A first course in quality engineering: integrating statistical and management methods of quality

Cover
Half Title
Title Page
Copyright Page
Table of Contents
	Preface to the Third Edition
	Preface to the Second Edition
	Preface to the First Edition
	Authors
	Chapter 1: Introduction to Quality
	1.1 A Historical Overview
	1.1.1 A Note about “Quality Engineering”
	1.2 Defining Quality
	1.2.1 Product Quality vs. Service Quality
	1.3 The Total Quality System
	1.4 Total Quality Management
	1.5 Economics of Quality
	1.6 Quality, Productivity, and Competitive Position
	1.7 Quality Costs
	1.7.1 Categories of Quality Costs
	1.7.1.1 Prevention Cost Expenses incurred in preventing the production of defective products are part of the prevention cost. Examples of expenditure under this category include:
	1.7.1.2 Appraisal Cost Appraisal cost is the cost incurred in appraising the condition of a product or material with reference to requirements or specifications. The following are the major components of this cost:
			1.7.1.3 Internal Failure Cost Internal failure cost is the cost arising from defective units produced but getting detected before being shipped to the customer. This is the cost arising from the lack, or failure, of process control methods, resulting in d
			1.7.1.4 External Failure Cost External failure cost is the cost arising from defective products reaching the customer. Making a bad product causes waste, and if it is allowed to reach the customer, it causes much more waste. The external failure cost incl
	1.7.2 Steps in Conducting a Quality Cost Study
	1.7.3 Projects Arising from a Quality Cost Study
	1.7.4 Quality Cost Scoreboard
	1.7.5 Quality Costs Not Included in the TQC
	1.7.6 Relationship among Quality Cost Categories
	1.7.7 Summary on Quality Costs
	1.7.8 A Case Study in Quality Costs
	1.8 Success Stories
	1.9 Exercise
	1.9.1 Practice Problems
	1.9.2 Mini-Projects
			Mini-Project 1.1 This exercise is meant to provide an experience in defining the quality of a “product.” Often, the first challenge for an engineer is to identify the “products” produced in an organization and the characteristics that define the quality o
			Mini-Project 1.2 This is an exercise in determining which expense belongs in which quality cost category and then analyzing the cost data to reveal any problems in the system. The data come from a chemical plant with annual sales of $100 million. Their op
			Mini-Project 1.3 A university ombudsman has the responsibility of resolving grievances brought to him/her by students relating to academic work, such as disputes on a grade, how the student was treated in the class, or any other issue related to academic
	Mini-Project 1.4 This project deals with a “large” system where the people involved are counted in millions, and the costs and benefits are counted in billions. It takes a bit of experience to get used to thinking in millions and billions.
	References
	Chapter 2: Statistics for Quality
	2.1 Variability in Populations
	2.2 Some Definitions
	2.2.1 The Population and a Sample
	2.2.2 Two Types of Data
	2.3 Quality vs. Variability
	2.4 Empirical Methods for Describing Populations
	2.4.1 The Frequency Distribution
			2.4.1.1 The Histogram The histogram is a very useful tool in the tool box of a quality engineer, as it helps in understanding the nature of the variability in a population. Since many quality problems relating to populations of products arise from excessi
			2.4.1.2 The Cumulative Frequency Distribution Another graph of the data that provides some very useful information is the cumulative frequency distribution. This is a graph of the cumulative percentages of frequencies shown in Table 2.2. Figure 2.3 shows
	2.4.2 Numerical Methods for Describing Populations
			2.4.2.1 Calculating the Average and Standard Deviation The average  is simply the arithmetic average or sum of all observations in the data divided by the number of observations. For the data shown in Table 2.1, . The standard deviation is the square ro
	2.4.3 Other Graphical Methods
			2.4.3.1 Stem-and-Leaf Diagram The stem-and-leaf (S&L) diagram is a graphical tool similar to the histogram, except it retains more information from the data. For a given set of data, the stems are somewhat equivalent to the cells of a histogram. The leave
			2.4.3.2 Box-and-Whisker Plot The box-and-whisker (B&W) plot is another compact way of representing a population with variability, and it is especially useful when comparing several distributions. The B&W plot is a graph showing several percentiles of a gi
	2.4.4 Other Numerical Measures
			2.4.4.1 Measures of Location The median, denoted by , is the middle value in the data. As stated before, it is the 50th percentile of the data. For data obtained from normally distributed populations, the median and the mean will be approximately the sam
	2.4.4.2 Measures of Dispersion The range, denoted by R, is the difference between the largest value and the smallest value in the data:
				Summary on Empirical Methods The empirical methods are based on observation, or data, gathered from population of interest in order to study where they are located and how they are distributed and be able to make predictions about their behavior. They do
	2.4.5 Exercises in Empirical Methods
	2.5 Mathematical Models for Describing Populations
	2.5.1 Probability
			2.5.1.1 Definition of Probability The term probability is always used with regards to the occurrence of an event. The probability of an event is a number between 0 and 1 that indicates the likelihood of occurrence of the event when the associated experime
	2.5.1.2 Computing the Probability of an Event There are two basic methods for computing the probability of events:
	2.5.1.2.1 Method of Analysis
	2.5.1.2.2 A Special Case If the sample space of an experiment consists of elements that are all equally likely, then:
	2.5.1.2.3 Method of Relative Frequency The method of relative frequency is used when the experimenter cannot assign weights to the outcomes in S from knowledge of the experiment. The probability has to be then determined through experimentation:
	2.5.1.3 Theorems on Probability
	2.5.1.3.1 Addition Theorem of Probability If A and B are any two events in a sample space, (i.e., A and B are two possible events when an experiment is performed), then the probability of A or B occurring:
	Corollary
	2.5.1.3.2 The Extension of the Addition Theorem When there are three events—A, B, and C, as shown in the sketch below—the addition theorem becomes:
	2.5.1.3.3 Complement Theorem of Probability Events A and Ac are said to be complement to each other if they are mutually exclusive and together make up the sample space (see the sketch below).
				2.5.1.3.4 Theorems on the Joint Occurrence of Events When we have to find the probability of the joint occurrence of two events—that is, the probability of A and B occurring together—we need to use one of the multiplication theorems given below, depending
				2.5.1.3.5 Conditional Probability Conditional probability can be described as follows: There are two events A and B that can occur when an experiment is performed. We are sometimes interested in knowing the probability of one event, say, A occurring given
	2.5.1.3.6 Independent Events Two events in a sample space are said to be independent if the occurrence of one does not affect the probability of occurrence of the other. Independence of events can be defined using conditional probabilities as follows:
	2.5.1.3.7 The Multiplication Theorems of Probability If A and B are any two events in a sample space, then:
				2.5.1.3.8 The Theorem of Total Probability Let B1, B2, …, Bk be partitions of a sample space S such that (B1 ∪ B2 ∪ … ∪ Bk) = S and (Bi ∩ Bj) = ∅ (null set) for any pair i and j (see diagram below). The partitions are mutually exclusive events that jointl
			2.5.1.4 Counting the Sample Points in a Sample Space The reader would have noticed, often, calculating the probability of an event involves first counting the number of sample points in the sample space and also counting the number of sample points in the
	2.5.1.4.1 The Multiplication Rule If an operation can be performed in n1 ways and another operation can be performed in n2 ways, then the two operations can be performed together in n1 × n2 ways.
				2.5.1.4.2 Permutations A permutation is an arrangement of all or part of a given set of objects. For example, the three objects a, b, and c can be permuted as follows, if all of them are taken together. So, there are six permutations of the three objects
	2.5.1.4.3 Theorem on Number of Permutations The number of permutations of n distinct objects taken r at a time denoted by nPr, is given by
				2.5.1.4.4 Combinations A combination is a group of a certain number of objects taken from a given collection of objects. Here, unlike in the permutations, no attention is paid to the arrangement or the relative location of the objects in the group. For ex
	2.5.1.4.5 Theorem on Number of Combinations The number of combinations of n distinct objects taken r at a time written as  (or written as nCr) is given by:
	Summary on Probability
	2.5.2 Exercises in Probability
	2.5.3 Probability Distributions
			2.5.3.1 Random Variable A random variable is a variable that assumes for its values the outcomes of a random experiment. A random variable takes only real numbers for its values. If an experiment produces outcomes that are not in numbers—that is, its outc
	2.5.3.2 Probability Mass Function If X is a discrete random variable, a function p(x) with the following properties is defined as the probability mass function (pmf ) of the random variable:
			2.5.3.3 Probability Density Function The method used above to describe a discrete random variable will not work in the case of a continuous random variable, because the continuous random variable takes an infinite number of possible values. The probabilit
	2.5.3.4 The Cumulative Distribution Function When we use mathematical models to describe populations, it is often convenient to have another function called the “cumulative distribution function” (CDF), denoted by F(x), defined as:
			2.5.3.5 The Mean and Variance of a Distribution A probability distribution function chosen appropriately for a random variable completely describes the behavior of the random variable. In the case of some distributions, however, the description can be acc
	2.5.4 Some Important Probability Distributions
			2.5.4.1 The Binomial Distribution The binomial distribution is the distribution of a random variable that represents the number of “successes” out of n independent trials when each trial can result in one of two possible outcomes—”success” or “failure.” T
	2.5.4.1.1 The Mean and Variance of a Binomial Variable If X ∼ Bi(n, p), it can be shown, using the definition for mean and variance, that:
			2.5.4.2 The Poisson Distribution The Poisson distribution has been found to be a good model to describe random variables that represent counts that can take values anywhere from zero to infinity, the following being a few examples of such random variables
	2.5.4.2.1 The Mean and Variance of the Poisson Distribution If X ~ Po(λ), then the mean of X can be shown to be
			2.5.4.3 The Normal Distribution The normal distribution has been found to be the model for many random variables found in the natural and man-made world. Many measurements such as amount of rainfall in a place, weight of newborn babies in a hospital, diam
				2.5.4.3.1 The Standard Normal Distribution A random variable that is normally distributed with μ = 0 and σ2 = 1 is called the “standard normal variable.” The standard normal variable is denoted by a unique label, Z. Its distribution is called the “standar
	2.5.4.3.2 Application of the Normal Distribution
			2.5.4.4 Distribution of the Sample Average  The sample average is a random variable. Suppose, for instance, that we take samples of size 16 repeatedly from a population and calculate the average from each sample, these averages will not all be the same.
			2.5.4.5 The Central Limit Theorem The previous theorem gave the distribution of  when the population is known to be normally distributed. What if the population is not normal or its distribution is not known? The central limit theorem (CLT) gives the ans
	Summary on Probability Distributions
	2.5.5 Exercises in Probability Distributions
	2.6 Inference of Population Quality from a Sample
	2.6.1 Definitions
	2.6.2 Confidence Intervals
	2.6.2.1 CI for the μ of a Normal Population When σ Is Known The estimator is . On the assumption that the population has N(μ, σ2), as stated earlier in the chapter, the  has the distribution N(μ, σ2/n), where n is the sample size. Then,
			2.6.2.2 Interpretation of CI We should note first that a conclusion has been drawn about the average of the entire day’s production from observations made on four sample observations. The CI is to be interpreted as follows: If 99% confidence intervals are
			2.6.2.3 CI for μ When σ Is Not Known In the above model of CI for μ, we assumed that the population standard deviation σ was known. Suppose the population standard deviation is not known; then, σ is replaced with S, the sample standard deviation, in the s
			2.6.2.4 CI for σ2 of a Normal Population When setting CI for the variance of the population σ2, the sample variance  is used as the estimator. The fact that the statistic (n −1)S2/σ2 has the χ2 (chi-squared) distribution with (n − 1) degrees of freedom i
	2.6.3 Hypothesis Testing
			Two Types of Errors A statistical test could result in any one of the four possible events shown in Figure 2.18. Of these four, two of the events lead to errors in conclusion. These errors are designated as Type I and Type II errors, as shown in Figure 2.
	2.6.3.1 Test Concerning the Mean µ of a Normal Population When σ Is Known The hypotheses are:
			2.6.3.2 Why Place the Claim Made about a Parameter in H1? We made a statement earlier that it is preferable to include the statement we want to verify (e.g., the average strength of nylon ropes >10 kg) in the alternate hypothesis H1. This is for the reaso
			2.6.3.3 The Three Possible Alternate Hypotheses We mentioned that the selection of the CR depends on the alternate hypothesis. There are only three possible scenarios that are encountered in practical testing and the applicable hypotheses for each are sho
			2.6.3.4 Test Concerning the Mean μ of a Normal Population When σ Is Not Known When σ is not known, we cannot use it in the test statistic and use the sample standard deviation S in its place. The new statistic with S replacing σ is known to have the t dis
			2.6.3.5 Test for Difference of Two Means When σs Are Known This test model is useful when two populations have to be compared with regards to their mean values. For example, the mean life of bulbs from one manufacturer may have to be compared with the mea
	2.6.4 Tests for Normality
			2.6.4.1 Use of the Normal Probability Plot Normal probability plotting involves plotting the cumulative percentage distribution of the data on specially designed normal probability paper (NPP). The NPP is designed such that if the data came from a normal
			2.6.4.2 Normal Probability Plot on the Computer Most statistical software packages provide a routine to make the normal probability plot for a given set of data. Figure 2.28 shows a normal probability plot made using the Minitab software. The computer sof
			2.6.4.3 A Goodness-of-Fit Test A goodness-of-fit test generally compares the actual empirical cumulative distribution of the data with that of the hypothesized distribution and obtains a quantity to signify how “far away” the empirical distribution is fro
	2.6.5 The P-Value
	Summary on Inference Methods
	2.6.6 Exercises in Inference Methods
	2.6.6.1 Confidence Intervals
	2.6.6.2 Hypothesis Testing
	2.6.6.3 Goodness-of-Fit Test
	2.7 Mini-Projects
	Mini-Project 2.1
	Mini-Project 2.2
	Mini-Project 2.3
	Mini-Project 2.4
	References
	Chapter 3: Quality in Design
	3.1 Planning for Quality
	3.1.1 The Product Creation Cycle
	3.2 Product Planning
	3.2.1 Finding Customer Needs
			3.2.1.1 Customer Survey A typical customer survey attempts to establish the customers’ needs and the level of importance that customers attach to the different needs. Many customers may also be lead users for a product or service category, and often expre
	3.2.2 Quality Function Deployment
			3.2.2.1 Customer Requirements and Design Features Next, the preference numbers reflecting the relative importance of the various customer preferences are determined. When a product has multiple customers—as in this case—a set of importance-weights represe
			3.2.2.2 Prioritizing Design Features For each design feature, the product of the numerical equivalent of the strength relationship and the importance-weight of the corresponding customer preference is obtained and added column-wise, and the total is place
			3.2.2.3 Choosing a Competitor as Benchmark Also shown on the right-hand side of the HOQ is the assessment on how well competitor products fare with respect to the established customer preferences. For this example, four books—A, B, C, and D—are identified
			3.2.2.4 Targets For the design features that have been prioritized as the most important, targets are selected for the new product based on a comparison with the benchmark. The benchmark is first evaluated by the planning team, and scored on a scale of 1
	3.2.3 Reliability Fundamentals
			3.2.3.1 Definition of Reliability Reliability is expressed as a function of time. (Reliability for a product at 1000 hrs. will be different from reliability at 5000 hrs.) The reliability of a product at time t, denoted as R(t), is defined as the probabili
			3.2.3.2 Hazard Function An important definition relating to reliability is that of the “hazard function,” which is also known as the “instantaneous failure rate” or “mortality rate” function. The hazard function, denoted by h(t), represents the rate at wh
			3.2.3.3 The Bathtub Curve Failure rate curves have been used to study the life behavior of many different types of equipment. Figures 3.5a to d show a few different types of failure rate curves experienced by different types of equipment. There are produc
			3.2.3.4 Distribution of Product Life The above discussion of the failure rate behavior of products and the phenomenon of the bathtub curve reminds us that the distribution of the life variable could change over the life of a product. At least three distin
	3.2.3.5 The Exponential Distribution If T represents a product life, the function form, or the pdf, of the exponential distribution is given by:
			3.2.3.6 Mean Time to Failure The mean of the distribution, or the average life of all units in the population, is called the “mean time to failure” (MTTF). The term MTTF is used for products that have only one life (i.e., those that are not repairable). F
	3.2.3.7 Reliability Engineering The discipline of applying reliability principles to evaluating, predicting, and enhancing reliability in products is called “reliability engineering.”
	3.3 Product Design
	3.3.1 Parameter Design
	3.3.2 Design of Experiments
			3.3.2.1 22 Factorial Design A factorial experiment is an experiment in which each level of one factor is combined with each level of every other factor to obtain all possible treatment-combinations at which trials are to be performed. The 22 design is a f
			3.3.2.2 Randomization Just as noise or extraneous factors can cause variability in replicates even if they are run with the same treatment combination, noise factors can also affect outcomes between two treatment combinations, either by masking or adding
			3.3.2.3 Experimental Results from a 22 Design Suppose the eight trials of the above experiment are run in a completely randomized manner and the results from the trials are as shown in Table 3.4. The results are presented in the graph in Figure 3.7, with
			3.3.2.4 Calculating the Factor Effects The data from the experiment under discussion are rearranged in Table 3.5 to facilitate the calculation of the effects. This table has a new column, with the heading “interaction,” added to those in Table 3.4. The or
			3.3.2.5 Main Effects Factors A and B are called the main factors in order to differentiate them from interactions that also arise as outcomes of experiments. The effect caused by a main factor, called a “main effect,” is calculated by subtracting the aver
			3.3.2.6 Interaction Effects Interaction between two factors exists if the effect of the two factors acting together is much more, or much less, than the sum of the effects caused by the individual factors acting alone. It is necessary to detect the existe
			3.3.2.7 A Shortcut for Calculating Effects The product of the column of treatment combination codes and the column of signs under any factor is called a “contrast” of that factor. For example, the product of the column of codes and column of signs under F
			3.3.2.8 Determining the Significance of Effects Next, we have to determine if the effects are significant. The analysis of variance (ANOVA) method is normally used to answer this question. The ANOVA method is explained in Chapter 5; here, we will use a qu
	3.3.2.9 The 23 Design The 23 design will be used when there are three factors affecting a response and each factor is studied at two levels. We will illustrate this design using another lawnmower example.
			3.3.2.10 Interpretation of the Results Looking at the confidence intervals, we see that all three main effects, A, B, and C, are significant because there is no 0 included in any of those intervals. The AB interaction is also significant. No other interac
			3.3.2.11 Model Building One of the important aspects of post-experiment data analysis is model building and exploring the treatment combination space for the optimal treatment combination that produces performances even better than those obtained in the e
			3.3.2.12 Taguchi Designs Taguchi designs, which have become very popular among engineers, are of the fractional factorial type, in which the designs are provided as orthogonal arrays, which are in fact the design columns, columns of signs for factors that
	3.3.3 Tolerance Design
			3.3.3.1 Traditional Approaches Traditionally, tolerancing has been done based on the experience of designers regarding what has worked for them satisfactorily. Experienced designers have documented the tolerances that have worked well for different manufa
			3.3.3.2 Tolerancing According to Dr. Taguchi Traditional tolerancing, which provides an upper and lower specification limit (USL and LSL, respectively) for a characteristic around a target, is based on the premise that the fitness of the product is good a
			3.3.3.3 Assembly Tolerances Many situations arise in industrial settings, where tolerance for an assembly has to be calculated from the tolerances of components. For example, the track for a bulldozer is made up of 50 links. The tolerances of the links ar
			3.3.3.4 The RSS Formula Suppose L1 and L2 are the lengths of two components (see Figure 3.12a) and L = L1 + L2. Let t1 and t2 be the tolerances on L1 and L2, respectively. Then the tolerance on L is given by . We need the assumption that L1 and L2 are in
			3.3.3.5 Natural Tolerance Limits “Specification limits” or “tolerance limits” usually refer to the limits of variability that a customer has imposed based on where and how the product is to be used. They might have been given by the customer, or chosen by
	3.3.4 Failure Mode and Effects Analysis
	3.3.5 Concurrent Engineering
			3.3.5.1 Design for Manufacturability/Assembly “[Design for manufacturability/­assembly] represents a new awareness of the importance of design as the first manufacturing step” (Syan and Swift 1994). This study is meant to make the design engineers conscio
			3.3.5.2 Design Reviews Design reviews are regularly scheduled meetings of a review team organized by design engineers. These meetings include representatives from manufacturing, quality, materials, suppliers, and customers, to review designs and monitor p
	3.4 Process Design
	3.4.1 The Process Flow Chart
	3.4.2 Process Parameter Selection: Experiments
	3.4.3 Floor Plan Layout
	3.4.4 Process FMEA
	3.4.5 Process Control Plan
	3.4.6 Other Process Plans
			3.4.6.1 Process Instructions Process instructions are documents describing how each operation in the process should be performed. Process instructions and control plans complement each other in defining how operations are to be performed. The process inst
			3.4.6.2 Packaging Standards Designing the packaging for the final product is an important activity in quality planning. The packaging design should ensure that the product performance will not be altered during handling and shipment to the customer. The q
			3.4.6.3 Preliminary Process Capabilities The meaning of process capability and the method of measuring process capabilities are explained in detail in Chapter 4. Process capability refers to the ability of a process to produce a desired product characteri
			3.4.6.4 Product and Process Validation This is the step before the product is launched into production, when the process is tried out to check if it is capable of producing the product according to design and in a manner to meet the customer’s needs. The
			3.4.6.5 Process Capability Results These are measured during the validation run and are compared with expected or targeted benchmarks. Changes may have to be made to the process if any shortfall occurs in the expected process capabilities or product quali
			3.4.6.6 Measurement System Analysis The requirements for measuring instruments in terms of accuracy, precision, resolution, and so on, and how the instruments are evaluated against these requirements, are described in Chapter 4. The objective of measuring
			3.4.6.7 Product/Process Approval Product/process approval refers to testing the product from the preproduction trial and verifying that the product meets the design specifications. Characteristics such as strength, chemistry, temperature rise, and fuel co
			3.4.6.8 Feedback, Assessment, and Corrective Action This is the stage when the effectiveness of all the quality planning work in previous stages is verified. The objective is to gather all the information together and assess the capabilities of the proces
	3.5 Exercise
	3.5.1 Practice Problems
	3.5.2 Mini-Projects
			Mini-Project 3.1 Create a customer survey instrument to find out from the group of IE (ME/EE/CE or any other) majors how satisfied they are with the services they receive in an IE (ME/EE/CE or any other) department in both academic and non-academic areas.
			Mini-Project 3.2 Based on the survey instrument created in Mini-Project 3.1, construct a HOQ, and choose the design features that you will use to enhance the services provided by the IE (ME/EE/CE or any other) department or the library. Assume some reason
	Mini-Project 3.3 This project is meant to verify the theoretical principles of the root-sum-of-squares formula used for obtaining assembly tolerance from parts tolerance. The theoretical results in the simplest form can be expressed as follows:
	References
	Chapter 4: Quality in Production—Process Control I
	4.1 Process Control
	4.2 The Control Charts
	4.2.1 Typical Control Chart
	4.2.2 Two Types of Data
	4.3 Measurement Control Charts
	4.3.1 - and R-Charts
	4.3.2 A Few Notes about the  and R-Charts
			4.3.2.1 The Many Uses of the Charts The - and R-chart combination is the most popular SPC method employed in industrial applications. Their simplicity and effectiveness in discovering significant assignable causes have made them very popular. The fact th
			4.3.2.2 Selecting the Variable for Charting Although control charts are very useful tools, they are expensive to maintain, especially if measurements have to be made manually. So, their use must be limited to where they are absolutely needed. Even if meas
			4.3.2.3 Preparing Instruments Before starting a control chart on any process, it is first necessary to decide the instrument to be used for measuring the variable and the level of accuracy at which the readings are to be recorded. It is also necessary to
			4.3.2.4 Preparing Check Sheets Proper check sheets or standard forms must be developed for recording relevant process information and for recording and analyzing data. Standard forms, such as that shown in Figure 4.4, that call for relevant data and provi
			4.3.2.5 False Alarm in the -Chart When a  value falls outside a control limit, we normally understand that an assignable cause has occurred and changed the process mean to a level different from the desired level. Just as with any statistical procedure,
			4.3.2.6 Determining Sample Size The typically recommended sample size for  and R-charts is four or five. Dr. Shewhart, when he first proposed the -chart, recommended the use of as small a sample as possible, because averaging over large samples would hi
			4.3.2.7 Why 3-Sigma Limits? Again, it was Dr. Shewhart who recommended use of the 3-sigma rule for calculating control limits when he originally proposed the control charts. He was particular that the chances for process interruption due to false alarms f
			4.3.2.8 Frequency of Sampling Again, economic studies can be made to determine how often the samples should be taken to optimize the control operation. Such studies try to balance the increased cost of sampling from increased frequency, with the benefits
			4.3.2.9 Rational Subgrouping In control chart parlance, the term “subgroup” and the term “sample” mean the same thing. Some prefer the former to the latter, however, because the former clearly implies there is more than one unit in it, which makes communi
			4.3.2.10 When the Sample Size Changes for - and R-Charts This refers to the situation in which samples of same size are not available consistently for charting purposes. Such a problem arises less often in the case of measurement control charts than in t
			4.3.2.11 Improving the Sensitivity of the -Chart In the next chapter, while discussing its operating characteristics, we will see that the -chart with 3-sigma limits and a sample size of four or five—referred to as the “conventional chart”—is capable of
			4.3.2.12 Increasing the Sample Size One way to increase the power of the - and R-charts is to use larger sample sizes. As will be shown in the next chapter, the power of the -chart increases with an increase in sample size, as does the power of the R-ch
			4.3.2.13 Use of Warning Limits Another way to improve the sensitivity of the -​chart is to use warning limits drawn at 1-sigma and 2-sigma distances on either side, between the centerline and the (3-sigma) control limits, as shown in Figure 4.8a. Rules a
			4.3.2.14 Use of Runs Yet another method to enhance the sensitivity of the -chart is to use runs. A run is a string of consecutive plots with some common properties. For example, if a sequence of consecutive  values occurs below the centerline, this sequ
			4.3.2.15 Patterns in Control Charts Besides runs, other telltale signs also appear in control charts. Some of these patterns are shown in Figure 4.9. The comments accompanying each pattern suggest how an analyst can interpret the patterns. The patterns in
			4.3.2.16 Control vs. Capability A process is in-control if it is operating consistently within its natural variability. A process is capable if it produces “all” the products within specification. When the process is in-control, it does not mean that the
	4.3.3  and S-Charts
	4.3.4 The Run Chart
	4.4 Attribute Control Charts
	4.4.1 The P-Chart
	4.4.2 The C-Chart
	4.4.3 Some Special Attribute Control Charts
			4.4.3.1 The P-Chart with Varying Sample Sizes The discussion of the P-chart earlier assumed that samples of size n could be drawn repeatedly from the process to be controlled and that the number of defective units in each of the equal-size samples could b
			4.4.3.2 The nP-Chart For the P-chart, the statistic P, proportion defectives in a sample, is calculated as P = D/n, where D is the number of defectives in a sample of size n. If P = D/n, then nP = D. Instead of plotting P, suppose we plot nP, which is sim
			4.4.3.3 The Percent Defective Chart (100P-Chart) If, instead of multiplying the limits of the P-chart by n, they are multiplied by 100, then the limits for percent defectives are obtained. For example, the charts with the following two sets of limits are
			4.4.3.4 The U-Chart The U-chart is a variation of the C-chart. The C-chart described previously can only be used if all the units inspected for the chart are identical—that is, if the opportunity for a defect is the same from unit to unit. Several situati
	4.4.4 A Few Notes about the Attribute Control Charts
			4.4.4.1 Meaning of the LCL on the P- or C-Chart The lower control limit does not have the same significance with the P-chart, or the C-chart, as it has with the -chart. If an  value falls below the LCL of an -chart, we will conclude that the process av
			4.4.4.2 P-Chart for Many Characteristics One of the advantages of the P-chart is that one chart can be used for several product characteristics. For example, if we are inspecting bolts, we can use one chart for several characteristics such as diameter, le
			4.4.4.3 Use of Runs The rules pertaining to runs above or below the CL, and to runs up or runs down, can also be used with the P-chart and the C-chart. These rules are especially useful when the average of P, or C, is decreasing and there is no LCL (i.e.,
			4.4.4.4 Rational Subgrouping As in - and R-charts, proper subgrouping is key to getting the most out of the P-chart and the C-chart. The subgrouping must be done so as to provide leads to discovering assignable causes when they are present. The following
	4.5 Summary on Control Charts
	4.5.1 Implementing SPC on Processes
	4.6 Process Capability
	4.6.1 Capability of a Process with Measurable Output
	4.6.2 Capability Indices Cp and Cpk
	4.6.3 Capability of a Process with Attribute Output
	4.7 Measurement System Analysis
	4.7.1 Properties of Instruments
	4.7.2 Measurement Standards
	4.7.3 Evaluating an Instrument
	4.7.3.1 Properties of a Good Instrument A good instrument should have:
			4.7.3.2 Evaluation Methods The methods (or testing procedures) for assessing the properties of instruments and the standards for their acceptability are described below. Many of the recommendations made here on instrument capability are those given by the
			4.7.3.3 Resolution As stated earlier, the resolution is the smallest division of a measurement unit that an instrument is capable of reading or capable of distinguishing. This is also known as “discrimination.” Good resolution is necessary to discover cha
			4.7.3.4 Bias The bias of an instrument is measured by taking repeated measurements on a product for which the “true” value is known. The true value can be obtained by measuring the product using a “tool-room” instrument (i.e., an instrument known to have
	4.7.3.5 Variability (Precision) Instrument variability, or error, could come from two sources:
			4.7.3.6 A Quick Check of Instrument Adequacy A quick way of comparing the variability in the instrument with the variability in the product, and thus of assessing the adequacy of the instrument, is as follows. Take repeated measurements using the instrume
	4.8 Exercise
	4.8.1 Practice Problems
	4.8.2 Mini-Projects
			Mini-Project 4.1 The table below shows data on the breaking strength of water-jacket cores used in making cylinder-head castings in a foundry. The jacket core is in three sections, identified as the B, C, and R sections. In other words, the three sections
			Mini-Project 4.2 The data below represent the proportion of rejected castings for various defects on a particular production line. Make a P-chart using the data on the daily proportion of defectives. Draw any useful information that can be gleaned from th
			Mini-Project 4.3 I have a suspicion that the speedometer in my 1984 VW van is not showing the correct speed of the vehicle, so I decided to evaluate it using the speedometer reading given on a GPS monitor. The following table shows the readings recorded f
	References
	Chapter 5: Quality in Production—Process Control II
	5.1 Derivation of Limits
	5.1.1 Limits for the -Chart
	5.1.2 Limits for the R-Chart
	5.1.3 Limits for the P-Chart
	5.1.4 Limits for the C-Chart
	5.2 Operating Characteristics of Control Charts
	5.2.1 Operating Characteristics of an -Chart
			5.2.1.1 Computing the OC Curve of an -Chart Figure 5.4 shows a process that has a mean equal to μ being moved by an assignable cause to a new location μ + kσ, where σ is the standard deviation of the process, which we assume remains unchanged. The proces
	5.2.2 OC Curve of an R-Chart
	5.2.3 Average Run Length
	5.2.4 OC Curve of a P-Chart
	5.2.5 OC Curve of a C-Chart
	5.3 Measurement Control Charts for Special Situations
	5.3.1 - and R-Charts When Standards for μ and/or σ Are Given
			5.3.1.1 Case I: μ Given, σ Not Given This case arises when the process mean should be controlled at a given standard average level but the variability at which the process must be controlled is not given. In this case, data will be collected as for the re
			5.3.1.2 Case II: μ and σ Given In this case, we do not use process data either to estimate the process mean or the process standard deviation. The formulas are calculated out of the given standards and reflect the limits of variability that we should expe
	5.3.2 Control Charts for Slow Processes
			5.3.2.1 Control Chart for Individuals (  X-Chart) The control chart for individuals, or the X-chart, is normally used along with a chart for successive differences, which is known as a moving range chart, or MR chart, with subgroup size n = 2. The MR char
			5.3.2.2 Moving Average and Moving Range Charts The Moving Average (MA) and Moving Range (MR) charts use sample averages and sample ranges, as do the regular - and R-charts, but the method of forming the samples or subgroups is different. Suppose the samp
	5.3.2.3 Notes on Moving Average and Moving Range Charts
				What Is a Good Value for n? The averaging done in the MA and MR charts reduces the noise (the variability) and helps in discovering the signals. Larger subgroup sizes tend to smooth out the variations in individual observations and bring out the signals b
				A Caution: While reacting to plots outside the control limits on the MA and MR charts, caution must be exercised when interpreting the charts. Suppose that an adjustment is made to a process because of a moving average falling outside a limit; the next co
	5.3.3 The Exponentially Weighted Moving Average Chart
	5.3.3.1 Limits for the EWMA Chart If wi = λxi + (1−λ)wi−1, it can then be shown (Lucas and Saccucci 1990) that
	5.3.4 Control Charts for Short Runs
	5.3.4.1 The DNOM Chart For the DNOM chart, the statistic plotted for the i-th sample is the deviation of the sample average from the nominal, or target:
			5.3.4.2 The Standardized DNOM Chart In the above DNOM chart, an assumption was made that the standard deviations of the measurements in each subgroup (i.e., in each part number), were the same and equal to σ. This assumption, however, may not always be tr
	5.4 Topics in Process Capability
	5.4.1 The Cpm Index
	5.4.2 Comparison of Cp , Cpk , and Cpm
	5.4.3 Confidence Interval for Capability Indices
	5.4.4 Motorola’s 6σ Capability
	5.5 Topics in the Design of Experiments
	5.5.1 Analysis of Variance
	5.5.2 The General 2k Design
	5.5.3 The 24 Design
	5.5.4 2k Design with Single Trial
	5.5.5 Fractional Factorials: One-Half Fractions
			5.5.5.1 Generating the One-Half Fraction With reference to Table 5.17a, when we have three factors and want a design with only four runs, we first write down the design for a 22 factorial design (which has four runs) using the (−) and (+) signs under A an
			5.5.5.2 Calculating the Effects To calculate the effects, we must first complete the calculation columns of the above tables showing the columns of signs for the other interaction terms. The effects of the factors and interactions can be then calculated a
	5.5.6 Resolution of a Design
	5.6 Exercise
	5.6.1 Practice Problems
	5.6.2 Mini-Projects
			Mini-Project 5.1 A printing shop that produces advertising specialties produces paper cubes of various sizes, of which the 3.5 in. cube is the most popular. The cubes are cut from a stack of paper on cutting presses. The two sides of the cube are determin
			Mini-Project 5.2 The data in Columns 2 and 5 of Table 5.8, being the process data from a normal distribution, were generated using the Minitab random number generator. The first 20 observations have μ = 10 and σ = 1.0, and the second set of 20 observation
	References
	Chapter 6: Managing for Quality
	6.1 Managing Human Resources
	6.1.1 Importance of Human Resources
	6.1.2 Organizations
			6.1.2.1 Organization Structures An organization is a collection of hard entities (e.g., facilities, machineries) and soft entities (e.g., methods, people, values) that are gathered together to accomplish a certain goal. Oakes and Westcott (2001) define an
			6.1.2.2 Organizational Culture The culture of an organization can be seen in how the members of the organization make decisions, how they treat one another within the organization, and how they treat their contacts outside the organization. The culture of
	6.1.3 Quality Leadership
			6.1.3.1 Characteristics of a Good Leader Good leaders have vision: they think and act for the future, not just for the present. They can envision what their organization should be like in order to meet the needs of the customer. They see the big picture,
	6.1.4 Customer Focus
	6.1.5 Open Communications
	6.1.6 Empowerment
	6.1.7 Education and Training
	6.1.7.1 Need for Training
				Necessity of Basic Skills A quality organization needs a quality workforce. The workforce must have capabilities in the basic skills of reading, writing, and arithmetic. In addition, they should be capable of logical thinking, and of analyzing and solving
				Global Competition In the twenty-first century marketplace, when competition comes from almost every part of the world, the edge over competitors comes from the knowledge and creativity of the workforce, for which education is a prerequisite. In global co
				Continuous Change in Technology Things no longer remain static in the workplace. What is new this year becomes obsolete the next year. This happens not only in the fast-changing communication and computer areas, but also throughout the industrial spectrum
				Diversity in the Workplace The new workforce dynamics in the United States will draw an unusual mix of nontraditional workers, including homemakers returning to work after raising families, minorities, and immigrants, to the workplace. This generates the
				Need to Improve Continuously Dr. Deming claimed that: “Competent men in every position, if they are doing their best, know all that there is to know about their work except how to improve it” (Deming 1986). That is a succinct statement of the fact that th
	6.1.7.2 Benefits from Training The benefits that come from a trained workforce are many. Some of the obvious benefits from a trained workforce are:
			6.1.7.3 Planning for Training Different organizations may have different needs for training. A manufacturer of heavy machinery may need a different training scheme from that needed by a software developer. Each has to assess their particular training need
			6.1.7.4 Training Methodology Training is best done in modules, each no more than two hours in length. The two-hour segments seem to be the most suitable to fit into the work schedules of managers and executives. The two-hour modules also provide a good br
			6.1.7.5 Finding Resources Training programs will usually be budgeted by the training department, with the quality department providing technical consultation in the development of syllabuses and in choosing instructors. Volunteers from within an organizat
			6.1.7.6 Evaluating Training Effectiveness The test of training is in the learning that the participants acquire. This learning usually becomes apparent in the results of the projects they complete or in the output of the teams in which they participate. I
	6.1.8 Teamwork
	6.1.8.1 Team Building Team building does not occur spontaneously; a certain effort is needed to create good teams, and there is a process for team building. The process includes the following steps.
			6.1.8.2 Selecting Team Members Members who have the most potential for contributing to the mission of a team based on their expertise, experience, and attitude to teamwork should be included. There must be diversity in all respects—education, salary grade
			6.1.8.3 Defining the Team Mission A mission statement, written by the team, describes the purpose of the team and is communicated to all in the organization. It should be broad enough to include all that is to be accomplished and should have just enough d
	6.1.8.4 Taking Stock of the Team’s Strength A team’s strength should be assessed initially, and at regular intervals, based on team members’ own perceptions. Strength is assessed in the following areas:
	6.1.8.5 Building the Team Team-building activities must be planned and implemented based on the results of the assessment made in the previous step:
	6.1.8.6 Basic Training for Quality Teams All team members should be familiar with the general problem-solving process and the basic tools for quality improvement. The problem-solving process includes the following steps:
	6.1.8.7 Desirable Characteristics among Team Members
				Trust Trust is a very important characteristic among team members because it can transform a group of people into a team. People will not share information openly and assume responsibilities unless they can trust their colleagues to keep their word. Trust
				Selflessness Selflessness is the characteristic in individuals that makes them subordinate their individual interests to the interest of the team and the organization. In societies, where individuality and personal success are encouraged and celebrated, c
				Responsibility Responsibility is the characteristic that distinguishes members who complete the tasks assigned to them on time with 100% satisfaction from those who offer excuses for not completing their assigned tasks. Responsible members will also offer
				Enthusiasm Enthusiastic individuals are the cheerleaders in a team. They are usually high-energy people with high productivity. They help the team to overcome roadblocks, and they get the team going when the going gets tough. A few such members are always
	Initiative People with initiative do not wait for tasks to be assigned to them; they offer to take up tasks where they can make contributions. They provide the starting momentum for the team to get moving.
				Resourcefulness Resourceful people are the ones with the ability to find creative ways of resolving difficult issues. They find ways to make the best use of available resources. They find ways to get to the destination when others feel they have reached a
				Tolerance There will be differences among team members based on their educational level, gender, age, or race. There will be occasions when people will think or act differently because of the differences arising from cultural, intellectual, or educational
				Perseverance Patience and perseverance are the means of success in any endeavor. There will be team members who are bright, creative, and enthusiastic, but who may get easily disappointed and depressed when the first failure occurs. This is when people wi
			6.1.8.8 Why a Team? We can easily see from the above discussion of the characteristics needed in team members why a team can succeed where individuals cannot. It is hard to find in one person all of the qualities that team members can bring to bear collec
	6.1.8.9 Ground Rules for Running a Team Meeting The following set of rules for conducting team meetings is summarized from The Team Handbook (Scholtes 1988) in which further elaboration of these rules can be found:
	6.1.8.10 Making the Teams Work A few suggestions are available from experts on how to avoid problems that may develop during the working of a team and, if problems do arise, how to minimize their effects and make progress toward team goals.
				Making Team Members Know One Another Introduce members of a team who have not had prior working relationships through informal introductory chats about their jobs, families, hobbies, and so on. Begin each meeting, especially the early ones, with warm-up e
				Resolving Conflicts Promptly Well-defined directions for the team and ground rules for the conduct of business, communicated clearly, will prevent serious conflicts. Disagreements among members are not bad in themselves; in fact, such disagreements may ev
	Setting an Example by the Organization The organization as a whole should set a good example as a team player by working together with the suppliers and customers, and by being a fair, sensitive, and responsible partner in the community.
				Rewarding Good Teams Monetary as well as nonmonetary rewards can be utilized. A wage system that rewards only individual performance may not promote teamwork. A compensation system based on three components—an individual base compensation, an individual i
	6.1.8.11 Different Types of Teams Teams acquire their names mainly based on the purpose for which they are constituted. Sometimes, the name reflects the constituents making up the team.
	Process Improvement Teams Process improvement teams are the most relevant type related to quality in an organization. They are created in order to address quality improvement and customer satisfaction.
				Cross-Functional Teams Cross-functional teams may be process improvement teams, product design teams, or teams for the improvement of safety in the workplace. These teams are created when there is a need for receiving input from various functions of the o
				Self-Managed Teams Self-managed teams consist of all members of a full department or division of an organization without a department or division head. They manage themselves in that, as a team, they have full responsibility for budgeting, staffing, procu
			6.1.8.12 Quality Circles Quality circles are teams of employees mainly working in one area and reporting to the same supervisor, who have come together voluntarily to solve problems relating to the quality of a product or service created in that departmen
	6.1.9 Motivation Methods
	6.1.10 Principles of Management
	6.2 Strategic Planning for Quality
	6.2.1 History of Planning
	6.2.2 Making the Strategic Plan
	6.2.3 Strategic Plan Deployment
	6.3 Exercise
	6.3.1 Practice Problems
	6.3.2 Mini-Project
			Mini-Project 6.1 Prepare an essay on any one of the topics, such as empowerment, motivation, strategic planning, and so on, which are covered in this chapter. There are many more references that relate to these topics that are not reviewed here. Limit the
	References
	Chapter 7: Quality in Procurement
	7.1 Importance of Quality in Supplies
	7.2 Establishing a Good Supplier Relationship
	7.2.1 Essentials of a Good Supplier Relationship
	7.3 Choosing and Certifying Suppliers
	7.3.1 Single vs. Multiple Suppliers
	7.3.2 Choosing a Supplier
	7.3.3 Certifying a Supplier
	7.4 Specifying the Supplies Completely
	7.5 Auditing the Supplier
	7.6 Supply Chain Optimization
	7.6.1 The Trilogy of Supplier Relationship
	7.6.2 Planning
	7.6.3 Control
	7.6.4 Improvement
	7.7 Using Statistical Sampling for Acceptance
	7.7.1 The Need for Sampling Inspection
	7.7.2 Single Sampling Plans for Attributes
			7.7.2.1 The Operating Characteristic Curve When we are faced with choosing a single sampling plan for a given situation, we have numerous alternative plans to choose from. For example, any of the combination of numbers (10, 0), (12, 0), or (24, 2) describ
			7.7.2.2 Calculating the OC Curve of a Single Sampling Plan Suppose that a lot with p fraction defectives is submitted to a single sampling plan with sample size n and acceptance number c. The probability of acceptance of this lot with p fraction defective
			7.7.2.3 Designing an SSP From the above discussion, we see how the OC curve for a sampling plan is computed and how it shows the discriminating ability of a sampling plan. Next, we will see how to select an SSP for a given OC curve. Often, a purchaser and
	7.7.2.4 Choosing a Suitable OC Curve An OC curve is specified by selecting two points that lie on it. A few definitions are needed:
	7.7.2.5 Choosing a Single Sampling Plan Let us denote AQL by p1 and LTPD by p2 (see Figure 7.5c). The problem is to determine the values of n and c such that:
	7.7.3 Double Sampling Plans for Attributes
			7.7.3.1 Why Use a DSP? The DSP will require, on average, a smaller amount of inspection than a comparable SSP, comparable in the sense that both have about “equal” OC curves. We will later quantify the amount of inspection needed by a sampling plan using
			7.7.3.2 The OC Curve of a DSP In the case of the DSP, there is more than one OC curve. In fact, a DSP has a primary OC curve and two secondary OC curves. The primary OC curve shows the relationship between the quality of a submitted lot and the probabilit
	7.7.4 The Average Sample Number of a Sampling Plan
	7.7.5 MIL-STD-105E (ANSI Z1.4)
	7.7.5.1 Selecting a Sampling Plan from MIL-STD-105E A sampling plan from the military standard is selected through the following steps:
	7.7.6 Average Outgoing Quality Limit
	7.7.7 Some Notes about Sampling Plans
			7.7.7.1 What Is a Good AQL? Most companies would settle for an AQL value for a product characteristic based on what has worked for them, both functionally and economically, for that product characteristic. Smaller AQL values will result in large sample si
			7.7.7.2 Available Choices for AQL Values in the MIL-STD-105E First, we want to remember that the AQL values shown at the head of the columns of the MIL-STD tables are in percentages. Some of them (AQL ≤ 10%) are in percent defectives, or number of defecti
			7.7.7.3 A Common Misconception about Sampling Plans The AQL is used as an index for indexing sampling plans in the MIL-STD-105E. Suppose that a sampling plan chosen based on an AQL of 1.5% is used at an incoming inspection station. This does not mean that
			7.7.7.4 Sampling Plans vs. Control Charts As mentioned in Chapter 4, the control charts are preventive tools that are used during production by the producer. Sampling plans are acceptance tools that are used by the customer before accepting the product fo
			7.7.7.5 Variable Sampling Plans The discussion in this chapter has been limited to attribute sampling plans only. Although they may not be the most efficient plans in terms of the number of units to be inspected for a given lot, they are simple to use. Wh
	7.8 Exercise
	References
	Chapter 8: Continuous Improvement of Quality
	8.1 The Need for Continuous Improvement
	8.2 The Problem-Solving Methodology
	8.2.1 Deming’s PDCA Cycle
	8.2.2 Juran’s Breakthrough Sequence
	8.2.3 The Generic Problem-Solving Methodology
	8.3 Quality Improvement Tools
	8.3.1 Cause-and-Effect Diagram
	8.3.2 Brainstorming
	8.3.3 Benchmarking
	8.3.4 Pareto Analysis
	8.3.5 Histogram
	8.3.6 Control Charts
	8.3.7 Scatter Plots
	8.3.8 Regression Analysis
			8.3.8.1 Simple Linear Regression In simple linear regression, we try to establish the relationship between the response Y and an independent variable X by fitting a straight line to define that relationship. We hypothesize that the relationship is defined
			8.3.8.2 Model Adequacy Model adequacy is measured using a quantity called the coefficient of determination, which is denoted by R2. The quantity R2 represents the proportion of total variability in the observations of Y that is explained by the regression
			8.3.8.3 Test of Significance The coefficients a and b as obtained from the formulas above, are the intercept and slope, respectively, of the fitted straight line representing the relationship between X and Y. A question to be answered now: Does a straight
	8.3.8.4 Multiple Linear Regression When there is reason to believe that a dependent variable Y is influenced by several independent variables, say, X1, X2, and X3, then we use the multiple linear regression. The regression model will be
			8.3.8.5 Nonlinear Regression When curvature is suspected in the relationship between two variables as disclosed in a scatter plot, curvilinear models can be fitted to the data. For example, the following model may be appropriate for a simple nonlinear reg
	8.3.9 Correlation Analysis
			8.3.9.1 Significance in Correlation For normally distributed variables, Table 8.4 gives the 95% critical values of r for selected sample sizes. If the absolute value of the calculated r from sample observations exceeds the quantity in the table for a give
	8.4 Lean Manufacturing
	8.4.1 Quality Control
	8.4.2 Quantity Control
	8.4.3 Waste and Cost Control
	8.4.4 Total Productive Maintenance
	8.4.5 Stable, Standardized Processes
	8.4.6 Visual Management
	8.4.7 Leveling and Balancing
	8.4.8 The Lean Culture
	8.5 Exercise
	8.5.1 Practice Problems
	8.5.2 Term Project
	References
	Chapter 9: A System for Quality
	9.1 The Systems Approach
	9.2 Dr. Deming’s System
	9.2.1 Long-Term Planning
			Point 1: Create Constancy of Purpose for Improvement of Product and Service This point relates to having a strategic, long-term vision for an organization regarding growth in quality and productivity, to become competitive, to stay in business, and to pro
	9.2.2 Cultural Change
			Point 2: Adopt the New Philosophy According to Dr. Deming, the old management philosophy practiced by the U.S. industry (during the 1960s and 1970s) allowed workers on jobs that they did not know how to perform, employed supervisors who neither knew the j
	9.2.3 Prevention Orientation
			Point 3: Cease Dependence on Mass Inspection Quality cannot be achieved through inspection. It must be built into the product through the use of the right material and the right processes by trained operators. Dr. Deming claimed that: “Quality comes not f
	9.2.4 Quality in Procurement
			Point 4: End the Practice of Awarding Business on the Basis of Price Tag Alone Dr. Deming placed great importance in buying quality material in order to produce quality products. Buying from the lowest bidder, with no regard for quality, is detrimental to
	9.2.5 Continuous Improvement
			Point 5: Continuously Improve the System of Production and Service Quality starts at the design stage, with a good “understanding of the customer’s needs and of the way he uses and misuses a product.” This understanding must be continuously updated, and a
	9.2.6 Training, Education, Empowerment, and Teamwork
			Point 6: Institute Training Everyone in an organization needs training on how to do his or her job. Managers should learn about the jobs on the production floor, in distribution, in process maintenance, in accounting, and so on. Supervisors should learn t
			Point 7: Adopt and Institute Leadership Managers should become leaders or coaches who facilitate and help their teams in performing their jobs. Management by objective (MBO), which is based solely on results or outcomes, should be replaced with leadership
			Point 8: Drive Out Fear Fear among employees prevents them from reporting problems in product design or problems arising from process deterioration that can cause poor quality products, which can then get shipped out to customers. Fear of losing one’s job
			Point 9: Break Down Barriers between Staff The work involved in achieving product quality is done at many places in an organization, and by many different people. They all have information that needs to be shared. For example, marketing people have inform
			Point 10: Eliminate Slogans, Exhortations, and Targets for the Workforce Posters and slogans on walls, mainly addressed to workers, do not have any positive results. In fact, they have negative effects, such as creating mistrust, frustration, and demorali
			Point 11(a): Eliminate Numerical Quotas for the Workforce According to Dr. Deming, a numerical quota or work standard—that is, requiring so many pieces to be produced per day—“is a fortress against improvement of quality and productivity.” Standards are o
			Point 11(b): Eliminate Numerical Goals for People in Management Numerical goals set for managers, such as decrease cost of warranty by 50% next year, with no plans as to how to accomplish it “is just a farce.” Those goals will never be accomplished. A man
			Point 12: Remove Barriers that Rob People of Pride of Workmanship According to Dr. Deming, every worker wants to do a good job and be proud of it. If defectives are produced and waste-generated, it is because the management does not provide the opportunit
			Point 13: Encourage Education and Self-Improvement for Everyone Dr. Deming claimed, “There is no shortage of good people, shortage exists at the high levels of knowledge, and this is true in every field. People must be continually educated, not for the sh
	Point 14: Take Action to Accomplish the Transformation Dr. Deming laid down an action plan for initiating and accomplishing quality in an organization:
	9.3 Dr. Juran’s System
	9.3.1 Quality Planning
	9.3.2 Quality Control
	9.3.3 Quality Improvement
	9.4 Dr. Feigenbaum’s System
	9.5 Baldrige Award Criteria
	9.5.1 Criterion 1: Leadership
	9.5.1.1 Senior Leadership
	9.5.1.2 Governance and Societal Responsibilities
	9.5.2 Criterion 2: Strategic Planning
	9.5.2.1 Strategy Development
	9.5.2.2 Strategy Implementation
	9.5.3 Criterion 3: Customers
	9.5.3.1 Voice of the Customer
	9.5.3.2 Customer Engagement
	9.5.4 Criterion 4: Measurement, Analysis, and Knowledge Management
	9.5.4.1 Measurement, Analysis, and Improvement of Organizational Performance
	9.5.4.2 Information and Knowledge Management
	9.5.5 Criterion 5: Workforce
	9.5.5.1 Workforce Environment
	9.5.5.2 Workforce Engagement
	9.5.6 Criterion 6: Operations
	9.5.6.1 Work Processes
	9.5.6.2 Operational Effectiveness
	9.5.7 Criterion 7: Results
	9.5.7.1 Product and Process Results
	9.5.7.2 Customer Results
	9.5.7.3 Workforce Results
	9.5.7.4 Leadership and Governance Results
	9.5.7.5 Financial and Market Outcomes
	9.6 ISO 9000 Quality Management Systems
	9.6.1 The ISO 9000 Standards
	9.6.2 The Seven Quality Management Principles
			QMP 1 Customer Focus Organizations should know who their customers are and understand the current and future needs of those customers and strive to meet and exceed the customers’ needs and expectations. Align organization’s objectives with the needs of th
			QMP 2 Leadership Leaders should establish the mission, vision, and strategy to achieve them, and communicate them throughout the organization. They should create a set of guiding values of fairness, ethical behavior, trust, and integrity in the organizati
			QMP 3 Engagement of People People at all levels should be involved in the pursuit of the chosen quality goals so that all their abilities are fully utilized for the benefit of the organization. This needs communication to people on the importance of their
			QMP 4 Process Approach A productive organization is a collection of interrelated and interacting processes, with each process transforming some inputs into outputs through the use of some resources. Figure 9.5 has been drawn to show that an enterprise is
			QMP 5 Improvement Organizations should continually look for opportunities to improve their processes in order to improve customer satisfaction and efficiency of internal processes. They should establish improvement objectives and train people at all level
			QMP 6 Evidence-Based Decision Making Organizations should encourage decision making based on evidence from data, or information gathered from processes, rather than on the feelings and beliefs of people. They should measure and monitor key indicators of p
			QMP 7 Relationship Management Organizations should enter into interdependent, mutually beneficial relationships with interested parties in order to enhance the abilities and create value for all. They should first identify their interested parties—supplie
	9.7 ISO 9001:2015 Quality Management Systems—Requirements
	9.7.1 Scope
	9.7.2 Normative Reference
	9.7.3 Terms and Definitions
	9.7.4 Context of the Organization
			9.7.4.1 Understanding the Organization and its Context This requirement says that the organization should be aware of its mission and vision and should know its strengths and weaknesses, internal and external, which have influence on achieving its vision
			9.7.4.2 Understanding the Needs and Expectation of Interested Parties This requirement says the organization should understand who the “interested parties” are to the QMS. The term “interested parties” includes the customers, supplies, investors, employee
			9.7.4.3 Determining the Scope of the Quality Management System The organization should determine the scope of the QMS and make it a part of the record defining the products covered and the boundaries and applicability of this International Standard. The s
			9.7.4.4 Quality Management System and its Processes This is the requirement where the standard stipulates that the organization shall establish, implement, maintain, and continually improve a QMS in accordance with this International Standard. It goes int
	9.7.5 Leadership
			9.7.5.1 Leadership and Commitment The top management shall demonstrate leadership and commitment to the QMS by taking accountability for effective performance of the system, formulating quality policy and objectives compatible within the context of the or
			9.7.5.2 Policy The top management shall establish, implement, and maintain the quality policy in accordance with their needs. The quality policy shall be maintained as a record and be communicated, understood, and applied. It should also be available to r
			9.7.5.3 Organizational Roles, Responsibilities, and Authorities The top management shall make sure that responsibilities and authorities for the relevant roles in the QMS are assigned, communicated, and understood within the organization, for making sure
	9.7.6. Planning
			9.7.6.1 Actions to Address Risks and Opportunities While planning a QMS taking into account the needs of its interested parties and its own strengths and weaknesses, the organization shall consider the risks and opportunities that need to be addressed to
			9.7.6.2 Quality Objectives and Planning to Achieve Them The organization shall establish quality objectives at relevant functions, levels, and processes as needed for the QMS. The objectives will be measurable, relevant to meeting customer requirements, a
			9.7.6.3 Planning of Changes If there is need to make changes to the QMS, it shall be done in a planned manner considering the purpose of the changes and their potential consequences and availability of resources, so that the integrity of the QMS remains i
	9.7.7. Support
	9.7.7.1 Resources The organization shall determine and provide the resources needed for establishing, implementing, maintaining, and continuous improvement of the QMS. This includes:
			9.7.7.2 Competence The organization shall determine competence of persons operating the system and make sure they have the necessary competencies to manage the system perform effectively. The persons should have appropriate education, training, or experie
			9.7.7.3 Awareness The organization shall make sure that the people working in the organization are aware of the quality policy, relevant quality objectives, the individual’s contribution to the effectiveness of the system, and the possible results arising
	9.7.7.4 Communication The organization shall determine the internal and external communications relative to the QMS, including what, when, how, with whom, and by whom, to communicate.
			9.7.7.5 Documented Information The organization shall maintain documented information (record) as required by this International Standard and the documentation deemed necessary by the organization for effective performance of the QMS, depending on the siz
	9.7.8. Operation
			9.7.8.1 Operation Planning and Control The organization shall plan, implement, and control the processes needed for providing the products. This includes determining requirements of the processes as well as the criteria for the process performance and pro
			9.7.8.2 Requirements for Products and Services The organization shall communicate with the customer for providing/receiving information relating to products and services, handling enquiries, concluding contracts, and orders. The organization shall determi
			9.7.8.3 Design and Development of Products and Services The organization shall establish, implement, and maintain a design and development process that is appropriate for the subsequent making and delivery of the products. The process will take into accou
			9.7.8.4 Control of Externally Provided Processes, Products, and Services The organization shall ensure that externally provided processes, products, and services conform to requirements, the externally provided processes remain within the control of its Q
			9.7.8.5 Production and Service Provision The organization shall implement production of products under controlled conditions. They shall make sure they have records available defining the characteristics of the products to be produced and the results to b
			9.7.8.6 Release of Products and Services The organization shall have arrangements at appropriate stages to verify that the product meet the requirements. The product release to the customer shall not proceed unless such verification is complete, unless ot
			9.7.8.7 Control of Nonconforming Outputs The organization shall ensure that outputs that do not conform to the requirements are identified and prevented from being delivered or used unintentionally. Nonconforming products can be either corrected, segregat
	9.7.9. Performance Evaluation
			9.7.9.1 Monitoring, Measurement Analysis, and Evaluation The organization shall evaluate the performance and effectiveness of the QMS and keep appropriate documentation of the outcomes. The organization shall determine what to monitor, using what measurem
			9.7.9.2 Internal Audit The organization shall conduct internal audits at scheduled intervals to verify if the QMS conforms to this International Standard as well as conforms to the organization’s own requirements, and if the system is effectively implemen
	9.7.9.3 Management Review The management shall review the QMS at suitable intervals for its adequacy, effectiveness, and suitability to the organization’s planned direction.
	9.7.10. Improvement
			9.7.10.1 General The organization shall Identify and select opportunities for improvement and make changes to meet customer requirements and enhance customer satisfaction. They shall include opportunities to improve products to meet customer requirements
			9.7.10.2 Nonconformity and Corrective Action When a nonconformity occurs, the organization shall take action to control and correct it and consider actions to eliminate the causes for the nonconformity, so that the nonconformity does not occur again or oc
			9.7.10.3 Continual Improvement The organization shall continually improve the QMS to improve its suitability, adequacy, and effectiveness using the results of analysis and evaluation from internal audits, process performance measures, customer satisfactio
	9.8 The Six Sigma System
	9.8.1 Six Themes of Six Sigma
			Theme 1: Focus on the Customer The Six Sigma process begins with the measurement of customer satisfaction on a dynamic basis, and the Six Sigma improvements are evaluated based on how they impact the customer. Customer requirements are assessed first, and
			Theme 2: Data and Fact-Driven Management The Six Sigma philosophy emphasizes the need for taking measurements on process performance, product performance, customer satisfaction, and so on. The Six Sigma process does not allow decisions based on opinions,
			Theme 3: Process Focus Every product or service is created or produced by a process, whether it is designing a product, preparing an invoice, answering a customer complaint, or solving a problem, the activities can be mapped as a process involving a seque
			Theme 4: Proactive Management Proactive management involves setting ambitious goals and clear priorities, reviewing them frequently, and implementing changes in the system to prevent errors and defects from reaching the customer. Waiting for customer comp
			Theme 5: Boundaryless Collaboration Lack of good communication among the functions in the design or production stage results in delays, redesigns, and rework—all of which cause wasted resources. Teamwork across the functions, toward the common goal of cus
			Theme 6: Drive for Perfection (with Tolerance for Failure) The most important theme of the Six Sigma process is driving toward near-perfection—that is, not more than 3.4 defects per million opportunities (DPMO) in every process. This is achieved through d
	9.8.2 The 6σ Measure
	9.8.3 The Three Strategies
	Process Improvement
	Process Design/Redesign
	Process Management
	9.8.4 The Two Improvement Processes
	9.8.5 The Five-Step Road Map
	9.8.6 The Organization for the Six Sigma System
	9.9 Summary of Quality Management Systems
	9.10 Exercise
	9.10.1 Practice Problems
	Deming System
	Juran System
	Baldrige System
	ISO 9000 System
	Six Sigma System
	9.10.2 Mini-Projects
			Mini-Project 9.1 The above set of 30 questions has been created to help students understand the various systems in good detail. However, it is only one of several possible sets. Generate another set of 30 questions, six from each system, similar to but di
	Mini-Project 9.2 Compare the three modern systems— Baldrige Award, ISO 9000, and Six Sigma—and identify their differences.
	References
	Appendix 1: Statistical Tables
	Appendix 2: Answers to Selected Exercises
	Index