Chapter 9 Quality Engineering

1
Chapter 9Quality Engineering
2
Agenda

• Quality
• Quality Dimensions
• Quality Costs
• A Framework for Quality Improvement
• Failure Mode and Effect Analysis
• Improving Product Quality During the Production
  Phase
• Automated Inspection Systems

3
Quality

• Quality is the totality of features and
  characteristics of a product or service that bear
  on its ability to satisfy a given need
• Features, characteristics ? performance
• Need customer requirements ad expectations
• QFD is a technique that is often used to
  translate the customer requirements or
  expectations into measurable quality
  characteristics

4
The Dimensions of Quality

• Performance primary operating characteristics
• Features secondary operating characteristics
• Time waiting time in line, lead time, time to
  complete a service
• Reliability extent of failure-free operation
• Durability amount of use until replacement is
  preferable to repair
• Uniformity low variation among repeated outcomes
  of a process
• Consistency match with documentation,
  advertising, deadlines or industry standards
• Serviceability resolution of problems and
  complaints
• Aesthetics characteristics related to senses
• Personal Interface punctuality, courtesy,
  professionalism
• Harmlessness characteristics related to safety,
  health, or environment
• Perceived Quality indirect measures or
  inferences about one or more of the dimensions,
  reputation

5
Quality Costs

• Preventive costs making the product right the
  first time. Elements quality planning and
  engineering, new product reviews, product and
  process design, process control, training,
  quality data acquisition and analysis
• Appraisal costs costs involved in measuring and
  evaluating, or auditing products, components, and
  purchased materials
• Internal failure costs when products fail before
  shipping to customers. Elements failure
  analysis, scrap, repair, etc.
• External failure costs when products do not
  function satisfactorily after shipping to
  customers. Elements warranty charges, lost of
  market share, complaint adjustment and dealing
  with returned products, etc.
• Quality costs 4 - 40 of sales

6
A Framework for Quality Improvement

• Types of quality improvement activities
• Off-line quality control activities conducted at
  the product and process design stages in the
  product development cycle
• On-line quality control activities conducted at
  the manufacturing stage (SPC)
• Activities conducted during the product usage
  stage to provide maintenance and after-sales
  product service
• Designing quality into products and processes
• Requirements it is necessary to consider
• All aspects of design that influence the
  deviations of functional characteristics from the
  target
• Methods for reducing undesirable and
  un-controllable factors that cause functional
  deviations

7
A Framework for Quality Improvement

• Taguchi method designing quality into products
  and process by a three-phase program
• System design new concepts, ideas, methods are
  synthesized to establish the basic design
  concept (selection of materials, parts,
  subassemblies, etc.). Technique QFD
• Parameter design input parameters are set, wide
  tolerances on noise factors are
  assumed/established. Techniques DOE, simulation,
  optimization
• Tolerance design tightening tolerances to reduce
  performance variationgtltmanufacturing cost

8
A Framework for Quality Improvement

• Robust design of products and processes achieve
  the target of quality characteristics and
  reduction of performance variation, ensure
  minimum quality loss
• Factors affect the target and the variance of a
  products quality characteristic
• Controllable factors factors controlled by the
  user/operator and factors controlled by designers
  (variability control, target control and neural
  factors)
• Uncontrollable factors (noise) outer noise,
  inner noise, between-product noise
• Taguchis approach to robust design select
  values of controllable factors ? remove or reduce
  the impact of noise factors

9
A Framework for Quality Improvement

• Parameter design the Taguchi approach
• Identify the quality characteristic to be
  observed and the objective function to be
  optimized
• Identify the design parameters and alternative
  levels
• Define possible interactions between these
  parameters
• Design the matrix experiment and the define-data
  analysis procedure
• Conduct the matrix experiment
• Analyze the data to determine the optimum levels
  of design parameters and verify through
  confirmation experiments
• Case study (reading)

10
Quality Loss
11
A Framework for Quality Improvement

• Taguchi loss function L(y) k(y-T)2
• where k quality loss coefficient, a constant
• y quality characteristic of a product
• T target value of y
• The average quality loss perform n measurements
  of the quality characteristic y.

12
A Framework for Quality Improvement

• Common variations of loss functions
• Nominal-the-best type L(y) k(y T)2k A/d2
• Smaller-the-better type L(y) k(y)2k A/d2
• Larger-the-better type L(y) k(1/y2) k Ad2
• The AQS of the nominal-the-best type
• L(y) k?2 (? T)2
• The AQS of the smaller-the-better type
• L(y) k?2 (?)2
• The AQS of the larger-the-better type
• L(y) k(1/ ?2)(13?2/?2)
• Where d is functional limit and associated
  quality loss A
• Applications of quality loss function used as a
  decision support tool in a number of situations
  involving variability.
• Determining best factory tolerances
• Product selection

13
Example 1
14
Example 2

• Company ABC is planning to buy a couple of
  thousand bolts to be used in their systems. The
  system requires highly reliable bolts. In case of
  bolt failure the system repair cost is estimated
  to be 15.00. Two companies that use different
  kinds of alloys in their products bid to supply
  the bolts. Company ABC decides to go for
  destructive testing using 20 specimens. The
  criterion used for testing is the ultimate
  tensile strength measured in kgf/mm2. The test
  data for both products are given in the following
  table.
• The lower specification limit is 11kgf/mm2, and
  purchase quantity is 20,000. The unit costs of
  product A and B are 0.14 and 0.13,
  respectively. Advise ABC company about its
  purchasing decision?

15
Example 2
16
Failure Mode and Effect Analysis

• FMEA is a technique that can be used to identify
  failure modes and analyze their effects on the
  system performance with the objective of
  developing a preventive strategy.
• FMEA can be used at any stage of product life
  cycle
• Our focus process FMEA evaluate the process
  for possible ways in which failures can occur.
• Objective eliminate potential production failure
  effects
• Conducted during quality planning and before
  beginning production
• Probability of failure occurrence (1-10) 1
  remote chance of failing, 10 very high chance
  of failing
• Degree of failure severity (1-10) 1 not
  noticeable to the customer, 10 critical failure
• Probability of failure detection (1-10) 1
  extremely low chance of escaping defects 10
  very high chance of escaping defects

17
Failure Mode and Effect Analysis Steps

• Step 1 A cross functional team determines the
  potential failure mode of a design or process
• Step 2 FMEA forms are distributed to the teams
  participants and all items on the form are
  explained and discussed
• Step 3 Prioritize the problems (failure modes)
  to be studied based on
• RPN (risk priority number) occurrence x
  severity x detection
• Step 4 The FMEA matrix is completed and checked
  for accuracy. Finally, recommended action items
  are assigned
• Step 5 The finalized FMEA matrix is dated and
  presented to the process owner

18
Example
19
Improving Product Quality During the Production
Phase

• SPC is a monitoring process stability and
  improving process capability by reducing
  variability
• SPC tools
• Histogram
• Check sheet
• Pareto chart
• Cause-and-effect diagram
• Defect concentration diagram
• Scatter diagram
• Control chart

20
Histogram

• Construct a histogram number of classes
• k int13.3log10n n sample size

21
Pareto Chart

• Paretos Rule 80 of the problem is created from
  20 of the causes
• Allows the company to concentrate resources on
  the jobs with the most problems

22
Cause-and-Effect Diagram

• Identified problem or undesirable result is the
  head
• Contributing factors are the bones

23
Control Chart

• The variation in the outcome of a process chance
  (common) causes and assignable (special) causes
• Control charts are effective means of detecting
  the occurrence of assignable causes
• Elements of control chart y is a sample
  statistic that measures some quality
  characteristic of interest
• UCL ?y k?y
• CL ?y
• LCL ?y - k?y
• Where k (3) is the distance of control limits
  from the center line

24
Control Chart

• Variable Control chart quality characteristic ?
  continuous var., e.g. x-bar chart (central
  tendency), R chart (sample range), S chart
  (sample SD)
• Attribute control chart conforming/nonconforming
  to specifications. E.g. p chart, np chart, c
  chart, u chart
• Benefits of control charts
• effective means of monitoring statistical control
• Help predict the performance of a process
• Provide a common language for communication about
  the process
• Help direct corrective measures in a logical
  manner

25
Control Chart

• Step 1 Determine the type of variance control
  chart to be used (e.g. p chart, c chart, etc.)
• Step 2 Collect at least 20 samples of data, 5
  measurements per sample. Sampling should be
  random and according to a set frequency over a
  period of time
• Step 3 prepare a type chart and record
  collected data
• Step 4 after all 20 subgroups (samples) have
  been recorded, perform all required calculations
• Step 5 Plot and connect plotted points to draw
  trend lines. Verify that trend line points
  reflect recorded averages and range R
• Step 6 Analyze plotted data for significant
  variance or patterns

26
Example
27
Automated Inspection

• Use to collect data on the features and
  characteristics of products and processes
• Objective to assure product conformance to given
  specifications and to detect machine malfunctions
  ? correct the process
• Types of inspection equipments
  on-line/in-process,
• On-line/in-process is performed during the
  manufacturing operation
• On-line/post-process is performed immediately
  following the production process
• Off-line does not make any physical contact with
  machine tools CMM

28
Five Basic Types of CMMs
29
Five Variations of Basic CMM Types
30
Automated Inspection

• Machine vision systems perform three functions
• Image acquisition and digitization
• Image processing and analysis
• Interpretation of the data for inspection
  purposes
• Elements of a machine vision system