Quality Improvement in the Modern in the Mocem Business Environment CHAPTER OUTLINE 1-1 THE MEANING OF QUALITY 1-4 OTHER ASPECTS OF QUALITY AND QUALITY IMPROVEMENT CONTROL AND IMPROVEMENT 11.1 Dimensions of Quality 1-4.1 Quality Philosophy and 1-1.2 Quality Engineering ‘Management Strategies 14.2 The Link Between Quality and Terminology 1-2. A BRIEF HISTORY OF QUALITY Productivity CONTROL AND IMPROVEMENT 1-4.3 Quality Costs 13. STATISTICAL METHODS FOR 1-44 Legal Aspects of Quality QUALITY CONTROL AND 1-45 Implementing Quality CHAPTER OVERVIEW Ihe use of statistical methods and other problem-solving techniques used by our society. These products consist of clothing, as well as services This book is about 1 to improve the quality of the products manufactured goods such as automobiles, computers, and rich as the generation and distribution of electrical energy, public transportation, bank- ing, and health care. Quality-improvement methods can be applied to any area within & company or organization, including manufacturing, process development, engineering design, finance and accounting, marketing, and field service of products, This text presents the technical tools that are needed to achieve quality improvement in these organizations.ONMENT vEME THE MODERN BUSINESS ENVIR UALITY IMPROVEMENT IN 2 CHAPTER 1 Ql ; ie fi ality, quality improve: satis impoeten menodology, and overview the statistical tools essential for mod. for implementing quality improvement is also given. V1.1 THE MEANING OF QUALITY AND QUALITY IMPROVEMENT - We may define quality in many ways. Most people have a conceptual understand quality as relating to one or more desirable characteristies that product or s should possess. Although this conceptual understanding is certainly a useful st point, we will give a more precise and useful definition, ; Quality has become one of the most important consumer decision factors in the Selection among competing products and services. The phenomenon is. widespread, regardless of whether the consumer is an individual, an industrial organization, a rea Store, or a military defense program. Consequently, understanding and improving Guality is a key factor leading to business success, growth, and an enhanced competitive position. There is a substantial return on investment from improved quality and from successfully employing quality as an integral part of overall business strategy. In ‘i section we give some operational definitions of quality and quality improvement, We _ begin with a brief discussion of the different dimensions of quality and some basic _ terminology. ing of erVicg: starting “LL. Dimensions of Qu: ee ty The quality of a product can be evaluated in several differentiate these different dimensionsvof quality. discussion of eight components or dimensions of q concerning these dimensions of quality as follows: ways. It is often very important Garvirl (1987) provides an excellen luality. We summarize his key point 1. Performance (will the product do the int otential customers usually evaluate a certain specific functions and determin ple, you could evaluate spreadsheet s which data manipulation Operations tl -. Outperforms another with Tespect to the execution, speed, 2) Reliability (how often does the product fail?) r 7 » Such as many appliances, automobiles, or airplanes, wil fe. For example, you shoul : Te occasional repair, but if the car require Frequent repair, we say that i is unreliable. This ic an industry in which th ustomer’s view of is greatly impacted by the reliability dimension = 5, tended job?) Product to determine if it will perfom how well it performs them/For ¢ oftware packages for a PC to determine hey perform. You may discover that ond \ THE MEANING OF QUALITY AND QUALITY IMPROVEMENT — 3 Durability (how long does the product last?) This is the effective service life of the product. Customers obviously want products that perform satisfactorily over a long period of time. Again, the au- tomobile and mijor appliance industries are examples of businesses where this dimension of quality is very important to most customers, 4. Serviceability (how easy is it to repair the product?) There are many industries in which the customer's view of quality is directly influenced by how quickly and economically a repair or routine maintenance activity can be accomplished, Examples include the appliance and automobile industries and many types of service industries (how long did it take a credit card company to correct an error in your 5. Aesthetics (what does the product look like?) ‘This is the visual appeal of the product, often taking into account factors such as style, color, shape, packaging alternatives, tactile characteristics, and other sensory features. For example, soft-drink beverage manufacturers have relied ‘on the visual appeal of their packaging to differentiate their product from other competitors. 6. Features (what does the product do?) Usually, customers associate high quality with products that have'added fea- tures; that_is_thase that have features beyond the basic performance of the competition. For example, you might consider a spreadsheet software package to be of superior quality if it had built-in statistical analysis features while its competitors did not. Perceived Quality (what is the reputation of the company or its product?) In many cases, customers rely on the past reputation of the company concern- of its products. This reputation is directly influenced by failures of the product that are highly visible to the public or that require product recalls, and by how the customer is treated when a quality-related problem with the product is reported. Perceived quality, customer loyalty, and repeated business re closely interconnected. For example, if you make regular business trips us- \d the flight almost always arrives on time and the ing a particular airline, an airline company does not lose or damage your luggage, you will probably pre- 0 fly on that carrier instead of its competitors. made exactly. as the designer A ing quality fer Conformance to Standards (is the product intended?) We usually think of a high-quality product as one that exactly meets the re- uirements placed on it. For example, how well does the hood fit on a new Gar? Is it perfectly flush with the fender height, and is the gap exactly the same on all sides? Manufactured parts that do not exactly meet the designers 1 Gquirements can cause significant quality problems when they are used as the components of a more complex assembly. An automobile consists of several thousand parts. If each one is just slightly to0 big oF to0 small, many of the components will not ft together properly, and the Vehicle (or its major subsys- tems) may not perform as the designer intended,+ HAPTER 1 QUALITY IMPROVEMENT INTHE MODERN BUSINESS ENVIRONMENT We see from the foregoing discussion that quality is indeed a multifaceted entity, Consequently, a simple answer to questions such as “What is quality?” or “What is qua ity improvement?” is not easy. The traditional definition of quality is based on the view. point that products and services must’ meet the requirements of those who use them. oint that products and services Mee Definition ( Quality means fitness for use.” ‘There are two general aspects of fitness for yse: quality of design and quality of conformance. All goods and services are produced in various grades or levels of qual- ity. These variations in grades or levels of quality are intentional, and, consequently, the appropriate technical term is quality of design. For example, all automobiles have as their basic objective providing safe transportation for the consumer. However, autonio- biles differ with respect to size, appointments, appearance, and performance. These dif- ferences are the result of intentional design differences between the types of automo- biles. These design differences include the types of materials used in construction, specifications on the components, reliability obtained through engineering development of engines and drive trains, and other accessories or equipment. The quality of conformance is how well the product conforms to the specifications required by the design. Quality of conformance is influenced by a number of factors, in- cluding the choice of marfafacturing processes, img and supervision of the work- force, the type of quality-assurance system used (process controls, tests, inspection ac- tivities, etc.), the extent to which these quality-assurance procedures are followed, and the motivation of the workforce to achieve quality. Unfortunately, this definition has become associated more with the conformance as- pect of quality than with design. This is in part due to the lack of formal education most designers and engineers receive in quality engineering methodology. This also leads to much less focus on the customer and more of a “conformance-to-specifications” ap- proach to quality, regardless of whether the product, even when produced to standards, was actually “fit-for-use” by the customer. Also, t ‘act there is still a widespread belief that uality is a problem that can be dealt with solely in manufacturing, ot that the only way quality can be improved is by “gold-plating” the product. ™ We prefer a modern definition of quality: Definition Quality is inyersely proportional to variability.‘THE MEANING OF QUALITY AND QUALITY IMPROVEMENT Unites Japan States me Figure 1-1 Warranty costs for transmissions. Note that this definition implies chair variability! in the important characteristics of a product decreases, the quality of the product increases) As an example of the opera: ago, one of the automobile com ales in the United States performed a comparative study of a transmission that was Fenutactared i domestic plant and by a Japanese supplier, An analysis of wat- “janty claims and repair Tosts indicated that there was a striking difference between ap Eas vouTves of production, With the Japanese-produced transmission having much Tagorcosis, as shown in Fig. I-1. As part of the study to discover the cause of this lected random samples of trans- difference in cost and performance, the company se Gausions from each plant, disassembled them, and measured several critical quali characteristics._ Figure 1-2 is generally representative of the results ofthis study. Note that the dis- tribution of the critical characteristics for the transmissions ‘manufactured in the United States Takes up about 75% of the width of the specifications, implying that very few Zonconforming units would be produced. In fact, the plant was roducing at a quali aris quite good, based on-the generally accepted view of quality within the jhich the same criti: level Enmpay Hone Tp a Ae eal Charar i about 25% of the specification band. As a result, there is considerably less variability in the critical qualt characteristics of the Japanese-built transmissions in comparison to those builtin the United States, eThere are two obvious questions here: Why di the Japanes do this? The answer to the “why” question is obvious from examination of Fig. 1-1. Re: Suced variability has directly translated into lower cost, Furthermore, the Japanese- Co SONS SHITED gears more TOGA, an SIiTed Fears more smoothly, ran more quietly, and were generally perceived by the customer As supe to those built domestically. Arewer repairs and warranty clajms means less fer e and the reduction of wasted time, @ffort, and money: Thus, quality truly is inversely proportional fo variability. Furthermore, it ean bbe com- municated very precisely in a language that everyone (particularly managers and execu tives) understands—namely, Money good. AS aon whch vai i cl ight ‘We ing to unwanted oF harmful variability. There are situations in which varial ae “{ really like Chinese food but I don’t want to eat it every’ ni ry good friend Bob Hogg hat pointed out do this? How did they112 SAArTHR ) QUALITY ANTROVIMINT 02 THE MODERN MUSINESS ENVIRONMENT rs Tepe vai Vipers 1-2 Diewmemione of ites dimensions foe tranemis: nee Now did the Japanese do this? The angwer lies in the systematic and effect " etlective use ‘he methosts described in this book. It also leads to the following defmition of improvement, . ™~ Excessive variability in process performance often results in waste. For example, con. “adr the Waste mromey- time ani @fT6et that is associated with the repai Fig. 1-1 Therefore, an alternate and highly useful definition is that quality improveme 2 the reduction of Wi his definition 1s particularly effective in service industries, Where there may not be as ‘many things that can be directly measured (like the transmis, sees crcal dimensions in Fig. 1-2) In service industries, a quality problem may be as SUE tisate the conection of which requites effort and expense. By improvig he service process, this wasted effort and ‘expense can be avoided, Ss frescat some quality engineering terminology that is used throughout te rc Quality Engincering Terminology Every product possesses a number of clements that jointly describe what the user of cor sumer thinks of as quality. These parameters are often called quality characteristics Quality characteristics may be of several types: 1, Physical: Length, weight, voltage, viscosity 2 Sensory: taste, appearance, color 3. Time Orientation: ‘liability, durability, serviceabilityb ‘THE MEANING OF QUALITY AND QUALITY IMPROVEMENT 7 Note that the different types of quality characteristics can relate directly or indirectly to the dimensions of quality discussed in the previous section. Quality engineering is the set of operational, managerial, and enginecring activities that a company uses to ensure that the quality characteristics of a product are at the nomi- nal or required levels. The techniques discussed in the book form much of the basic ‘methodology used by engineers and other technical professionals to achieve these goals. Most organizations find it difficult (and expensive) to provide the customer with Products that have quality characteristics that are always identical from unit to unit, or are at levels that match customer expectations. A major reason for this is variability. There is a certain amount of variability in every product; consequently, no two products are ever identical. For example, the thickness of the blades on a jet turbine engine im- eller is not identical even on the same impeller. Blade thickness will also differ be- tween impellers. If this variation in blade thickness is small, then it may have no impact on the customer. However, if the variation is large, then the customer may perceive the unit to be undesirable and unacceptable. Sources of this variability include differences in materials, differences in the performance and operation of the manufacturing equip- ment, and differences in the way the operators perform their tasks, This line of thinking led to the previous definition of quality improvement. Since variability can only be described in statistical terms, statistical methods play a central role in quality improvement efforts. In the application of statistical methods to quality engineering, it is fairly typical to classify data on quality characteristics as either attributes or variables data. Variables data are usually continuous measurements, such as length, voltage, or viscosity. Attributes data, on the other hand, are usually discrete data, often taking the form of counts. We will describe statistical-based quality engi- neering tools for dealing with both types of data. Quality characteristics are often evaluated relative to specifications. For a manufac- tured product, the specifications are the desired measurements for the quality character- istics on the components and subassemblies that make up the product, as well as the desired values for the quality characteristics in the final product, For example, the diam- eter of a shaft used in an automobile transmission cannot be foo large or it will not fit into the mating bearing, nor can it be too small, resulting in a loose fit, causing vibra- tion, wear, and early failure of the assembly. In the service industries, specifications are typically in terms of the maximum amount of time to process an order or to provide a particular service. ; ; ‘A value of a measurement that corresponds to the desired value for that qualit chavacteristic is called the nominal or target value for that characteristic. These target values are usually bounded by a range ‘of values that, most typically, we believe will be sufficiently close to the target so as to not impact the function or performance of the wroduct if the quality characteristic is in that range,)The lar quality ‘haracteristic is called the upper specification limit (USL), and the smallest al- i lled the lower iL), iowable value for a quality characteristic is called t i ty characteristics have specification limits on only one side of the target. For cramps te gompresiv guengh of component sed in an automobile bunpe ey areal valutand a lower. ‘ification limit, but not an upper specification limit, specif. Cem Aye Ged 0 chek defective W LL nye (tend - oo non defect chee K Conk iw te weal - ' ake 4/ ke 8 CHAPTER | QUALITY IMPROVEMENT IN THE MODERN BUSINESS ENVIRONMENT Specifications are usually the result of the engineering design process for the uct, Traditionally, design engineers have arrived at a product design confi through the use of engineering science principles, which often results in the des, specifying the target values for the critical design parameters. Then prototype cons tion and testing follow, This testing is often done in a very unstructured manner, yi the use of statistically bused experimental design procedures, and without much interat tion with or knowledge of the manufacturing processes that must produce the com nent parts and final product. However, through this general procedure the specifica limits are usually determined by the design engineer. Then the final products releaey to manufacturing. We refer to this as the over-the-wall approach to design, Problems in produet quality usually are greater when the over-he-Wall approach design is used. In this approach, specifications are often set without regard to 7 ; the inher. ent variability that exists in materials, praegsses, and other pats of the system, which sults in components or products that are woncohforming; that is, tht fail 10 meet cs more of its specifications cific type of failure is called" nonconformity’ A nop. conforming OTE a6 neces uit fr use; for example, a tetergent may have 4 concentration of active ingredient that is below the lower specification limit, but it nay still perform acceptably if the customer uses a greater amount of the product. A noncon. forming product is considered defective if it has one or more Aefects, which are non. conformities that are serious enough to significantly affect the’safe or effective use of the product. Obviously, failure on the part of a company to improve its manufacturing processes can also cause nonconformities and defects. The over-the-wall design process has been the subject of much attention in the last 20 years. CAD/CAM systems have done much to automate the design process and to more effectively translate specifications into manufacturing activities and processes. De- sign for manufacturability and assembly have emerged as an important part of overcon: ing the inherent problems with the over-the-wall approach to design, and most engineers receive some background on those areas today as part of their formal education. The recent emphasis on concurrent engineering has stressed a team approach to design, with specialists in manufacturing, quality engineering, and other disciplines working together with the product designer at the earliest stages of the product design process. Furthermore, the effective use of the quality improvement methodology in this book, at all levels of the process used in product design, development, and manufacturing, plays acrucial role in quality improvement. BUTation, 1-2 A BRIEF HISTORY OF QUALITY CONTROL AND IMPROVEMENT Quality always has been an integral part of virtually all products and services. Howeveh Our awareness of its importance and the introduction of formal methods for quality com trol and improvement have been an evolutionary development, Table 1-1 presents * timeline of some of the important milestones in this evolutionary process. We wil briefly discuss some of the events on this timeline. -~‘A BRIEF HISTORY OF QUALITY CONTROL AND IMPROVEMENT = 9 ‘Table 1-1 A Timeline of Quality Methods 1700-1900 ity i i Quay a lel determined by the efforts of an individual craftsman, es Eig bodies standardized, interchangeable pars to simplify assembly. Frederick W. Tay 4 a a “Scientie Management” principles divide work into smaller, ‘units—the first approach to dealing with more e ee ing with more comple products and processes. The focus was on productivity. Later contributors were Cire and , Gant. 1900-1930 Henry Ford—th i ity and quality; he nmemiiy i= Pete refinement of work methods to improve productiv- process inspection, sped mistake-proof assembly concepts, self-checking, and in- 1901 First standards laboratories é ¢ 1907-1908 AT&T begins el testing of , 1908 Ar sTosea (oars suieal products and materials. Mealy Sonrolat Guess slew) introduces the t-distribution—results from his work on ee WWI— British government begins a supplier certfiation program, fears a Saspecion Asotin formed in England; his later becomes the Institute of 19206 agar Labor forms quality department —emphasizing quay inspection and B. P, Dudding at General Electric in England uses statistial methods to contr} the quality of electric lamps. at 1922-1923 R.A. Fisher publishes series of fundamental papers on designed experiments and their application tothe agricultural sciences. 1924 WA, Shewhart introduces the control chart concept in a Bell Laboratories technical ‘memorandum. 1928 sampling methodology i developed and refined by H. F. Dodge and HG. Romig at Bell Labs. 1931 W.A.Shewhart publishes Economic Control of Quality of Manufactured Product—ou edstieal methods for use in production and control chart methods 1932 WA. Shewhart gives lectares on statistical methods in production and control charts at the ai $ University of London. 1932-1933 British textile and wooden industry and German chemical indusry begin use of designed experiments for product/ process development. 1933 ‘The Royal Statistical Society forms the Industrial and ‘Agricultural Research Section. 1938 WE. Deming invites Shewhart to present seminars on control charts at the U.S. Department of Agriculture Graduate School, 1940 The US. War Department publishes guide for using ‘control charts to analyze process data. 1940-1943 Bell Labs develop the forerunners ofthe miliary standard sampling plans for the U.S. Army. 1942 In Great Britain, the Ministry of Supply Advising Service on Statistical Methods and Quality Control is formed. state: ; 1942-1946 ‘Training courses on statistical quality control are given to industry: more than 15 quality tocietes are formed ia North America ‘al Quali jns publication. 1944 ‘edustrial Quality Control begins PUD , 1946 ‘The American Society for Quality ‘Control (ASQC) is formed as the merger of various ality societies. i seeming i invited 1 Japan by the Economic and Scientific Services Section ofthe US. War : eupation forces in rebuilding Japanese industry. formed. Department to help n ‘on of Scientists and Engineers (JUSE) is Tapanese Union The Japancte sco give statistical quality control seminars 10 Japanese industry. 1946-1949 ae)10 Tae 1-1 reas, 1980 19508 1951 1951+ 1954 1957 1959 1960 1961 1960s 1969 1970s 1975-1978 1980s 1984 1986 1988 1989 1990s CHAPTER 1 QUALITY (Continued) WVEMENT IN THE MODERN BUSINESS ENVIRONME\y IMPRO en esas slucaton of fpanse indus mansges tats Deming begins in to be widely taught in Japan. eho ae rk Ishikawa introduces the cause-and-effect diagram, Professor son statistical quality control by Eugene Grant and A.J. Duncan 2peae, Dea. V-Feigenbaum publishes the frstedton of kis book, ont Quality Cone JUSE eases the “Deming Prize" fr significant achievement in quality cone “ i jodology. " : eres oie Publish fundamental work on using designed expe: & onse surface methodology for process Optimization; focus is on chemical ind ng plications of designed experiments in the chemical industry grow steadily ater gis Ap Dr. Joseph M. Juran is invited by the Japanese to give some lectures‘on Quality m, and improvement. anaBe me, trol chart, Al ality conta British statistician E. S. Page introduces the cumulative sum (CUSUM) cont J.-M. Juran and F. M. Gryna’s Quality Control Handbook is first Published, Technomerris (a journal of statistics forthe physical, chemical, and Bnei scien, established; J. Stuart Hunter is the founding editor. is’ S. Roberts introduces the exponentially weighted moving avera G.E.P. Box and J. S, Hunter write fundamental Papers on 2'-P factorial designs. ‘The quality control circle concep is introduced in Japan by K. Ishikawa, National Council for Quality and Productivity is formed in British Productivity Council Courses in statistical quality control become wi programs. Zero defects (ZD) programs are introduced in certain U.S, industries, Great Britain as part of the "idespread in Industrial Engineering academic ‘The American Statistical Associat and Productivity; this later becor Box and others visit tistical methods, ‘The Malcolm Baldrige National The journal Quality Engineeriy Motorola’s six-sigma initiat 1S0 9000 certification activi Grow steadily; many states y tion (ASA) establishes the Ad Hoc Committee on Quality mes a full Section of the ASA. Japan, noting the extensive use of. designed experiments and other st I Quality Award is established by the U.S. Congress. ing appears, ve begins, ¢ ties increase in U.S, industry; applicants for the Baldrige ® ’Ponsor quality awards based on the Baldrige criteria.1995 1997 1998 ‘A BRIEF HISTORY OF QUALITY CONTROL AND IMPROVEMENT 11 Many undergraduate engineting programs require foal courses instil techniques focusing on bse methods fr process chareerzation and improves " Moo’ sicsuma approach peat to oer nds American Society for Quali mercan Soci a ity Control becomes the A ity, at tempting to indicat the broader spe of he qully improvement fd Frederick W. Taylor introduced some principles of scientific management as mass Prod cee beaan . develop prior to 1900. Taylor pioneered dividing work at aa at rd could be manufactured and assembled more easily. His work ed! to substantial improvements in productivity. Also, because of standardized pro- sembly methods, the quality of manufactured goods was positively im- pacted as well. However, along with the standardization of work methods came the con- cept of work standards—a standard time to accomplish the work, or a specified number of units that must be produced per period. Frank Gilbreth and others extended this con- cept to the study of motion and work design. Much of this had a positive impact on pro- ductivity, but it often de-emphasized the quality aspect of work. Furthermore, if carried to extremes, work standards have the risk of halting innovation and continuous improve- ment, which we recognize today as being a vital aspect of all work activities. Statistical methods and their application in quality improvement have had a long his tory. In 1924, Walter A. Shewhart of the Bell Telephone Laboratories developed the statis- tical control-chart concept, which is often considered the formal beginning of statistical of the 1920s, Harold F. Dodge and Harry G. Romig, both developed statistically based acceptance sampling as an the middle of the 1930s, statistical quality-control methods were in wide use at Westem Electric, the manufacturing arm of the Bell System. eof statistical quality control was not widely recognized by industry. World War II saw a greatly expanded use and acceptance of statistical quality- control concepts in manufacturing industries. Wartime experience made it apparent that statistical techniques were necessary to control and improve product quality. The ‘Ameri- can Society for Quality Control was formed in 1946. This organization promotes the use of quality improvement techniques for all types of products and services. It offers a umber of conferences, technical publications, and training programs in quality assur- of reliability engineering, the introduc- ‘ance. The 1950s and 1960s saw the emergence the in tion of several important textbooks on statistical quality control, and the viewpoint that quality is a way of managing the organization. ; : Tn the 1950s, designed experiments for product and process improvement were first introduced in the United States. The initial applications wets in the chemical industry. These methods were widely exploited in the chemical industry, and they are often cited as one ofthe primary reasons that the U.S. chemical industry is one of the most compet Tittle business to foreign companies. The spréad ofthese itive in the world and has lost ; : methods urside the chemical industry was relatively slow until the late 1970s oF ‘Western companies discovered that their Japanese ComPOY: rents since the 1960s for process early 1980s, when many Western comP D tors had been systematically using designed exper troubleshooting, new process development, evalua quality control. Toward the end of Bell Telephone Laboratories, alternative to 100% inspection. By However, the value tion of new product designs,oN “HAPTER | QUALITY IMPROVEMENT IN THE MODERN BUSINESS ENVIRON Egy improvement of reliability and field performance of products, product design, includit ‘Sof sparked furd y sned c ety efforts to introduce the methodology in engineering and development organizag Me industry, as well as in academic engineering curricula. WZations iy 1980 there has been a profound growth in the w provement inthe United States. This has been motivated in a oe loss of business and markets suffered by many domestic compan; Wi during the 1970s, For example, the U.S. automobile industry was neatly de hs foreign competition during this period. One domestic automobite ag 1 destroyed by operating losses a nearly $1 milion per hour in 1980. The dogg Pay eae cal methods have played a central role in the re-emergence of US. induc OF statis, management srvtures have also emerged a frameworks in which js Va improvement. In the next two sections we bri iefly discuss the statistic the central focus of this book and give an overview of some i = SC Of statistical met Plement quai ‘Methods that w ey aspects of TQM, VK cexristica METHODS FOR QUALITY CONTROL AND IMPROVEMENT This textbook concentrates on statistical and engineering technology useful improvement. Specifically, we focus on thtee major areas: statistical Pp design of experiments, and (to a lesser extent) acce these techniques, a number of other statistical tools lems and improving the performance of production tools is illustrated in Fig. 1-3, which presents a PI in quatiy FOLESS C0 tance sampling. In addition | ‘re useful in analyzing quality prob, Processes. The role of some of these roduction process as a system with, Controllable inputs, 4 % % Input raw materials, components, = = YF and ER subassemblies : = Gualy characteristic Output reduc | 4 2 % Uncontrlabie inputs Figure 1-3 Production process inputs and outputs.STATISTICAL METHODS FOR QUALITY CONTROL AND IMPROVEMENT 13, uct} Sample average Let Time (or sample number) Figure 1-$ A typical contol char. set of inputs and an output. The inputs.x),.x,, . . - » %» are controllable factors, such as temperatures, pressures, feed rates, and other process variables. The inputs z1, 22, - - - + are uncontrollable (or difficult to control) inputs, such as environmental factors or properties of raw materials submitted by the vendor. The manufacturing process trans- forms these inputs into a finished product that has several quality characteristics. The ‘output variable y is a measure of process quality. ‘A control chart is one of the primary techniques of statistical process control or SPC. A typical control chart is shown in Fig. 1-4. This chart plots the averages of mea~ surements of a quality characteristic in samples taken from the process versus time (or the sample number). The chart has a center line (CL) and upper and lower ‘control limits (UCL and LCL in Fig. 1-4). The center line represents where this process characteristic should fallif there are no unusual sources of variability present. The control limits are determined from some simple statistical considerations that we will discuss in Chapters 4, 5, and 6. Classically, control charts are applied to the output variable(s) in a system such as in Fig. 14, However, in some cases they can be usefully applied to the inputs as well toring technique; when unusual The control chart is a very useful process monit sources of variability are present, sample averages will plot outside the control limits. ‘This is a signal that some investigation of the process should be made and corrective ac- tion to remove these unusual sources of variability taken. Systematic use of a control chart is an excellent way to reduce variability. tremely helpful in discovering the key variables influ- ‘A designed experiment is ex! ul ables in encing the quality characteristics of interest inthe process. A designed experiment is an approach to systematically varying the controllable input factors in the process and de- termining the effect these factors have on the output product parameters, Statistically designed experiments are invaluable in reducing the variability in the quality character fsties and in determining the levels of the controllable variables that optimize process type of designed experiment is the factorial design, varied together in such a way that all possible combinations of factor Figure 1-5 shows two possible factorial designs forthe process in Fig. 1-3, fo the 2 and p = 3 controllable factors. In Fig. 1-Sa the factors have two levels, in which factors are levels are tested. the cases lowry TIN BUSINESS ENVIRON} IMPROVEMENT IN THE MODERN IAL CHAPTER 1 QU! 1 " ign] Low High (a) Two factors, x, and xg ——_ T a a T (by Tee factors, x, 4, and Figure 1-5 Factorial designs forthe process in Fig. and high, and the four possible test combinations in this factorial experiment form th) comers of a square. In Fig. 1-5b there are three factors each at two levels, giving ane) Periment with eight test combinations arranged at the corners of a cube. The distiby tions at the comers of the cube represent the process performance at each combinatit Of the controllable factors x;, x», and x3, It is clear that some combinations of factt levels produce better results than others. For example, increasing x, from low to hit’, increases the average level of the Process output and could shift it off the target va (7), Furthermore, process variability seems to be substantially reduced when we opert! the process along the back edge of the cube, where 42 and x, are at their high levels. ee experiments are a major off-line quality-control tool, because they are oft ares rent activities and the early Stages of manufacturing, rather than &! Once we have identified «In sa: TY Pay a crucial role in reducing varzbiliy | P of important variables th: utp! is usually necessary to model the relationshi les that affect the process outpt J nship between the influential input variables ® | { | | iLSTATISTICAL METHODS FOR QUALITY CONTROL AND IMPROVEMENT 15 peas Salty characteristics. Statistical techniques useful in constructing such mod- experimen pa analysis and time series analysis. Detailed discussions of designed Montgomery Ba s ee estate Cr series modeling are in Montgomery (1997), When the important variables tase eae aaeat ees 5 en identified and the ionshi between the important variables and the process output has been quanfed, then snow line statistical process-control technique for monitoring and surveillance of the process can be employed with considerable effectiveness. Techniques such as control charts can be used to monitor the process output and detect when changes in the inputs are required to bring the process back to an in-control state. The models that relate the influ- ential inputs to process outputs help determine the nature and magnitude of the adjust- ments required. In many processes, once the dynamic nature of the relationships be- tween the inputs and the outputs are understood, it may be possible to routinely adjust the process so that future values of the product characteristics will be approximately on- target. This routine adjustment is often called engineering control, automatic control, or feedback control. We will briefly discuss these types of process control schemes in Chapter 11, and illustrate how SPC methods can be successfully integrated into a manu- facturing system in which engineering control is in use. The third area of quality control and improvement that we discuss is acceptance sampling. This is closely connected with inspection and testing of product, which is one of the earliest aspects of quality control, dating back to long before statistical methodol- ‘ogy was developed for quality improvement. Inspection can occur at many points in a process. Acceptance sampling, defined as the inspection and classification of a sample of units selected at random froi -oF lot-and theultimate decision about Gisposition of the lot, usually occurs at two points: incoming raw materials or compo- ents, or final production. Several different variations of acceptance sampling are shown in Fig. 1-6. In Fig. I-6a, the inspection operation is performed immediately following production, be fore the product is shipped to the customer. This is usually called outgoing inspection. Figure 1-66 iustates Snoring inspection a is, a situation in which lots of batches of product are sampled as they are received from the supplier. Various ot-dispositioning decisions are illustrated in Fig. 1-6c. Sampled lots may either be accepted or rejected. Items in a rejected lot are typically either scrapped or recycled, or they may be reworked or replaced with good units. This latter case is often called rectifying inspection. Modern quality assurance systems usually place less emphasis on acceptance sam- pling and attempt to make statistical process control and designed experiments the focus of their efforts. Acceptance sampling tends to reinforce the “conformance to specification’ view of quality and does not have any feedback into either the production process or engi- neering design or development that would necessarily lead to quality improvement. Figure 1-7 shows the typical evolution in the usé of these techniques in most orgami- zations, At the lowest level of maturity, management may be completely unaware cr quality issues, and there is likely to be no effective organized quality improves effort. Frequently there will be some modest applications of acceptance-samp! me methods, usually in receiving inspection. The first activity as maturity increases 1S16 CHAPTER 1 QUALITY IMPROVEMENT IN THE MODERN BUSINESS ENVIRONMENT: peter et] “ee (2) Outgoing inspection Ship rm (1) Receivingfineoming inspection Accept Ship -O {€) Disposition of lots Figure 1-6 Variations of acceptance sampling. | intensify the use of sampling inspection. The use of sampling will increase until we re. alize that quality cannot be inspected or tested into the product. ‘At that point the organization usually begins to focus on process improvement Statistical process control and experimental design potentially have major impact o | ‘manufacturing, product design activities, and process development. The systematic in- troduction of these methods usually marks the start of substantial quality, cost, and pro-| ductivity improvements in the organization. At the highest levels of maturity, companie,_ use designed experiments and statistical process control methods intensively, and ma | relatively modest use of acceptance sampling. 100, Acceptance sompling Percent of appication Design of experiments { Time | Figure 1-7 Phase diagram of the - eee ee use of quality iOTHER ASPECTS OF QUALITY CONTROL AND IMPROVEMENT 17 Upper speciation - limit T Process mean, — a Lower specification limit ‘Acceptance Statistical Design of sampling process contol experiments Figure 1-8 Application of qualty-engineering techniques and the systematic re- duction of process variability. The primary objective of quality engineering efforts is the systematic reduction of variability in the key quality characteristics of the product. Figure 1-8 shows how this happens over time. In the early stages, when acceptance sampling is the major technique in use, process “fallout,” or units that do not conform to the specifications, constitute a high percentage of the process output. The introduction of statistical process control will stabilize the process and reduce the variability. However, it is not satisfactory just to meet requirements —further reduction of variability usually leads to better product per- formance and enhanced competitive position, as was vividly demonstrated in the auto- mobile transmission example discussed earlier. Statistically designed experiments can be employed in conjunction with statistical process control to minimize process variabil- ity in nearly all industrial settings. 1-4 OTHER ASPECTS OF QUALITY CONTROL AND IMPROVEMENT ues are the critical technical tools for quality control and improvement, to be used most effectively they must be implemented within and be part of a management system that is quality driven, In effect, the management system ust direct the quality improvement philosophy and ensure its implementation in all rial frameworks used to accomplish this is aspects of the business. One of the manage though other widely used names include ), and six- Although statistical techniq sigma. In this section we give a brief overview of some of the key elements of TQM. We discuss some of the important quality philosophies, the link between quality and productivity, economic and legal implication of quality, and some aspects of implementation. /18 te ESS ENVIRONME: sjuapTeR | QUALITY IMPROVEMENT IN THE MODERN BUSINESS ENVIRONMENT Quality Philosophy and Management Strategies il e statist thodology of quality improvem, le have contributed to the statistical methodolog) proven, Pe fn terms of implementition and management philosophy, three individuals emerge as the leaders: W. E, Deming, J. M. Juran, and A. Y. Feigenbaum, We now view the approaches and philosophy of those leaders in quality management, : 7 Oe son that Tpanese manufacturers exhibited broad capabilities in statistical qualyy control and improvement during the 1970s and 1980s is the work of Dr. W. Edwards Deming. During World War Il, Dr. Deming worked for the War Department and the Census Bureau, Following the war, he became a consultant to ee ae and convinced their top management of the power of statistical mncthode and te importance of quality as a competitive weapon. This commitment to and use of statist on methods has been a key element in the expansion of Japan’s industry and economy The Jae Union of Scientists and Engineers created the Deming Prize for quality improvement in his honor, Until his death in 1994, Dr. Deming was an active conse anc speaker, he \was an inspirational force for quality improvement in this country and around the word, He firmly believed that the responsibility for quality rests with management; that is, ‘most of the opportunities for quality improvement require management action, and very few opportunities lie at the workforce or operator level. Dr. Deming was a harsh critic of ican management practices. ; mae ilosophy of Dr Deming is an important frainework for implementing quality and productivity improvement. This philosophy is summarized in his 14 points for man- agement. We now give a brief statement and discussion of Dr. Deming’s 14 points: 1, Create a constancy of purpose focused on the improvement of products and services. Constantly try to improve product design and performance. Invest ‘ment in research, development, and innovation will have long-term payback to the organization, 2, Adopt a new philosophy of rejecting poor workmanship, defective products, or bad service. It costs as much to produce a defective unit as it does to pro duce a good one (and sometimes more). The cost of dealing with scrap, re- work, and other losses created by defectives is an enormous drain on company resources, 3. Do not rely on mass inspection to “control” Sort out defectives, and at this point it is too 1; to produce these defectives. Inspection typic it is expensive, and it is often ineffective, defectives through process improvement, n 4. Do not award business to suppliers on th sider quality. Price is a meaningful meas is considered in relation to a measure of Of the item must be considered, not just quality. All inspection can do is late because we have already paid ally occurs too late in the process, Quality results from prevention of }ot inspection, 1e basis of price alone, but also com ure of a supplier's product only if it quality. In other words, the total cost the purchase price. When quality isOTHER ASPECTS OF QUALITY CONTROL AND IMPROVEMENT 19. considered, the lowest bidder frequently is not the low-cost supplier. Prefer- ar ou eI be Gitta {© suppliers who use modem methods of quality vemnent in.their business and who can demons ae lemonstrate process control and 5. ae on continuous improvement. Constantly try to improve the production and service system. Involve the workforce in these activities and make use of statistical methods, particularly the statistically based problem-solving tools discussed in this book. 6. Practice modern training methods and invest in training for all employees. Everyone should be trained in the technical aspects of their job, and in modern quality- and productivity-improvement methods as well. The training should encourage all employees to practice these methods every day. 7. Practice moder supervision methods. Supervision should not consist merely of passive surveillance of workers but should be focused on helping the em ployees improve the system in which they work. The number one goal of su- pervision should be to improve the work system and the product. 8. Drive out fear. Many workers are afraid to ask questions, report problems, or point out conditions that are barriers to quality and effective production. In many organizations the economic loss associated with fear is large; only man- agement can eliminate fear. 9. Break down the barriers between functional areas of the business. Teamwork among different organizational units is essential for effective quality and pro- ductivity improvement to take place. Eliminate targets, slogans, and numerical goals for the workforce. A target such as “zero defects” is useless without a plan for the achievement of this ob- jective. In fact, these slogans and “programs” are usually counterproductive, Work to improve the system and provide information on that. id work standards. These standards have-histori- toms of 10. 11, Eliminate numerical quotas an cally been set without regard to quality. Work standards are often symy management's inability to understand the work process and to provide an ef- fective management system focused on improving this process. 12. Remove the barriers that discourage employees from doing their jobs. Man- agement must listen to employee suggestions, comments, and complaints. The person who is doing the job knows the most about it and usually has valuable ideas about how to make the process work more effectively. The workforce is an important participant in the business, and not just an opponent in collective bargaining. hema - it ir ini .ducation for all employees. Ed- 13. Institute an ongoing program of training and e wcatlon in simple, powerful statistical techniques should be mandatory for all employees. Use of the basic SPC problem-solving tools, particularly the con- trol chart, should become widespread in the business. As these charts Cr widespread and as employees understand their uses, they will be more likelyMENT 20 CHAPTER 1 QUALITY IMPROVEMENT IN THE MODERN BUSINESS ENVIRON! ji i ss improver to look for the causes of poor quality and to Ce ee tig on me, Education is a way of making everyone partners in rent process. / = 14, Create a structure in top management that will vigorously advocate the frst 1, points, ings: First, there is a strong em, As we read Dr. Deming’s 14 points we notice two ae eae cienae proce ce phasis on change. Second, the role of management in su a how should this change dominating importance. However, what should be change ae yield of a semiconducise roots be stared? For example, i we want improve We YO 0 | Scene manufacturing process, what should we do? Ie in this rea Hat SSUNCA len come into play most frequently. To improve the seni nn ee of detective ai mine which sontolable os ae on the process ea se how the Lseeacgilledae ee ge in the process variables, Statistical methods, such as designe peo ‘and control charts, can contribute to these activities. es os of te founding fathers of statistical quality control. He worked fer Dy Water she at ATST Bell Labortis and as bee athe lang eget aualiy inprovement ever since. He was invited to speak to Japanese industry leader they began ther industrial tansfomation inthe early 1950s. He is the co-author (wit Frank M. Gryna) of the Quality Control Handbook, a standard reference for quality methods and improvement since its initial publication in 1957. Dr. Juranis less focused than Dr. Deming on statistical methods, The phy is based on organization for change and the im through what he calls “managerial breakthrough.” structured problem-solving process, Dr. Juran is a is required to improve quality. He shares with Dr. Turan philoso. plementation of improvement The breakthrough sequence is really 2 firm believer that management action Deming the viewpoint that most of the Dr. Armand V. Feigenbaum Dr. Feigenbaum first introduced the concept of com} i i in his hi i ipany-wide quality control in his his ie a eal Quatiy Control (the first edition was published in 1951), This book it luenced much of the early Philosophy of qualit man: it an i @ 1950s. In fact, many Japanese companies used the names tee eBay Scribe their efforts, sed the name “total quality control” to Dr. Feigenbaum is more co Proach to ii ‘ational aspects of Dr. Feigenbaum's work #OTHER ASPECTS OF QUALITY CONTROL AND IMPROVEMENT 21 important, as quality improvement does not usually spring forth as a “grass roots” activ- ity; it requires a lot of management commitment to make it work, ti ie oe descriptions of the philosophies of Deming, Juran, and Feigenbaum have Hghlighted both the common aspects and differences of their viewpoints. In this au- thor’s opinion, there are more similarities than differences among them, and the similar- ities are what is important. All three of these pioneers stress the importance of quality as an essential competitive weapon, the important role that management must play in im- plementing quality improvement, and the importance of statistical methods and tech- niques in the “quality transformation” of an organization, Sha Quality Management Total quality management (or TQM) is a strategy fo: implementing and managing qual- ity improvement activities on an organization-Wide basis, TQM began in the early 1980s, with the philosophies of Deming and Juran as the focal point. It evolved into a broader spectrum of concepts and ideas, involving participative organizations and work culture, customer focus, supplier quality improvement, integration of the quality system with business goals, and many other activities to focus all elements of the organization around the quality improvement goal. Typically, organizations that have implemented a TQM approach to quality improvement have quality councils or high-level teams that deal with strategic quality initiatives, workforce-level teams that focus on routine pro- duction or business activities, and cross-functional teams that address specific quality improvement issues. TQM has only had moderate success for a variety of reasons, but frequently be- cause there is insufficient effort devoted to widespread utilization of the technical tools of variability reduction. Many organizations saw the mission of TQM as one of training. Consequently, many TQM efforts engaged in widespread training of the workforce in the philosophy of quality improvement and a few basic methods. This training was usu- ally placed in the hands of human resources departinents, and much of it was.ineffec- tive. The trainers often had no real idea about what methods should be taught, and suc- cess was usually measured by the percentage of the workforce that had been “trained,” not by whether any measurable impact on business results had been achieved. Some general reasons for the lack of conspicuous success of TQM include (1) lack of top- down, high-level management commitment and involvement; (2) inadequate use of sta- tistical methods and insufficient recognition of variability reduction as a prime objec~ tive; (3) diffuse as opposed to focused, specific objectives; and (4) too much emphasis on widespread training as opposed to focused technical education, ‘Another reason for the erratic success of TQM is that many managers and execu- tives have regarded it as just another “program” to improve quality. During the 1950s defects and value engineering abounded, but they .d productivity improvement. During the heyday of ative, in which and 1960s, programs such as zero had little real impact on quality an ement, During TOM in the 1980s, another popular program was the quality is fre ini in which identifying the cost of quality (or the cost of nonquality, management worked on identifying ate ee “quality is free” devotees so cleverly put it). Indeed, Identificat22 ISINESS ENVIRONMENT LITY IMPROVEMENT IN THE MODERN BU! IN CHATTER 1 QU: i the “quality is free” cuss quality costs in Section 1-4.3), _ a ts . Be : bbs 4 ve very use ve seus gual eu pro may Pe fa iners often had no ie Leaders of this initiative had no Know edge about sy ST Cio + M completely Tailed to understand is role in quality improve tical methodolo; When TQM is wrappe' effective mm such as this, disaster is often 4 eer in " i .d around an ineffective progral nent yen is result. 'e0 the it id Registration “ ; Ce a Standards Organization (ISO) has developed a series of quality sen, reesaetading the ISO 9000 series, which is also an American National Standards ty stitute and an ASQ standard as well The focus of these standards is the quality system, including components such as 1. Management responsibility for quality Design control pee }. Document and data control 4, . Purchasing and contract management 5, Product identification and traceability 6. equipment Inspection and testing, including control of measurement and inspection 71, Process control 8. Handling of nonconforming product, corrective and preventive actions 9. Handling, storage, packagin; i and delivery of product, including serie activities, 10. Control of quality records A. Internal audits 12. Training 13. Statistical methodology ; for these and some other activiti i ave required their suppli tified under ISO 900 iva | 10 oF one of the derivate Standards, such as those developed i y the North Am ile i | 9000. Consequently, there is currently a Seen orlitn 4 substantial effort j eee Come cetfed to the appropriate standard, Think pany ose et OrBanizations 1 | Much of the focus of 1SO 9000 (and th ‘party registrars handle certification. i i ind the derivative standards) is on formal dos Trenton of the quality system. Organizati s) is on their documentation into Ii aeOTHER ASPECTS OF QUALITY CONTROL AND IMPROVEMENT 23 the most elementary techniques. Therefore, they concentrate largely on the bookkeep- ing, records, and paperwork aspects of certification. Ichas been estimated that ISO registration activities are (approximately) a 40 billion dollar annual business, worldwide. Much of this money flows to the registrars, auditors, and consultants. This amount does not include the intemal costs incurred by organiza- tions to achieve registration, such as the thousands of hours of engineering and manage- ‘ment effort, travel, internal training, and internal auditing. It is not clear whether any significant fraction of this expenditure has made its way to the bottom line of the regis- tered organizations. Many quality engineering authorities feel that ISO registration is largely a waste of effort. Often organizations would be far better off to “just say no to ISO” and spend a small fraction of that 40 billion dollars on their quality systems and another larger fraction on meaningful variability reduction efforts, develop their own in- ternal (or perhaps industry-based) quality standards, rigorously enforce them, and pocket the difference. The focus of these standards should be principally on variability reduction and yield enhancement efforts, with systems and paperwork playing a sup- porting role. Six-Sigma High-technology products with many complex components typically have many oppor- tunities for failure or defects to occur. Motorola developed the six-sigma program in the late 1980s as a response to the demand for these products. The focus of six-sigma is re- ducing variability in key product quality characteristics to the level at which failure or defects are extremely unlikely. Figure 1-9a shows a normal probability distribution as a model for a quality charac- teristic with the specification limits at three standard deviations on either side of the mean, Now it tums out that in this situation the probability of producing a product within these specifications is 0.9973, which corresponds to 2700 parts per million (ppm) defec- tive. This is referred_to_as_three-sigma quality performance, and it actually sounds pretty good. However, suppose we have a product that consists of an assembly of 100 ‘components or parts and all 100 of these parts must be nondefective for the product to function satisfactorily. The probability that any specific unit of product is nondefective is 0.9973 X 0.9973 X +++ X 0.9973 = (0.9973)! = 0.7631 That is, about 23.7% of the products produced under three-sigma quality will be defec- tive. This is not an acceptable situation, because many high-technology products are made up of thousands of components. An automobile has about 200,000 components and an airplane has several million! The Motorola six-sigma concept is to reduce the variability in the process so that the specification limits are six standard deviations from the mean, Then, as shown in Fig. 1-9a, there will only be about 2 parts per billion defective. Under six-sigma quality, the probability that any specific unit of the hypothetical product above is non- defective is 0.999998, or 0.2 ppm, a much better situation. When the six-sigma concept was initially developed, an assumption was made that when the process reached the six-sigma quality level, the process mean was still subject