Douglas C. Montgomery

1
Introduction to Statistical Quality Control, 4th
Edition

• Douglas C. Montgomery
• Arizona State University

With my revisions
2
Chapter 1

• Quality Improvement in the Modern Business
  Environment

3
1-1. The Meaning of Quality and Quality
Improvement

• 1-1.1 Dimensions of Quality
• 1-1.2 Quality Engineering Technology

4
1-1.1 Dimensions of Quality

• Aesthetics
• Features
• Perceived Quality
• Conformance to standards

• Performance
• Reliability
• Durability
• Serviceability

5
Performance

• Will the product perform its intended function?
• Will this industrial scale measure up to 4000
  pounds in 5 pound graduations?

6
Reliability

• How often does the product fail?
• How often do I have to take this Toyota Camry to
  the service department?

7
Durability

• How long will the product last?
• The product should perform satisfactorily over a
  long period of life
• If I buy that 1999 Toyota 4Runner when the lease
  expires, will it give me good service for 10 more
  years?

8
Serviceability

• How easy is it to repair the product?
• If amazon.com sends the wrong book, how hard is
  it to get this error corrected?

9
Aesthetics

• Is the product pleasing to the senses?
• Do you like the box in which Timberland Shoes are
  packaged?

10
Features

• What will the product do beyond the basics?
• Does the new Ford Focus come with a CD changer,
  or do you have to pay extra for that?

11
Perceived quality

• What is the reputation of the company selling
  this product?
• Is Delta really ready when you are?
• Why do people buy Gucci handbags?

12
Conformance to standards

• Is the product made as designed?
• Why do they have to beat on those Ford Taurus
  bodies when they are being assembled in Hapeville?

13
1-1.1 Dimensions of Quality

• Definitions of Quality
• Quality means fitness for use
• - quality of design
• - quality of conformance
• Quality is inversely proportional to
  variability.

14
Quality of design

• Automobile differences
• Materials used in construction
• Specifications of the components
• Reliability of drive train components
• Reliability of accessories

15
Quality of conformance

• How well does the product conform to the
  specifications required by the design?
• Choice of manufacturing processes
• Training of the workers
• Supervision of the workers
• Motivation of the workers
• Quality-assurance procedures that were used

16
Quality is inversely proportional to variability

• Toyota Lexus versus Ford Taurus
• That transmission noise (or lack of it) is wasted
  energy caused by components that dont fit
  precisely
• Imprecise components lead to wear and tear
• Ford Taurus/Mercury Sable transmission lasts for
  64,000 miles

17
1-1.1 Dimensions of Quality
Transmission Example
N(100,10) vs N(100,4)
18
1-1.1 Dimensions of Quality

• Quality Improvement
• Quality improvement is the reduction of
  variability in processes and products.
• Alternatively, quality improvement is also
  seen as waste reduction.

19
1-1.2 Quality Engineering Terminology

• Quality Characteristics
• Physical - length, weight, voltage, viscosity
• Sensory - taste, appearance, color
• Time Orientation - reliability, durability,
  serviceability

20
1-1.2 Quality Engineering Terminology

• Quality engineering is the set of operational,
  managerial, and engineering activities that a
  company uses to ensure that the quality
  characteristics of a product are at the nominal
  or required levels.

21
Inherent variability

• No two products are ever identical
• Slight differences in materials
• Slight differences in machine settings
• Slight differences in operators
• Slight differences in ambient temperature during
  production
• Papermate Pens not much different
• Fins on a turbine engine quite a lot

22
1-1.2 Quality Engineering Terminology

• Two types of data
• Attributes Data - discrete data, often in the
  form of counts
• Variables Data - continuous measurements such as
  length, weight
• Both types will be discussed in the course

23
1-1.2 Quality Engineering Terminology

• Specifications
• Quality characteristics being measured are
  often compared to standards or specifications.
• Desired measure for the quality characteristic
• Example Shaft and bearing
• Too loose the assembly will wobble causing
  wear
• Too tight, and the assembly can not be made,
  no clearance

24
1-1.2 Quality Engineering Terminology

• Specifications
• Nominal or target value
• Desired value for a quality characteristic

25
1-1.2 Quality Engineering Terminology

• Specifications
• Upper Specification Limit (USL)
• Lower Specification Limit (LSL)
• Largest and smallest allowable values

26
1-1.2 Quality Engineering Terminology

• Specifications
• Upper Specification Limit (USL)
• Lower Specification Limit (LSL)
• One-sided
• The compression strength of a Coke bottle must be
  greater than a given psi value
• Two-sided
• The weight of potato chips in the bag can be
  between 7.8 and 8.3 ounces

27
Design specifications

• Over the wall
• From design to manufacturing
• Cooperatively
• Between design and manufacturing

28
1-1.2 Quality Engineering Terminology

• When a component or product does not meet
  specifications, it is considered to be
  nonconforming.
• A nonconforming product is considered defective
  if it has one or more nonconformities that may
  seriously affect the safe or effective use of the
  product.

29
1-1.2 Quality Engineering Terminology

• A new car is purchased
• A bubble in the paint on the door is noticed
• Nonconformity yes
• Defective car - no

30
1-1.2 Quality Engineering Terminology

• Concurrent Engineering
• Team approach to design. Specialists from
  manufacturing, quality engineering, management,
  etc. work together for product or process
  improvement.

31
1-2. A Brief History of Quality Control
and Improvement

• (Refer to Table 1-1)
• Frederick Taylor (1875) introduces the principles
  of scientific management dividing work into
  tasks with standardized procedures
• The Gilbreths developed standard times and
  motions (1920s)

32
1-2. A Brief History of Quality Control
and Improvement

• (Refer to Table 1-1)
• Walter Shewhart (1924) introduced statistical
  control chart concepts and QC begins
• Dodge and Romig (1928), Bell Labs, develop
  acceptance sampling as an alternate to 100
  inspection
• During WW II the shells didnt fit the howitzers
  leading to development of MIL-STDs

33
1-2. A Brief History of Quality Control
and Improvement

• (Refer to Table 1-1)
• The American Society for Quality Control formed
  in 1946 now known as the American Society for
  Quality (ASQ)
• 1950s and 1960s saw an increase in reliability
  engineering, experimental design, and statistical
  quality control

34
1-2. A Brief History of Quality Control
and Improvement

• (Refer to Table 1-1)
• Competition from foreign industries (Japan)
  increases during the 1970s and 1980s.
• Statistical methods for quality improvement use
  increases in the United States during the 1980s
• Total Quality Management (TQM) emerges during
  1970s and into the 1980s as an important
  management tool to implement statistical methods.

35
1-2. A Brief History of Quality Control
and Improvement

• Malcolm Baldrige National Quality Award is
  established in 1988.
• ISO 9000 certification activities increase in
  U.S. industry in the 1990s.
• Motorolas Six-Sigma initiative begins in the
  1990s.

36
1-3. Statistical Methods for Quality Control and
Improvement

• Three major areas
• Statistical process control (SPC)
• Design of experiments (DOE)
• Acceptance sampling

37
1-3. Statistical Methods for Quality Control and
Improvement

• Statistical Process
• Control (SPC)
• Control charts are used for process monitoring
  and variability reduction.
• SPC is an on-line quality control tool.

38
1-3. Statistical Methods for Quality Control and
Improvement

• Design of Experiments
• Experimental design is an approach to
  systematically varying the controllable input
  factors in the process then determining the
  effect these factors have on the output
  responses.
• Experimental designs are off-line quality tools.
• Crucial for variability reduction.

39
1-3. Statistical Methods for Quality Control and
Improvement

• Acceptance Sampling
• Acceptance sampling is the inspection and
  classification of a sample of the product
  selected at random from a larger batch or lot and
  the ultimate decision about disposition of the
  lot.
• Two types
• 1. Outgoing inspection - follows production
• 2. Incoming inspection - before use in
  production

40
Quality cannot be inspected into the product

• When the organization realizes this, process
  improvement efforts begin

41
The objective

• Systematic reduction of variability
• First, by using acceptance sampling
• Then, by using SPC
• Finally, by using DOE
• We dont stop when requirements are met
• Further reductions in variability lead to better
  performance
• Avoid the Taurus transmission

42
1-4. Other Aspects of Quality Control and
Improvement

• Total Quality Management (TQM)
• TQM is a managerial framework to accomplish
  quality improvement.
• Other names and related approaches
• Company-Wide Quality Control (CWQC)
• Total Quality Assurance (TQA)
• Six-Sigma

43
1-4. Other Aspects of Quality Control and
Improvement

• 1-4.1 Quality Philosophy and Management
• Strategies
• 1-4.2 The Link Between Quality and
• Productivity
• 1-4.3 Quality Costs
• 1-4.4 Legal Aspects of Quality
• 1-4.5 Implementing Quality Improvement

44
1-4.1 Quality Philosophy and Management
Strategies

• Three Important Leaders
• W. Edwards Deming
• - Emphasis on statistical methods in quality
  improvement
• Joseph Juran
• - Emphasis on managerial role in quality
  implementation
• Armand V. Feigenbaum
• - Emphasis on organizational structure

45
W. Edwards Deming

• After WW II, he consulted with Japanese
  industries
• Demanded management commitment to use statistical
  methods
• Deming Prize in Japan
• For quality improvement
• Deming was a harsh critic of US management
  practices

46
1. Create a constancy of purpose

• Focus on the improvement of products and services
• Constantly improve product design and performance
• Invest in RD
• Innovate

47
2. Adopt a new philosophy

• Eliminate defective products
• It costs as much to produce a defective unit as a
  good one
• Dealing with scrap and rework is very expensive

48
3. Dont rely on inspection

• Inspection only sorts out defectives
• Already have paid to produce them
• Inspection is too late in the process
• Its also ineffective
• Prevent defectives through process improvement

49
4. Dont award business on price alone

• Consider supplier quality as well
• Give preference to those suppliers that
  demonstrate process control and process capability

50
5. Focus on continuous improvement

• Involve the workforce
• Use statistical techniques

51
6. Invest in training

• Everyone should be trained in the technical
  aspects of their job, QC, and process improvement
• Workers should be encouraged to put this training
  to use

52
7. Practice modern supervision methods

• Help the employees improve the system in which
  they work

53
8. Drive out fear

• Create an environment where the workers will ask
  questions, report problems, or point out
  conditions that are barriers to quality

54
9. Break down the barriers

• Break down the barriers between the functional
  areas of the business
• Only through teamwork can quality and process
  improvement take place

55
10. Eliminate targets and slogans

• Useless without a plan for the achievement of the
  target or goal
• Instead, improve the system and provide
  information on that

56
11. Eliminate quotas

• Numerical quotas and work standards often
  conflict with quality control

57
12. Encourage employees to do their job

• Remove the barriers
• Listen to the workers
• The person doing the job knows more about it than
  anyone else

58
13. Have ongoing education and training

• Teach them simple yet powerful statistical
  techniques
• Use the basic SPC tools, particularly the control
  chart

59
14. Involve top management

• Management should be advocates for these points

60
Dr. Joseph Juran

• A founder of SQC
• Co-author of QC Handbook (1957)
• His philosophy is based on management of the
  quality function
• Says that 80 of the opportunities for quality
  improvement can be addressed by management and
  only 20 by the workers

61
Dr. Armand Feigenbaum

• Total Quality Control (1951)
• Concern is with organizational structure to
  improve quality
• Says that QC should be concentrated in a
  specialized department
• Conflicts with Deming on this point

62
1-4.1 Quality Philosophy and Management
Strategies

• Total Quality Management (TQM)
• Quality Standards and Registration
• ISO 9000
• Six Sigma
• Just-In-Time, Lean Manufacturing, Poka-Yoke, etc.

63
TQM

• Began in the early 80s based on the philosophies
  of Deming and Juran
• Evolved into wide spectrum of ideas
• Participation in quality groups
• Work culture
• Customer focus
• Supplier quality improvement
• Cross-functional teams concerned with quality

64
TQM

• A success?
• Moderately
• Why not?
• Not enough concern for reduction of variability
• Ineffective training conducted by HR people
• No knowledge of what is important
• Success measured by of workforce trained
• Management not committed

65
TQM

• More reasons for lack of success
• Zero defects, value engineering, quality is free
• Programs with no emphasis on reducing variability

66
ISO 9000

• Quality system oriented
• Say what you do, do what you say
• Much effort devoted to paperwork and bookkeeping
• Not much to reducing variability and improving
  processes

67
ISO 9000 paragraphs

• 1. Management responsibility for quality
• 2. Design control
• 3. Document and data control
• 4. Purchasing and contract management
• 5. Product identification and traceability
• 6. Inspection and testing, including control of
  measurement and inspection equipment

68
ISO 9000 paragraphs

• 7. Process control
• 8. Handling of nonconforming product
• 9. Handling, storage, packaging and delivery of
  product
• 10. Control of quality records
• 11. Internal audits
• 12. Training
• 13. Statistical methodology

69
ISO 9000

• US40 billion annual business worldwide
• Registrars, auditors, consultants
• Plus, 1000s of hours of internal costs
• Effective?
• Does it reduce variability?

70
Six Sigma

• Developed by Motorola in the late 80s
• Consider that 3s provides 0.00135 in each tail,
  or 0.00270 in the two tales
• So, in 1 million parts, 2700 would be defective

71
Six Sigma

• Consider an assembly of 100 parts that must all
  function for the assembly to function
• .9973 x .9973 x ..9973 (.9973)100 .7631
• Thus, about 23.7 of the products under 3s will
  fail
• Not usually an acceptable situation

72
Six Sigma

• But, 6s results in 0.999999998 inside specs
• (0.999999998)100 .9999998
• Or, 2 parts/billion defective
• i.e., 0.2 ppm
• Much better than 3s

73
Six Sigma

• Has moved beyond Motorola
• Has come to encompass much more
• Has become a method for improving corporate
  business performance
• Companies involved in Six Sigma use teams that
  work on projects involving quality and costs

74
Six Sigma

• More successful than TQM
• More managerial commitment
• Involves costs
• But, its still another slogan and program
• Better to train everyone in quality tools and
  make efforts to reduce variability

75
JIT, Lean Manufacturing, etc.

• Programs that devote too little attention to
  variance reduction
• For example, JIT
• The variability in demand results in a need for
  inventory
• Reduce the variability and reduce the inventory

76
1-4.2 The Link Between Quality and
Productivity

• Effective quality improvement can be instrumental
  in increasing productivity and reducing cost.
• The cost of achieving quality improvements and
  increased productivity is often negligible.

77
An example

• Data
• 100 parts/day are manufactured
• 75 are conforming
• 60 of the nonconforming can be reworked for a
  cost of 4
• Remainder are scrapped
• Direct manufacturing cost is 20/part

78
An example

• Cost/conforming part
• 20 (100) 4 (15)/90 22.89
• Note that the yield is 90 conforming/day

79
An example

• New process us introduced
• Fallout is 5
• 60 can be reworked
• Cost/conforming part
• 20 (100) 4 (3)/98 20.53
• Note that the yield is 98 conforming/day
• Up from 90/day
• And, costs are reduced by 10.3

80
1-4.3 Quality Costs

• Quality Costs are those categories of costs that
  are associated with producing, identifying,
  avoiding, or repairing products that do not meet
  requirements. These costs are
• Prevention Costs
• Appraisal Costs
• Internal Failure Costs
• External Failure Costs

81
Quality costs

• Prevention costs
• Quality planning and engineering
• New products review
• Product/process design
• Burn-in
• Training
• Quality data acquisition and analysis

82
Appraisal costs

• Inspection and test of incoming material
• Product inspection and test
• Materials and services consumed
• Maintaining accuracy of test equipment

83
Internal failure costs

• Scrap
• Rework
• Retest
• Failure analysis
• Downtime
• Yield losses
• Downgrading (off-specing)

84
External failure costs

• Complaint adjustment
• Returned product/material
• Warranty charges
• Liability costs
• Indirect costs

85
Pareto analysis

• Cost reduction through identifying improvement
  opportunities
• Identifying quality costs by category, or by
  product, or by type of defect or nonconformity

86
Monthly quality costs for PCB assembly
87
Pareto analysis

• Insufficient solder
• 42 of defects and 52 of scrap and rework costs
• Work on that defect first
• Most of the cost reductions will come from
  attacking the few problems that are responsible
  for the majority of the quality costs

88
Appraisal or prevention

• Many firms spend far too much of their quality
  management budget on appraisal and not enough on
  prevention
• Money spent on prevention has a much better
  payoff than money spent on appraisal

89
1-4.4 Legal Aspects of Quality

• The re-emergence of quality assurance as an
  important business strategy is in part a result
  of
• Consumerism
• Product Liability

90
Consumerism

• Virtually every product line of today is superior
  to that of yesterday
• But, many consumers see it otherwise
• Consumer tolerance for minor defects aesthetic
  problems has decreased considerably
• Blemishes, surface-finish defects, noises,
  appearance problems

91
Consumerism

• Many manufacturers introduce new designs before
  they are fully evaluated and tested
• To remain competitive
• Unproved designs

92
Product liability

• Manufacturers and sellers are likely to incur a
  liability when they have been unreasonably
  careless or negligent in what they have designed,
  or produced, or how they have produced it

93
More stringent Strict liability

• 1. There is a strong responsibility for both
  manufacturer and merchandiser requiring immediate
  responsiveness to unsatisfactory quality through
  product service, repair, or replacement of
  defective product
• Extends into the period of use by the consumer
• By producing the product, manufacturer and seller
  must accept responsibility for use

94
More stringent Strict liability

• 2. All advertising statements must be supportable
  by valid company quality or certification data

95
1-4.5 Implementing Quality Improvement

• Strategic management of quality
• Almost all successful efforts have been
  management-driven.
• Too much emphasis on registration and
  certification programs (ISO, QS)
• Insufficient focus on quality planning and
  design, quality improvement, overemphasis on
  quality assurance
• Poor use of available resources

96
End