Total Quality Management

Module : Quality Control and improvements

Dr Appaso M Gadade
Assistant Professor, MED
Email: appaso.gadade@thapar.edu
 Syllabus
1. Causes of variation: special and chance causes
2. Sporadic and chronic problems
3. Juran’s Quality Triology: Different zones and their functions
4. Quality planning activities
5. Crossby’s and Deming’s quality viewpoint
6. Variable Vs Attribute data
7. Different distributions, their significance and applications
8. Control Charts for variables
9. Underlying principle, advantages, limitations and applications of: X-bar and R charts, X-bar and S charts, X
and MR charts
10. Control charts for attributes
11. Underlying principle, advantages, limitations and applications of: p, np, c and u charts
 References
1. Juran, J.M. and Gryna, F.M., Quality Planning & Analysis, 3rd Edition, McGraw Hill (2001).
2. Grant, E.L., Statistical Quality Control, 3rd Edition, McGraw Hill (2014).
3. Beckwith, T.G., Marangoni, R.D., and Lienhard, J.H., Mechanical Measurements, 6th Edition, Pearson
Education (2006).
Why process quality improvement is important?
 The key to success in an organization is to understand how their processes work to deliver
the required outputs. If process improves
We will get the best output or responses and as a consequence best desired quality product for the end
customer.
 Any process should add value and unnecessary waste activities should be eliminated from the
process steps.

 General Model of a process:

 Determining the best setting for controllable variables and reduction of variation is the
primary focus of process quality improvement activity.
 Is
Process
Stable
The Define
?

Quality Process

Improvement Select
Measures
Model
Collect &
Interpret
Data

Is Process Stable? Is No Investigate &

Process
Fix Special
Stable
Causes
Purpose: Determine the ?

stability of key Yes

measures of the
Is
product. Improve
Process
No
Process
Capable
Capability
?

Yes
Use SPC to
Maintain
Current
Process

6-5
 Causes of variation: special and chance causes
 Once the design of quality has been specified, the whole production process will be adjusted
to manufacture the product according to the specifications.
 At this stage, the problem of conformance ( producing items that conforms to the design
specifications) starts.
 The problem arise from the fact that every manufacturing process exhibits variability.
 Variation in the quality of manufactured product in the repetitive process in industry is
inherent and unavoidable.
Variation may be due to
tool wear, machine vibrations, loose bearings, faulty fixtures,
humidity, poor ventilation, poor quality of raw materials,
carelessness of labours, inexperienced operators, weather, voltage
fluctuations, poor maintenance, measuring errors etc.
 Causes of variation: special and chance causes
 Variations affecting a production process are broadly classified as being due to two causes
– Chance causes (common causes/random causes)
The natural or inherent
- Assignable causes(non random causes or special causes) variability of the process is the

 Chance causes cumulative effect of these

uncontrollable chance causes.
 Minor causes that behaves in a random manner.
 The variation due to these causes is beyond the control of human hand
 Cannot be prevented or eliminated under any circumstances
The range of such variations (called
 Many in number
allowable variation) is known as natural
 Each contribute a little to the variation tolerance of the production process.
 Are continuously active in the process and are built in as a part of the process.
 Causes of variation: special and chance causes
 Assignable causes
 Generally few in number
 Each has a significant effect in the variation of the output of the process.
 Can be detected
 Can be eliminated from the process

Such variations are termed as preventable variation.

Natural or Un natural or
Total Variation in
Inherent or External or
the production
Allowable Preventable
process
Variation Variation
Sporadic and chronic problems
Sporadic and chronic problems
Sporadic and chronic problems
Various process quality Control and improvement techniques
 Part A: Popular Tools of QC & SPC
 Pareto Diagram, Process Flow Diagram, Cause and Effect Diagram, Check Sheets,
Histogram, Run Charts, and Scatter diagram, Control charts, Acceptance Sampling.
 Hypothesis testing, Design of Experiment (DOE) and ANOVA (Analysis of Variance).
Failure Mode Effect Analysis (FMEA).
 Part B: Additional Tools and Techniques under the various Quality management approaches :
 Juran, Deming, and Crosby quality philosophies, 5S principle, Total Quality Management
Six Sigma Philosophy, etc.
 Continuous Quality Improvement (CQI), Benchmarking, Quality Function Deployment
(QFD), Plan-Do-Check-Act (PDCA) Cycle, etc.
Failure Mode and Effects Analysis (FMEA)
 It is also known as failure modes, effects and criticality analysis (FMECA).
FMEA is a step-by-step approach for identifying all possible failures in a design,
a manufacturing or assembly process, or a product or service.
"Failure modes" means the ways, or modes, in which something might fail. Failures are
any errors or defects, especially ones that affect the customer, and can be potential or
actual.
"Effects analysis" refers to studying the consequences of those failures.
Failures are prioritized according to what are the risks of failures & how serious
their consequences are, how frequently they occur, and how easily they can be
detected.
The purpose of the FMEA is to take actions to eliminate or reduce failures, starting with the
highest-priority ones.
The 10-steps of traditional FMEA (RPN)
 Traditional FMEAs
 Traditional FMEAs establish an
Risk Priority Number (RPN) for
each failure mode and its resulting
effect(s).
 RPN = Severity rating X
Occurrence rating X Detection
rating;
The RPN can range from a low of 1
to a high of 1,000
 Once the RPNs are determined, an
Action Plan is developed to
reduce the risks of failure modes
of unacceptably high RPNs.
Design of Experiment (DoE)
 What is DoE?
A designed experiment is a test or series of tests in which
purposeful changes are made to the input variables of a process
so that we may observe and identify corresponding changes in
the output response.
Experimental design methods may be used either in process
development or process troubleshooting to improve process
performance.
In the context of DoE, controllable parameters are called factors, whereas uncontrollable
parameters are often known as noise
 Guidelines for DoE

15
Problem 1:
The safe operation of an automobile is dependent on several subsystems (e.g.,engine, transmission, braking
mechanism). Construct a cause-and-effect diagram for automobile accidents. Conduct a failure mode and effects
criticality analysis and comment on areas of emphasis for prevention of accidents.

16
Problem 1:
The safe operation of an automobile is dependent on several subsystems (e.g.,engine, transmission, braking
mechanism). Construct a cause-and-effect diagram for automobile accidents. Conduct a failure mode and effects
criticality analysis and comment on areas of emphasis for prevention of accidents.

Comments: Rating scores on severity,

occurrence, and detection are
assigned and the risk priority number
(RPN) is calculated. From the
calculated RPN values, the highest
value (315) is associated with
transmission failure due to broken
belts. Some action plans need to be
designed to detect such broken or
imminent to break belts during routine
or preventive
maintenance.

17
Problem 2:
Consider previous Exercise on the prevention of automobile accidents. However, in this exercise, consider the driver of the
automobile. Construct a cause-and-effect diagram for accidents influenced by the driver. Conduct a failure model and
effects criticality analysis considering issues related to the driver, assuming that the automobile is in fine condition.

18
Problem 2:
Consider previous Exercise on the prevention of automobile accidents. However, in this exercise, consider the driver of the
automobile. Construct a cause-and-effect diagram for accidents influenced by the driver. Conduct a failure model and
effects criticality analysis considering issues related to the driver, assuming that the automobile is in fine condition.
Comments: A FMECA analysis
is shown in Table 3-4. Rating
scores on severity, occurrences,
and detection are assigned and
the risk priority number (RPN) is
calculated. From the calculated
RPN values, the highest value
(640) is associated with
emotionally unfit due to personal
issues, followed by emotionally
unfit due to disturbed work
environment. Detection of such
causes are difficult, specially
personal issues. This leads to
high RPN values which draws
attention to create action items
to address these issues.

19
 Juran’s Quality Triology
 Managing for quality consists of three basic quality-oriented processes:
 quality planning,
 quality control, and
 quality improvement.
 The role of quality planning is to design a process that will be able to meet established goals
under operating conditions.
 The role of quality control is to operate and when necessary correct the process so that it
performs with optimal effectiveness.
 The role of quality improvement is to devise ways to take the process to unprecedented
levels of performance.
 Juran’s Quality Triology: Planning
 Quality planning stems from a unity of purpose that spans all functions of an organization.
 The subject of planning can be anything -- an engineering process for designing new
products, a production process for making goods, or a service process for responding to
customer requests.
 Quality Planning involves
 Identifying customers, both internal and external
 Determining their needs
 Specifying the product features that satisfy those needs at minimum cost.
 Designing the processes that can reliably produce those features.
 Proving that the process can achieve its goals under operating conditions.
 Juran’s Quality Triology: Control
 The process of managing operations to meet quality goals.
 The process of Quality Control involves:
 Choosing control subjects
 Choosing units of measurement
 Establishing a measurement procedure
 Measuring
 Interpreting differences between measurement and goal.
 Taking action to correct significant differences
 Juran’s Quality Triology: Improvement
 Assuming the process is under control, any waste that occurs must be inherent in the design
of the process.
 The object of quality improvement is to reduce chronic waste to a much lower level.
 The steps in Quality Improvement:
 Prove the need for improvement
 Identify specific projects for improvement
 Organize to guide the projects
 Organize for diagnosis -- discovery of causes
 Diagnose the causes
 Provide remedies
 Prove that the remedies are effective under operating conditions
 Provide for control to maintain the gains.
THE JURAN TRILOGY
Quality Planning: Summary

24
THE JURAN TRILOGY
Quality Control: Summary

25
THE JURAN TRILOGY
Quality Improvement: Summary

26
Deming’s Plan-Do-Check-Act (PDCA) cycle
This is a continuous cycle of process improvement.
Later on Deming modified the “Check” step “Study”: PDSAcycle.

• Recognize Problem
• Ensure performance • Form quality improvement teams
• Continuous • Define Problem
improvement • Analyze Problem
• Determine Possible Causes
• Identify Possible Solutions

Implement
Evaluate
Solution
Solution
27
Statistical Hypothesis Testing
 A statistical hypothesis is a statement about parameter(s) of one or more populations.
It is a statement about the population or distribution under study, AND not statements about the
sample.
However, the hypothesis is tested based on the observation of sample.
 Now, If the information obtained from sampling is consistent with the null hypothesis then
we will conclude that the null hypothesis is true.
However, if this information is inconsistent with null hypothesis, we will conclude that there is
little evidence to support null hypothesis.
We will reject the null hypothesis.
 For example, suppose that for an application our interest focuses on mean shaft diameter (a
parameter of this distribution).
Specifically, we are interested in deciding whether or not the mean diameter is 60 mm.
For this purpose, mathematically we can make the following statement

The statement Ho: =60 mm is called the null hypothesis, and the statement H1: =/60 mm is so-
called the alternative (here it is two-sided alternative)
28
hypothesis.
Analysis of variance (ANOVA)
 Analysis of variance (ANOVA) is the parametric statistical technique that is used
to check if the means of two or more groups are significantly different from each other
to understand the influence of the factors on variability of output characteristic.
 In short, ANOVA is a statistical tool used in several ways to develop and confirm an
explanation for the observed data.
 ANOVA is used to support other statistical tools
Once the test is finished, an analyst performs additional testing on the methodical factors that
measurably contribute to the data set's inconsistency.

 Types of ANOVA
One way, Two way, K-way

29
Other Tools, Techniques & Strategies used for Quality improvement
 Continuous Quality Improvement (CQI): Is a management philosophy which contends that most
things can be improved by the efforts from the entire team, from management to the workers to
eliminate waste.
 Benchmarking: Systematic search for best practices, innovative ideas, and highly effective
operating practices.
 QFD: is a structured approach to defining customer needs/requirements and translating them
into specific plans to produce products to meet those needs.
 Quality Standards: ISO9000 series, ASQ
 5S principle: It is one of the most widely used and fundamental components of Lean Manufacturing
& are pre-requisites for any improvement program
 Total Quality Management: TQM is a strategy for implementing and managing quality
improvement activities on all dimensions of products and services (organization-wide) that are
important to the customer—which will lead to customer satisfaction.
 Other Tools, Techniques & Strategies used for Quality improvement
 Juran, Deming, and Crosby quality philosophies
All of these three quality gurus preached for the
following common points for a successful program
in quality.
Should be Customer Focused
Top management support and commitment are
essential.
Most problems associated with quality can be attributed
to management policy or action.
There are no shortcuts to quality and the pursuit of
quality must be a continuous effort.
Education and training must be continuous.
Effective communication and teamwork at all levels
are essential. 15
 Variables and Attributes
 Quality characteristics fall into two broad classes: variables and attributes
 Characteristics that are measurable and are expressed on a numerical scale are called
variables
 Example: The waiting time in a bank before being served, expressed in minutes, density
of a liquid in grams per cubic centimeter and the processing speed of a computer.
 A quality characteristic is said to be an attribute if it is classified as either conforming or
nonconforming to a stipulated specification
 A nonconforming unit has one or more nonconformities such that the unit is unable to
meet the intended standards and is unable to function as required
 Example : A CI pipe whose ID and weight both fail to satisfy specifications, thereby
making the unit dysfunctional
 Statistics
 A population is the set of all items that possess a certain characteristic of interest.
 A sample is a subset of a population
 A parameter is a characteristic of a population, something that describes it
 Statistics is the science that deals with the collection, classification, analysis, and making of inferences from
data or information. Types: Descriptive statistics and inferential statistics.
 Descriptive statistics describes the characteristics of a product or process using information collected on it
 Inferential statistics draws conclusions on unknown product or process parameters based on information
contained in a sample
 Data on quality characteristics are described by a random variable and are categorized as continuous or
discrete
 Continuous variable that can assume any value on a continuous scale within a range is said to be continuous.
 Discrete Variable that can assume a finite or countably infinite number of values are said to be discrete.
 Statistics
 Mean: The mean is the simple average of the observations in a dataset
 The sample mean, or average (denoted by X ), is found by adding all observations
in a sample and dividing by the number of observations (n) in that sample

 The population mean(μ) is found by adding all the data values

in the population and dividing by the size of the population(N).
 The population mean(μ) is sometimes denoted as E(X), the expected value of the random variable X. It is also
called the mean of the probability distribution of X.
 Median: The median is the value in the middle when the observations are ranked

 Mode: The mode is the value that occurs most frequently in the dataset

 Trimmed Mean: The trimmed mean is a robust estimator of the central tendency of a set of observations.
 Range: A widely used measure of dispersion in quality control is the range, which is the difference between the
largest and smallest values in a dataset
 Statistics
 Variance: The variance measures the fluctuation of the observations around the mean.
The larger the value, the greater the fluctuation. The population variance σ2 is given by
 The sample variance s2 is given by Modified version

 Standard deviation: measures the variability of the observations

around the mean. It is equal to the positive square root of the
variance. The population standard deviation σ is given by

 Sample standard deviation s is given by

 Skewness Coefficient : describes the asymmetry of the data set about the mean

 Kurtosis Coefficient: measure of the peakedness of the data set. It is also viewed as a measure of the “heaviness”
of the tails of a distribution.
 Statistics
 A correlation coefficient is a measure of the strength of the linear relationship between two variables in
bivariate data. If two variables are denoted by X and Y, the correlation coefficient of a sample of observations
is found from

 Alternative version:
 Probability Distributions
 For a discrete random variable X, which takes on the values x1, x2, and so on, a probability distribution
function p(x) has the following properties:

 When X is a continuous random variable, the probability density function is represented by f(x), which has
the following properties:
 Cumulative Distribution Function
 The cumulative distribution function (cdf) is usually denoted by F(x) and represents the probability of the
random variable X taking on a value less than or equal to x, that is,

 For a discrete random variable,

 If X is a continuous random variable,

 Note that F(x) is a nondecreasing function of x such that,

 Expected Value
 The expected value or mean of a random variable is given by

 and,

 The variance of a random variable X is given by,

Discrete Distributions: Hypergeometric Distribution
 The discrete class of probability distributions deals with those random variables that can take on a finite or
countably infinite number of values.
 A hypergeometric distribution is useful in sampling from a finite population (or lot) without replacement (i.e.,
without placing the sample elements back in the population) when the items or outcomes can be categorized
in to one of two groups (usually called success and failure).

 If we consider finding a nonconforming item a success, the probability distribution of the number of
nonconforming items (x) in the sample is given by

 where D = number of nonconforming items in the population, N = size of the population, n=size of the sample,
x = number of nonconforming items in the sample, and = combination of D items taken x at a time,
D!/(x!(D-x)!)
Discrete Distributions: Hypergeometric Distribution
 The factorial of a positive integer x is written as x!=x(x–1)(x–2)∙∙∙3 2 1, and 0! is defined to be 1. The mean (or
expected value) of a hypergeometric distribution is given by

 The variance of a hypergeometric random variable is given by

 Discrete Distributions: Binomial Distribution
 Consider a series of independent trials where each trial results in one of two outcomes.

 These outcomes are labeled as either a success or a failure.

 The probability p of success on any trial is assumed to be constant

 Let X denote the number of successes if n such trials are conducted. Then the probability of x successes is
given by

 and X is said to have a binomial distribution. The mean of the binomial random variable is given by

 and the variance is expressed as

 A binomial distribution is a distribution using the two parameters n and p. If the values of these parameters are
known, all information associated with the binomial distribution can be determined.
 Discrete Distributions: Poisson Distribution
 A Poisson distribution is used to model the number of events that happen within a product unit (number of
defective rivets in an airplane wing), space or volume (blemishes per 200 square meters of fabric), or time
period (machine breakdowns per month).
 It is assumed that the events happen randomly and independently.
 The Poisson random variable is denoted by X. An observed value of X is represented by x. The probability
distribution(or mass) function of the number of events(x) is given by
 Dr Appaso M Gadade
Assistant Professor, MED
Email: appaso.gadade@thapar.edu