unit produced (not the % defective). Use when the sample size is NOT constant. It is similar to c chart except that the number of defects is expressed on a per unit basis. U chart must be used if the sample size can vary from period to period. The details of instructions for the u chart are virtually identical to those for the p chart; exceptions are noted bellow: Gather the Data i. Sample sizes do not need to be constant from subgroup to subgroup, although maintaining them within 5% above or bellow the average simplifies the calculation of control limits. 15

153 ii. Record and plot the nonconformities per unit in each subgroup ( u ). c u n where c is the number of nonconformities found, and n is the sample size (number of inspection reporting units) of the subgroup; c and n should also be recorded on the form. Calculate Control Limits i. Calculate the process average nonconformities per unit ( u ): ii. u = c1 c n n 1... c... n k k, where c 1, c, c k and n 1, n,, nonconformities and sample size of each of the k subgroups. Calculate the upper and lower control limits: UCL u u 3 LCL u u 3 u n u n, where n is the average sample size. n k are the number of Note: If any individual subgroup size is more than 5% above or bellow the average sample size, the control limits must be recalculated, but use of variable control limits cumbersome and potentially confusing so it is much better using constant subgroup sample sizes wherever possible. All other details are identical to those for the p chart. VIII.4.d. Out of Control Action Plans If SPC is used, one of the most important things is to properly act in cases when the SPC tool gives a warning signal. To have a consequent and structured way of acting Out of Control Action Plans (OOCAP) are applied. We do not discuss these tools in details here. We ust would like to emphasize their importance. SPC and Process Improvement There are (at least) two applications of SPC charts in process improvement: These tools can be used as data sources to identify where the problems are. SPC charts are powerful tools for controlling processes after improvements are done. These are the cases when SPC tools are used to sustain the gains of process improvements. 153

154 IX. Capability Assessment A capability assessment: Allows us to quantify the nature of the problem. Allows the organization to predict its true quality levels for all goods and services. Capability can be assessed for either variable or attribute data collected either in the shortterm or in long-term. Variable data represent a quantitative process characteristic, while attribute data are usually used to characterize the goodness of a process by counting the number of defects of defective parts. The next flowchart shows how to conduct the capability assessment studies. Start No Is the process characteristic quantitative? Use the CADT to decide. Yes Take a sample. Check the specification limits. No Do the data represent a short term? Yes Take a sample on the process variable. Compute the PPM rate. Compute the PPM rate. No Do the data represent a short term? Yes Compute the long term z-score. Compute the short term z-score. Calculate the Pp, Ppk indices. Calculate the Cp, Cpk indices. Estimate the Ppk index. Estimate the Cpk index. Estimate the PPM rate. Estimate the PPM rate. Store the results in the corresponding line of CA result form. (Use the ID for appropriate identification.) 38. Figure 154

155 IX.1. Short-term data vs. long-term data The processes usually show different performance on short and long-term. This is simply due to the fact that there are less assignable causes or there are no assignable causes at all, which might be impacting the processes over a short term. However, the likelihood of experiencing assignable causes is getting higher as the process is observed over a longer term. Here we summarize the characteristics of short and long-term data samples. IX.1.a. Short-term Data We may assume that a data set represents a short-term sample, if this data set is a subset of the population that is taken over a short enough time to eliminate the special causes of variation. Its main properties are: Free of assignable causes, Represents the effect of random causes only, Collected across a narrow inference space, Across one shift of production, Using only one machine, With one operator, Using raw components from only one lot of raw material, etc IX.1.b. Long-term Data A data set can be considered as a long-term one, if it is presumable that the process is not free of assignable effects. Main properties of long-term samples are: Reflects the influence of random causes as well as assignable phenomena, Taken across a broad inference space, Across many shifts of production, Using many machines, With many operators, Using many lots of raw material, etc. IX.. Capability Assessment for Quantitative Process Characteristics We conduct a capability study on a quantitative process characteristic (variable) through the next steps. Setp1: Check the specification given for the process variable. Step: Take a sample on the process variable. Step3: Decide if the sample represents either a short-term or a long-term production. Step4: Calculate the C p, C pk, Sigma Level indices to express the short-term capability, and compute the P p, P pk indices to evaluate the long-term capability. Step5: Estimate the PPM rate. 155

156 IX..a. The C p, C pk and Sigma Level Indices These indices are used to express the short-term capability of a process. Let be a key characteristic of a process, and let LSL and USL be the lower and upper specification limits of respectively. Let s take a short- term sample on and compute the empirical mean and the corrected empirical standard deviation from the sample. S shortterm X S short term X 1 n n i1 i, 1 n i X n 1 i1 where 1,..., n are the elements of the sample, and n is the sample size. The C p index C p is the capability potential of the process assuming that the process is free of nonrandom variability, that is there are no assignable effects. C p is estimated by the next equation C p USL LSL S 6* short term Since C p is independent of X the C p central the process is. That is why we use the index does not provide any information about how C pk expected value is from the center of the specification interval. indicator, which expresses how far the The C pk index C pk( USL) 3* USL X S shortterm C pk LSL 3* X LSL S short term The C pk index is calculated as pk C min C, C pk LSL pkusl. C pk It means that the index measures how far the mean is from the nearest specification limit in terms of the triple of short-term standard deviation. Please note that the equation above is C pk an estimation of because its formula contains the estimated mean and estimated standard deviation. 156

157 The Sigma Level The sigma level is calculated in a very similar way as the is that we use Sshort term level equals to the triple of C pk index, the only one difference in the denominator instead of its tripled value. Therefore, the sigma C pk. Sigma Level 3* C pk IX..b. Estimation of PPM Rate When we already have the values of estimated mean S X and estimated standard deviation, we can estimate that fraction of the total population, which is outside the specification. For this estimation, we assume that the process characteristic as a random variable has normal probability distribution with mean and standard deviation. This assumption is a rightful one because most of the physical characteristics as random variables have normal probability distribution. Since these two parameters are unknown the best what we can do is to estimate them with X and S respectively. When we are looking for an estimation of the PPM rate, we need to compute the probability. Since LSL P and LSL The Z-transformation (Standardization) P LSL or USL are mutually exclusive events LSL or USL P LSL P USL. If is a random variable with normal probability distribution and mean deviation, then the and standard z random variable has normal probability distribution with mean of zero and standard deviation of one. This probability distribution is called standard normal and its values are available from statistical tables or can be calculated by MS Excel. Now, we can standardize LSL and USL, and than based on those, we can compute the probability we are looking for. z LSL LSL X S z USL USL X S P LSL or USL P LSL P USL P z z LSL P z z z z 1 Pz z z z 1 P, LSL where is the probability distribution function of standard normal probability distribution. USL LSL USL USL 157

158 Since P LSL or USL P LSL P USL represents the likelihood that is outside the specification, if we multiply this by million, we get the estimation of PPM rate. PPM 6 z LSL z 1*10 Please note that depending on if X and are calculated from short-term or long-term samples, the appropriate PPM rate can be calculated by using the previous expression. S USL IX..c. The P p, P pk Indices These two indices are used to express the long-term capability of a process that is usually referred to as process performance. Thus, these two indices are called performance potential indices. To calculate P p and long-term standard deviation sample. P, we need to have estimations of the long-term mean pk S longterm X and, which can be computed based on a long-term P p USL LSL S 6* short term P pk( USL) 3* USL X S pk longterm pk P pk LSL P min P, P 3* LSL pkusl X LSL S longterm IX..d. Capability Assessment for Qualitative Process Characteristics The most common qualitative characteristics have the values of good/not good, so they qualify the product. If we have such a qualitative characteristic, then we can calculate the PPM rate for the whole population, if 100% inspection is applied. Now, we discuss how to estimate the P and sigma level indices when the PPM rate is given. We do that C pk, pk through the next steps. The z-score Step1: Decide if the sample represents either a short-term or a long-term production. number of defective products Step: Compute the PPM rate as PPM. total number of productsproduced Step3: Compute the so-called z-scores (short-term and long-term z-scores). Step4: Estimate the, and Sigma Level indices. C pk P pk When we know the PPM rate, we can derive the probability to have bad products. The z-score is calculated as PPM p

159 z 1 1 This z-score is either the short-term or the long-term one, depending on what type of sample we used. Based on empirical data the following equations show the relation between the short-term and long-term z-scores. p If we have short-term data, then z 11 p zst is the short-term z-score, and the longterm z LT z-score is computed as If we have long-term data, then is the long-term z-score, and the shortterm z ST z-score is computed as z LT z ST 1.5 z 1 1 p z ST z LT. LT 1.5 z. XVIII. Table z-score we want to know Short-term Long-term Type of sample Short-term OK -1.5 Long-term +1.5 OK Estimation of Sigma Level z ST is an estimator of the short-term sigma level, and sigma level. z LT is and estimator of the long-term Estimation of C pk and P pk C pk z ST 3 P pk z LT 3 IX..e. Acceptance Guideline Here we provide a short acceptance guideline that is based on the industrial experience. Therefore, the particular acceptances may differ from that because customers requirements result different acceptance levels. XIX. Table C pk Sigma Level (short-term) World Class Level >1.33 >4 159

160 Acceptable, depends on criticality and cost Poor capability/performance 1 to to 4 <1 <3 IX..f. Trainings Effective usage of SPC process is possible only in case of all the involved personnel have sufficient knowledge at least about the pertinent process step. Operators are filling in the SPC charts and technicians or engineers are analyzing results must be qualified for SPC on the required level. The SPCS is responsible to organize the necessary trainings. The training matrix has to refer to the SPC training. 160

161 X. Introduction to measurement system analysis mathematical basis As you will see that, we are going to characterize the performance of a measurement system by determining its repeatability and reproducibility. Since one of the most reliable methods to do it is the so-called analysis of variance, we will take a closer look at that at first. X.1. Notation We will use certain notations in this chapter. Most of them are commonly used in many mathematical writings. The sign stands for the end of a unit such as definition, theorem, proof, etc. X.. Analysis of Variance (ANOVA) The analysis of variance is one of the most powerful methods to investigate the significance of impact of different factors. The fundamental idea behind the ANOVA method is simple. It is to measure the difference of means of subgroups determined by different levels of the participating factors. Depending on how many factors are considered in our models, we can differentiate between one-, two- or multiply ways of ANOVA. In our first step, we are going to go through the mathematics of one-way ANOVA. X..a. One-way ANOVA Let f be a factor with numbers. Grouping C k possible levels. Let C by levels of f, we get the groups. Due to the method how we got the C C,..., 1, be a set that contains the C C,..., 1, C k C k x x,..., x 1, sets, which we will refer to as groups, they have the next properties. n i. ii. Ci C i, i, 1,,..., k k i1 C i C These mean that the groups are representing a disoint decomposition of set C. Our basic idea is to characterize the significance of factor f by expressing the difference between the means of groups. Hence, let us use the following definitions. Definition Let I be the index set of all elements of set C group, and let n N, and let I be the index set of the be the number of elements of group C 1,,...,k. C 161

162 Definition x : 1 n I x, that is x is the average of elements in group C. Definition 1 x : n I x, that is number of elements in set C. x is the total average of all elements in C, where n is the total Sometimes, we will also use the 1 x : n k x 1 I expression that is identical with the original definition of x. Now, we defined the average of elements in each group average of all elements in set C x x, 1,,..., k and the total. As you probably might assume, now we have to define those characteristics, which will inform us about the significance of factor Definition f. S T : I x x S T is the quantity that is in the literature commonly referred to as sum of squared totals. This quantity measures the total variation on C, and it is n 1 times greater than the empirical variance of x 1, x,..., xn numbers. Definition S B S B k 1 x : n x is the sum of squares between the averages of elements of groups and the total average of elements of C. The differences between groups are getting more significant as S B is increasing, or with other wording, S B can be considered as that part of the total variation, which we can explain as the impact of factor f. 16

163 163 Definition 1 : k I W x x S S W is the sum of squares within the groups. It represents that natural variation, which cannot be interpreted as the effect of f. S W is also known as sum of square of residuals. The next theorem summarizes the above-mentioned properties of S T, S B and W S. Theorem W B T S S S Proof Using the definitions up to now and applying simple transformations on S T : k I k I I T x x x x x x x x S 1 1 k I x x x x x x x x 1 k I k I k I x x x x x x x x k I W k x x x x S x x n 1 1 k I W B x x x x S S 1 k I I W B x x x x S S 1 k I W B x n x x x S S 1

164 164 W B W B k I I W B S S S S n x n x x x S S 0 1 The biggest practical advantage of this theorem is that it is enough to compute any two of S T, S B, S W, and the third one can be easily determined. On the other hand, it might be still cumbersome to calculate the sum of squares in the above shown way. The next two lemmas help us to simplify the calculation of S T and S B. Lemma nx x S I T Proof Using the definition of x, and applying simple transformations: I I I I I T x x x x x x x x x x S I I I I I nx xnx x nx n x xn x nx x x x I nx x Lemma 1 nx x n S k B Proof Using the definition of S B, x and x, S B can be rewritten according to the next steps: k k B x x x x n x x n S 1 1

165 k 1 k 1 k n x k 1 n x x k 1 n x k k k I n x x n x nx n x x n nx 1 1 k k k 1 n x x x nx n x xn x nx 1 k 1 1 I 1 n x 1 xnx nx n x nx k 1 1 n n I x Mean Square Variances Each of the S T, S B, S W statistics has an associated quantity that is called degree of freedom. The degree of freedom of sample minus one. The degree of freedom of S T is n 1, that is it equals the total number of elements in the S B equals k 1, that is it equals the number of possible levels of factor f minus one, which is less by one than the number of groups. The degree of freedom of S W and degree of freedom of For example, if factor f is defined as the difference between the degree of freedom of S B then the degrees of freedom of 14 1 d f S W Definition., so, it equals n 1 1 k n k has three different levels, and each group contains five elements, S T, S B, S W are. 14 d f S T 15 1, S T 31 d f S B The mean square of variance is defined as the sum of squares divided by the corresponding degree of freedom. So, the mean square variances of,, are the S T S B S W, M ST n 1 T, M S B k 1 B, M W S W n k statistics. Balanced ANOVA It follows the definitions of n that k 1 n n. (Remember that n indicates the number of elements in the th group, and the groups are representing a disoint decomposition of set C 165

166 .) In most of the practical applications data are collected so that each group contains the same number of data. This version of ANOVA is known as balanced ANOVA. Hypothesis Test We would like to decide if the grouping by levels of factor means of groups. Thus, our null hypothesis is f causes significant differences in H 0 : the f have the same mean; factor based disoint decomposition of population results groups which all while the alternative hypothesis is H a : there are at least two groups having different means. The test is based on that the M F M B W S B k 1 S W n k freedom, if the ratio follows x x F 1,,... k, I probability distribution with k 1, n k degrees of residuals have normal probability distribution with mean of zero. Normal distribution of residuals is an assumption of ANOVA. The F-test is robust in the face of violations of this assumption. If there is extreme skewness or extreme kurtosis, then the reliability of the F-test becomes questionable. Procedure of ANOVA Test i. Choose an appropriate significance level. The most commonly used values are. The confidence level of our test equals. ii. iii. iv. 0.1; 0.05; Take a sample, and create the groups according to the levels of factor Calculate the Calculate the v. Determine vi. S T, S B M F M, and B W S W S B k 1 S W n k 1 F k1, nk 1 statistics. test statistic., where probability distribution function with F is the inverse function of F 1 k1, nk k 1, n k degrees of freedom. f. F is usually referred to as critical value, and its particular values are accessible from tables, or can be calculated by using the MS Excel or some statistical programs. Accept, if, otherwise reect. H 0 The ANOVA Table F F The information from a test can be summarized in a table known as an ANOVA table. H 0 166

167 XX. Table Source of variation Sum of Squares Degree of Freedom Mean Square Variance F ratio P value Between groups S B k 1 S B k 1 S B k 1 S W n k 1 k 1, nk S B k 1 SW n k Within groups Total S W S T n k n 1 W ns k S T n 1 There is a so-called where k 1, nk p is the F value associated to the F probability distribution function with ratio. It is calculated as 1 k 1, nk S B k 1 SW n k k 1, n k degrees of freedom. It is known that the decision on the null hypothesis can be based on the Namely, H 0 is accepted, if p, otherwise H 0 must be reected. p value as well. Acceptance of H 0 means that the grouping by different levels of factor f, results groups which have the same mean. That is, f does not contribute significantly to the total variation of the population being investigated. Thus, we can summarize it using the p value so that f is significant, if Example p. Three operators measure the length of five shafts using a calibrated caliper rule. Each operator measures each shaft once. The results are collected in the next table. (Data are in mms.) XXI. Table Operator1 Operator Operator

168 Test the null hypothesis that the means of measured values are the same in case of each operator. To perform an ANOVA test, we need to calculate the notation used up to now, n 15, k 3, n n n S T., S B, and S W statistics. Using the x , x x x n i1 x i S T n x i1 i nx * S B 3 n x 1 nx 5* * * * S W S S, T B S W Now, we calculate the F statistic, and determine the F critical value. F S B k 1 S W n k F can be determined by using the Tables of F Critical Values in the Appendix. Let be n k k 1 (degree of freedom for the numerator), (degree of freedom for the denominator), so the critical value is that is less than the calculated F statistic, thus we must reect the null hypothesis. 168

169 The ANOVA table Source of Sum of Degree of Mean Square F ratio P value variation Squares Freedom Variance Between groups Within groups Total The practical conclusion is that the operator as a factor significantly impacts onto the measured values, so this way of measuring is not proper enough. Exercise You have got three testers, which among other things measure the output voltage of an electrical circuit. You would like to decide whether your testers are identical in terms of the measured averages, or there is a significant difference between them. You take 5 circuits and make measure the output voltage of each circuit by each tester. Your data sheet looks like this. Tester1 Tester Tester Complete the appropriate ANOVA table, and give the practical interpretation of results! X..b. Two-way ANOVA We learnt how to conduct a one-way ANOVA test to characterize the significance of one factor. In many practical applications, we need to consider the oint effect of two or more factors. Now, we are going to study the case, when we consider the common effect of two factors in our ANOVA model. For instance, in the example of previous chapter we investigated the effect of operators on the measured length of shafts. We may assume that the observed values vary not only due to the different operators but the lengths themselves are not exactly the same, they represent a source of variability as well. It means that we face two sources of variability: the part-to-part difference among the parts and the variability caused by 169

170 imprecision of operators. Later on, we will discuss the meaning of imprecision more exactly. The part-to-part variation is something that can be considered as a natural variation, while the variation caused by the measurement system is something undesirable what we wish to avoid as much as possible. The two-way ANOVA method gives us a powerful tool to identify the sources of variation as well as quantifies their strength. In two-way ANOVA we have two factors, let them be k f and g. Let different levels. It means that the oint grouping by levels of different groups. Now, we introduce the necessary notations and definitions. Definition Definition Definition Let C be a set that contains the x x,..., 1, x n numbers and let be the groups generated by grouping the elements of C g. Let I be the index set of all elements of set C, and let of the f g f g C, group, and let 1,,..., k k C, 1. f g n, N f f C have and k 1 and g have g results k 1 k f, g f, g f, g, C,..., C 1 k1k by levels of both of f and I f, g be the index set be the number of elements of group 1 x : n I x, that is total number of elements in set C. x is the total average of all elements of set C, where n is the Notice that x has the same definition as in case of one-way ANOVA. Definition x 1 n f, g: f, g x f, g I f g C,. 1,,..., k k 1, that is x f, g is the average of elements in group You can see that the definition of x f, g is analogous with the definition of x that we defined in one-way ANOVA method. Similar analogies are visible in the next definitions of S T, f g B S, and Definition f g W S,. 170

171 S T x x I S T is the sum of square total that characterizes the total variation of data in set C. Definition S k k 1 f, g f, g n x f, g B 1 x f g S, B is the sum of squares between the averages of elements of groups and the total average of elements of set C. It expresses the impact of grouping by variation. Definition S k k 1 f, g x x f, g W 1 f, g I f and g onto the total f g S, W like in case of one-way ANOVA represents the natural part of variation that cannot be interpreted as the effect of grouping by f and g. There is an important property of the oint grouping by find a factor with levels of both of k 1 k f and 171 f and g. Namely, we can always levels that can generate the same grouping as the oint grouping by g. This observation means that the remains valid for the two-way ANOVA as well. Furthermore, the S k k 1 f, g f, g n x f g B 1, nx computations remain correct as well. S T S S f, g B T S x I f, g W nx equation, and So, we can separate the variation caused by the oint effect of factors from the natural variation of data. We might ask it with good reason, if there is any way to separate the effect of and in terms of the variation with which they separately contribute to the total f g variation. The answer is positive. Now, we introduce those quantities, which characterize the effect of each factor. Definition Let f f f C, C,..., Ck 1 1 be the groups generated by grouping the elements of C by levels of factor f. Let f f f I, I,..., I k1 1 be the index sets of elements of

172 f f f C1, C,..., Ck 1 elements of sets respectively, and let f f f C, C,..., Ck 1 1 sets. f f f n1, n,..., nk 1 be the number of Definition x 1 n f : f 1,,..., k 1 x f I, that is x f is the average of elements in group f C. Definition S 1 x f x f B k 1 n f f S B f is the sum of squares between the averages of elements of groups generated by levels of. This is the same quantity as we used in one-way ANOVA method to characterize the effect of one factor. Definition Let g g g C1, C,..., Ck levels of factor g g g C1, C,..., Ck elements of Definition g be the groups generated by grouping the elements of C by. Let g g g I1, I,..., I k sets respectively, and let g g g C1, C,..., Ck sets. be the index sets of elements of g g g I1, I,..., I k be the number of x g 1 : n g 1,,..., k x g I, that is x g is the average of elements in group g C. Definition S k x g g n x g B 1 17

173 g B S is the sum of squares between the averages of elements of groups generated by levels of g. Interaction of Factors It can be proven that of squares for interaction. f, g f g S B SB SB S f * g where the S f * g quantity is the so-called sum An interaction is the variation among the differences between means for different levels of one factor over different levels of the other factor. It means that the effect of one factor depends on the current level of the other one. For example, suppose that two operators, A and B measure two parts, 1 and with the next accuracy results. XXII. Table Operator / Part 1 A Good Bad B Bad Good Accuracy Operator A Operator B 1 Part 39. Figure Which operator measures more accurately? We cannot decide that because their accuracy depends on the measured part as well. So, these two factors interact to each other. It is even visible better, if we put the results into the next chart. Since there are usually several different levels of each factor, their interaction might cause a considerable variation. We do not prove the this document. Since the to define S * : f g S f, g f g S B SB SB S f * g S f B S and g B f, g f g S f * g B B B. equation, it would be beyond the purpose of S quantities are easily computable, we use this equation S 173

174 Two-way ANOVA model with interaction We discussed that S T S S T S f, g B f g f, g B S B S f * g S W S f, g W and f, g f g S B SB SB S f * g, so our model is. Thus, the two-way ANOVA with interactions separates four sources of the variation, and these sources are characterized by the f g S, W statistics. f S B, g S B, S f * g and Similarly to the one-way ANOVA, each of freedom. We summarize them in the next table. S T, f S B, g S B, S f * g and f g S, W has its degree of XXIII. Table Sum of Squares Degree of Freedom f S B k 1 1 g S B k 1 S f * g k k f g S, W n k 1 k S T n 1 The mean square variances are calculated so as we discussed earlier. There is an F ratio for each of the f B S, g B S, S f * g statistics and its defined as a fraction with numerator of the corresponding mean square variance and denominator of the mean square variance of The p values are computed as in the case of one-way ANOVA as well. The Two-way ANOVA Table with Interaction XXIV. Table f g S, W. Source of Sum of Degree of Mean Square F ratio P value variation Squares Freedom Variance Factor f f S B k 1 1 f M B k S f B 1 1 M M f B f, g W 1 k1, nk M M f B f, g W Factor g g S k 1 B g M B k S g B 1 1 M M g B f, g W 1 k1, nk M M g B f, g W 174

175 Interaction of and g f S f * g k k M f * g k k 1 1 S f * g 1 M M f * g f, g W 1 k1, nk M M f * g f, g W Error f g S, W n k 1 k M f, g W S f, g W 1 n k k Total S T n 1 S T n 1 Note that the above shown calculations of F ratios are used when the factors are not random variables. We will see another possibility to calculate the ratio for factors, which have random values rather than fixed ones. Two-way ANOVA without Interaction When we perform a two-way ANOVA considering the interaction of factors, it can happen that the value of the interaction term indicates that it is not significant. We account the p interaction not significant, if its p F value is greater than 0.5. In such a case we can reduce our model. We do it so that we add the sum of square of interaction to the sum of square of error, and we simply drop the interaction term from the model. It yields the new square of errors that is calculated as remains the same and the reduced model is of f,g * S W using the * f, g f, g S W SW S f * g S T S f,g * S W sum of. Therefore, the sum of square totals * f B is calculated as n k k 1). The new f,g * S W quantity. ( 1 The Two-way ANOVA Table without Interaction XXV. Table S F g B S f,g W ratios and. The degree of freedom p values are calculated Source of variation Sum of Squares Degree of Freedom Mean Square Variance F ratio P value Factor f f S B k 1 1 M f B k S f B 1 1 M M f,g * W f B 1 f M B f, g M W k 1, nk * Factor g g B S 1 k g M B k S g B 1 1 M M f,g * W g B 1 g M B f, g M W k 1, nk * 175

176 Error * f,g S W n k k 1 1 M f, g W * S n k 1 f, g W * k 1 Total S T n 1 S T n 1 176

177 XI. Measurement Systems XI.1. Categories of Measurement Systems A measurement system is used to get information about a product or process characteristic. This characteristic can be either quantitative or qualitative. If a characteristic can be quantified and expressed in numbers, then we call it quantitative. On the other hand, if a characteristic expresses one or more attributes of a product or process, then we call it qualitative characteristic. For example, the length of a shaft is a quantitative characteristic, because it can always be expressed in numbers. At the same time, if we ust need to decide if a product is good or not, then this characteristic is an attribute of the product. The quantitative variables can be furthermore categorized into the categories of discrete and continuous variables. A variable is continuous, if it can have infinite different values from a finite or infinite interval. E.g. the length of shafts is a quantitative variable, which is continuous. A variable is discrete, if it can have finite or countable infinite different values. E.g. the number of failures on a product is a quantitative and discrete variable. Based on the characteristic that measurement systems are to measure, we differentiate between quantitative and qualitative measurement systems. There are two subcategories of quantitative measurement systems depending on if the measured variable is continuous or discrete. The next chart summarizes the categories mentioned above. Measurement Systems Qualitative (attribute) Quantitative (variable) Discrete Continuous 40. Figure XI.. Analysis of Quantitative Measurement Systems Our fundamental purpose is to evaluate the goodness of a measurement system. We use a simple approach to decide if a measurement system is good or not. The less the measurement system distorts the observed values, the better its goodness is. The goodness of a measurement system can be expressed through the categories of accuracy and precision. 177

178 Let be a variable of a process or product. We assume that is a random variable with mean and figures and we get the variance. When we measure, our measurement system distorts the observed Observed and Observed This can be modeled by the next equations. The Observed Observed Product Product values that usually differ from their real values. Measurement System Measurement System Measurement System quantity is commonly referred to as measurement offset. This can be caused by improper linearity, stability and accuracy of the measurement instrument. Measurement System is known as the imprecision of the measurement system. It represents that undesirable variation of observed data, which is caused by the measurement system, and which we want to identify and separate from the total variation in order to find how much the measurement system contributes to the total variation. We consider the repeatability and reproducibility as the main components of imprecision. So, we will characterize the imprecision by expressing the repeatability and reproducibility, which we commonly call gauge R&R. The next chart summarizes the possible sources of observed variation. Observed process or product variation Process or product variation Measurement variation Due to operator Reproducibility Due to measurement device Repeatability Precision (Gauge R&R) Accuracy Stability Linearity Measurement offset (Accuracy) 41. Figure 178

179 XI..a. Measurement Offset (Accuracy) The measurement offset is usually called as accuracy, because all its components are strongly related to the accuracy of the measurement device (gauge). Do not mislead you that we use the accuracy as an individual measurement offset component as well. This is so because the other two components, the linearity and stability can be considered as subcategories of accuracy. Now, we define and explain the components of measurement system offset. Gauge Accuracy Accuracy is the difference applied between the observed average of measurements and the true value. 4. Figure An adequate calibration program can ensure the accuracy of the device is maintained. A gauge calibrated well is always considered as an accurate one. Gauge Stability Stability is the difference in the accuracy obtained on the same parts taken at different times. Gauge Linearity 43. Figure Linearity is the difference in the accuracy of values throughout the expected operating range. 179

180 44. Figure 45. Figure The calculation of accuracy, stability or linearity is not difficult. However, it might be challenging because to obtain the true value is usually not easy. It requires a high accurate gauge, which we can trust in. Normally, it should be a master gauge from a laboratory or a calibrated one, whose accuracy is guaranteed. XI..b. Quantitative Gauge R&R (Precision) As we discussed, we are going to express the precision (or imprecision) of a measurement system by expressing its repeatability and reproducibility. According to the Observed Product Measurement System model, we wish to separate the part of variation, which is caused by the measurement system from the total observed variation. As we discussed earlier, we will look for the sum of variances due to the repeatability and reproducibility: Measuremen t System Repeatabil ity Reproducib ility Measurement System as Thus our model is Observed. Product Repeatabil ity Reproducib ility Now, it is time to define the meanings of repeatability and reproducibility. Repeatability Repeatability of a gauge is a measure of the variation when one operator uses the same device to repeatedly measure the identical characteristic on the same part. Repeatability must also account an automated piece of test equipment. 180

181 46. Figure Reproducibility Reproducibility is the variation in the averages of measurements caused by different operators using the same device when measuring identical characteristics of the same parts. Reproducibility must be accounted for variation between different measuring devices as well. The Gauge R&R Methodology 47. Figure To conduct a gauge R&R study, we need to define a unique identifier for each part we want to measure. We call it part ID. We also need to select the operators (testers) we wish to evaluate. The next procedure defines how to collect the necessary data to evaluate the measurement system. i. Collect a minimum of 10 samples that represent the full range of products. ii. Identify the operators who use the measuring instrument daily, or choose the testers, which you want to evaluate. iii. Calibrate the gauge or verify if the last calibration date is valid. iv. Create a data collection sheet that looks like this Part ID Operator Trial Measurement v. Ask the first operator to measure all the parts once in random order. Blind sampling, in which operator does not know the identity of each part. vi. Have the second operator measure all the samples once in random order, and continue until all the operators have measured the samples once. 181

182 vii. viii. ix. Repeat step v.) and vi.) for the required number of trials. Put the data into the collection sheet. Analyze the results by using an appropriate mathematical method. Mathematical methods for Gauge R&R Studies There are two well-known mathematical methods used commonly to conduct a gauge R&R study. These are the so-called X-bar and range and the ANOVA methods. We will use the ANOVA method. Two-way ANOVA in the Gauge R&R Study Now, we have two factors, the operator and the part ID. These two factors are considered in our two-way ANOVA. The ANOVA table looks like this. XXVI. Table Source of Sum of Degree of Mean Square F ratio P value variation Squares Freedom Variance Part ID (Factor f ) f S B k 1 1 M f B k S f B 1 1 M M f B f * g 1 k1, nk M M f B f * g Operator (Factor g ) S g B k 1 M g B k S g B 1 1 M M g B f * g 1 k1, nk M M g B f * g Operator* Part ID Interaction of and g f S f * g k k M f * g k k 1 1 S f * g 1 M M f * g f, g W 1 k1, nk M M f * g f, g W Repeatability f g S, W n k 1 k M f, g W S f, g W 1 n k k Total S T n 1 S T n 1 If you look at this table attentively, you can see that it differs from the one we have introduced in chapter 1. The difference is in the computation of F ratios for the operator and part ID. This is so because now both factors are random variables, and in this case, the mean square variance of interaction is used in the denominator for calculating the F ratios for random factors. (Remember, if a factor is fixed, then the mean square variance of error is used in the denominator to calculate the F ratio.) Of course, the corresponding p values are recalculated according to the new F ratios as well. 18

183 The ANOVA table immediately informs us whether the operator and the operator*part ID are significant as well as we can see how much the sum of squares of repeatability contributes to the total variation. Variance components The mean square variances are used to calculate the variance components. The formulas depend on whether the value for the operator*part ID interaction is greater than 0.5. If it p is so, then this term is dropped from the model because it is not significant. According to the notation used in the ANOVA table k 1 is the number of parts, number of operators. Let r be the number of replicates, that is the number of trials. k is the Calculation of variance components when the operator*part ID term is included in the model. XXVII. Table Variance component Part ID Operator Definition M M f B g B M k r M k r 1 f * g f * g Comment MS Part ID - MS Operator*Part ID number of operators* number of replicates MS Operator - MS Operator*Part ID number of parts* number of replicates Operator *Part ID M f * g r M f, g w MS Operator*Part ID - MS Repeatability number of replicates Repeatabil ity MS f g M, W Repeatabil ity Reproducib ility Operator Operator *Part ID GaugeR&R Measurement System Repeatabil ity Reproducib ility Total Part ID GaugeR&R Calculation of variance components when the operator*part ID term is not included in the model. XXVIII. Table Variance component Definition Comment 183

184 Part ID M f B M k r f, g W * MS Part ID - MS Repeatability number of operators* number of replicates Operator M g f, g B M k r 1 W * MS Operator - MS Repeatability number of parts* number of replicates Repeatabil ity M W f,g * MS Repeatabil ity Reproducib ility Operator GaugeR&R Measurement System Repeatabil ity Reproducib ility Total Part ID GaugeR&R Measurement System Metrics We introduce the metrics can be used to evaluate the precision of a measurement system. We also provide a guideline for acceptance of a measurement system. That is we provide information on how to decide if a measurement system is acceptable, marginal or unacceptable using different metrics. Total Gauge R&R % Contribution The % contribution is the percentage of total variation that is accounted for by each variance component. It expresses how much the variance components contribute to the total variance. The total gauge R&R % contribution is the % contribution of calculated as: total gauge R&R % contribution Measurement System Total *100. Measurement System to Total, so it is The Study Variation The study variation of each variance component is calculated as times the root square of the variance component. The most common value for X is There is a statistical reason behind this value. Namely, approximately 99% of the measurements are in the interval, which is centered to the observed mean of the measurements and has the width of 5.15 times the standard deviation on both sides of the mean. Specially, if the measurement has normal probability distribution, then 99.97% of the data are in the interval. X 5. Observed 15 Observed We can calculate the so-called % study variation quantity for each variance component as the standard deviation of the component divided by the total standard deviation. 184

185 The Precision to Total Variation Ratio (P/TV) The study variation for the gauge R&R component is known as the precision to total variation ratio (P/TV). It is calculated as P/TV GaugeR&R Total The Precision to Tolerance Ratio (P/T) The precision to tolerance ratio is used to express what percent of the process tolerance is taken up by imprecision of the measurement system. P/T = 5.15 USL LSL GaugeR&R, where USL is the upper and LSL is the lower specification limit. Number of Distinct Categories The number of distinct categories is the number of groups within the process data that the measurement system can discern. Number of distinct categories Total 1 GaugeR& R If the number of distinct categories is fewer than two, the measurement system is unable to distinguish any part from any other one. Acceptance guideline Here we summarize the guiding principles to qualify a measurement system. We use the % contribution, % study variation and the number of distinct categories as metrics. 3 XXIX. Table % Contribution % Study Variation (P/TV%) Distinct Categories No issues with the measurement system Depends on criticality and cost <5% <10% >10 5% to 15% 10% to 30% 4 to 9 Reect the measurement system >15% >30% <4 Notice that the above shown qualification is ust a recommendation. In practice, there must be a regulation, which specifies the acceptance considering the process specific factors as well. 3 The sign stands for the round down function. 185

186 Example (Quantitative Gauge R&R) We have a laser microscope that is used for measuring the height of soldering paste printed by a stencil printer. Two operators use the microscope. We collected a sample having 10 parts, and we asked the first operator to measure the paste height on each part in random order first. After that we asked the second operator to do the same. Finally, we asked them to repeat the measuring once more. Data were collected in the next data sheet. Part ID Operator Trial Measurement (microns)

187 We will perform a gauge R&R study by computing the % contribution, % study variation and number of distinct categories. ANOVA Table Two-Way ANOVA Table With Interaction Source DF SS MS F P Part ID Operator Part ID * Operator Repeatability Total The p value of interaction indicates that the interaction term is not significant, so we reduce the model. Two-Way ANOVA Table Without Interaction Source DF SS MS F P Part ID Operator Repeatability Total Variance components Using the notation we used up to know: k 1 10 (number of parts), k (number of operators), r factor (number of replicates), f is the Part ID, factor g is the Operator. Part ID M f f, g B M k r W * * Operator M g f, g B M k r 1 W * * Repeatability * f, g M W Reproducib ility Operator

188 Gauge R&R Repeatabil ity Reproducib ility Total Part ID GaugeR&R % Contribution Source Variance Component Contribution % Part ID Operator Repeatability Reproducibility Total Gage R&R Total Variation % Contribution Contribution in % Part ID Operator Repeatability Reproducibility Total Gage R&R Source of Variation Total Variation According to the acceptance guideline, the Total Gauge R&R contribution of 44.5% indicates that the measurement system is unsuitable. % Study Variation Source Variance Component Std. Deviation Study Variation (5.15*Std. Dev.) Study Variation % Part ID % Operator % 188

189 Repeatability % Reproducibility % Total Gage R&R % Total Variation % Study Variation in % % 90.00% 80.00% 70.00% 60.00% 50.00% 40.00% 30.00% 0.00% 10.00% 0.00% 74.49% Part ID % Study Variation 31.16% Operator Repeatability 59.00% Reproducibility 31.16% Total Gage R&R Source of Variation 66.7% Total Variation % The P/TV ratio of 66.7% (% study variation for the total gauge R&R) means, that the measurement system is very imprecise. Number of Distinct Categories Number of distinct categories Total GaugeR& R * The number of distinct categories of 1 also shows the poor performance of the measurement system. XI.3. Analysis of Qualitative Measurement Systems (Attribute Gauge Study) A qualitative measurement system is one that compares each part to a standard and accepts the part if this standard is met. Commonly, it is a kind of inspection done by operators or a pass/fail gauge. An attribute measurement system provides pass/fail or other types of categorical data such as good, not good or 1, 0. The effectiveness of screening of an attribute measurement system is the ability to differentiate between good and bad parts. There are two types of errors: good parts are

190 reected or bad parts are accepted. The first error type is called customer bias, while the second one is known as producer bias. An attribute measurement system is as good as much it is free of these two biases. Now, we review the so-called attribute gauge R&R tool that is used to evaluate the precision of qualitative measurement systems. XI.3.a. Purpose of the Attribute Gauge R&R Since the inspectors (or a pass/fail gauge) make decisions on their own it is extremely important to have the same criteria to make right udgments. It means that each of the inspectors has to be well trained, and has to be able to precisely apply the known standards on making decisions. This kind of desired precision also means that they have to have the ability to accurately repeat their inspection decisions. The main purposes of an attribute gauge R&R study are to identify how well inspectors/gauges conform to a known master, and how well they conform to their own and to the others decisions. XI.3.b. The Attribute Gauge R&R Methodology i. Select a minimum of 30 parts to do the study 4 ii. ~ 50% of the parts should have defects, iii. ~ 50% of the parts should be defect free, iv. ~ if possible, select borderline good and bad parts, too. v. Identify the inspectors (who should be qualified and experienced) and the gages. vi. Have each inspector, independently and in random order, assess these parts and determine whether or not they pass or fail. vii. Repeat step iii.) for the required number of replicates. viii. Enter the data into the attribute R&R spreadsheet to report the effectiveness of the measurement system. ix. Evaluate the results and define actions, if necessary. x. Re-run the study to verify the fix. 4 The number of parts is arbitrary. More samples result higher confidence of the study. 190

191 Attribute R&R Spreadsheet for Three Inspectors Attribute Gage R & R Effectiveness Attribute Legend 5 (used in computations) SCORING REPORT DATE: NAME: 1 pass PRODUCT: All operators fail BUSINESS: agree within and All Operators between each agree with Other standard Known Population Operator #1 Operator # Operator #3 Y/N Y/N Sample # Attribute Try #1 Try # Try #1 Try # Try #1 Try # Agree Agree 1 pass pass pass pass pass fail fail N N pass pass pass pass pass fail fail N N 3 fail fail fail fail pass fail fail N N 4 fail fail fail fail fail fail fail Y Y 5 fail fail fail pass fail fail fail N N 6 pass pass pass pass pass pass pass Y Y 7 pass fail fail fail fail fail fail Y N 8 pass pass pass pass pass pass pass Y Y 9 fail pass pass pass pass pass pass Y N 10 fail pass pass fail fail fail fail N N 11 pass pass pass pass pass pass pass Y Y 1 pass pass pass pass pass pass pass Y Y 13 fail fail fail fail fail fail fail Y Y 14 fail fail fail pass fail fail fail N N Statistical Results % Appraiser 1 %Score vs Attribute Source Operator #1 Operator # Operator #3 Operator #1 Operator # Operator #3 Total Inspected # Matched False Negative (operator reected good product) False Positive (operator accepted bad product) 1 1 Mixed % UCL 100.0% 95.3% 100.0% 95.3% 87.% 91.6% Calculated Score 100.0% 78.6% 100.0% 78.6% 64.3% 71.4% 95% LCL 76.8% 49.% 76.8% 49.% 35.1% 41.9% Screen % Effective Score 3 Screen % Effective Score vs Attribute 4 Total Inspected # in Agreement % UCL 8.3% 71.1% Calculated Score 57.1% 4.9% 95% LCL 8.9% 17.7% % Appraiser Score This quantity is calculated for each inspector. It expresses the % of occasions when an inspector agrees with him/herself on both trials, so it indicates the repeatability of the inspector (operator) in percentage. You can also see a confidence interval calculated for each % appraiser score. The limits of confidence intervals are calculated as 191

192 UCL = log 10 n 1 10, if k 0 * 1 k 1, nk 1, if 0 k n 1, if k n LCL = 0, if k 0 * 1 n1 k, k 1, if log10 n 10, if k n 0 k n where k is the number of agreements, n of Beta probability distribution function with, level (we use * 0.05). is the number of parts, 1, parameters, and is the inverse function * is the significance % Score vs. Attribute This metric tells for each inspector the % of occasions when the inspector agrees on both trials with the known standard. Screen % Effective Score This quantity expresses the % of occasions when all inspectors agreed within and between themselves. Screen % Effective Score vs. Attribute This metric is used to express the % of occasions when all inspectors agreed within and between themselves, and agreed with the known standard. There are confidence intervals calculated for the screen % effective score and the screen % effective score vs. attribute indices. These are calculated in the same way as the confidence intervals for the appraiser scores. Interpretation of Results The metrics introduced can tell us how good our attribute measurement system is, as well as they show at the same time where the weaknesses are. Therefore, these metrics can be used to draw the appropriate conclusions. (E.g. we can highlight, if we face a lack of training or experience of inspectors.) 19

193 XII. Appendix XII.1. The F Probability Distribution Let,..., 1, n and,..., 1, probability distribution. Then distribution with m, n m F m, n degrees of freedom. be independent random variables with standard normal m m n n is a random variable that has F XII.1.a. Tables of F critical values 0.1 Degree of freedom for denominator Degrees of freedom for numerator

194 0.05 Degree of freedom for denominator Degrees of freedom for numerator

195 0.01 Degree of freedom for denominator Degrees of freedom for numerator

196 XIII. References Automotive Industry Action Group, American Society for Quality Control. Supplier Quality Requirements Task Force: Fundamental statistical process control: reference manual, AIAG, 1991, p. 161 Automotive Industry Action Group, Chrysler Corporation, Ford Motor Company, General Motors Corporation: Statistical process control (SPC): reference manual, AIAG, 1995, p Kani, G. K., Arif, H. O.: Statistical Process Control, Wisdom House, 010, p. 0 Forrest W. Breyfogle III: Implementing Six Sigma: Smarter Solutions Using Statistical Methods. Wiley and Sons, Inc., New York, 1999 Lawrence D. Brown, T. Tony Cai and Anirban DasGupta. Interval Estimation for Binomial Proportion Statistical Science 001, Vol. 16, No., , Douglas Downing, Ph.D., and Jeffery Clark. Ph.D. Business Statistics. Barron s Educational Series. Inc Six Sigma Black Belt Training Books. Six Sigma Qualtec , Tenner, A.R., DeToro, I.J. (1993): Total Quality Management, Three Steps to Continuous Improvement, Addison-Wesley Publishing Company, Reading, Massachusetts Kövesi. J., Topár, J. (szerk) (006): A minőségmenedzsment alapai. Typotex, Budapest Measurement Systems Analysis Reference Manual, nd ed., (1995). Chrysler Corp., Ford Motor Corp., General Motors Corp. Pegels, C.C. (1995): Total Quality Management, A survey of its important aspects, Boyd & Fraser Publishing Company, Danvers, Massachusetts Prékopa András. Valószínűségelmélet. Mûszaki Kiadó 1974 Meszéna György-Ziermann Margit. Valószínûségelmélet és matematikai statisztika. Közgazdasági és Jogi Kiadó Tandori Károly. Valószínűségszámítás. József Attila Tudományegyetem, Szeged (Egyetemi egyzet), Ágnes Szendrei. Diszkrét Matematika. (Logika, Algebra, Kombinatorika) Polygon Jegyzettár,