ISQA 572/ 449 Models for Quality Control/ Process Control and Improvement

1
ISQA 572/ 449Models for Quality Control/ Process
Control and Improvement

• Dr. David Raffo
• Tel [phone], Fax [phone]
• Email davidr_at_sba.pdx.edu

2
Agenda

• Announcements
• Questions
• HW1
• Control Charts (Variables) Cont.
• Quality Costs
• SPC Vs SQC (Inspections and Acceptance Sampling)
• Control Charts (Attributes)
• Process Capability

3
Quality Costs
4
Quality Costs
5
Quality Cost Traditional View TM2-5
6
Competitive Benefits of TQM Exhibit 2-8
7
SPC Vs SQC (Inspections and Acceptance Sampling)
8
Approaches to Quality TM 4-1
9
Quality Control Modes TM 4-2
10
Statistical Process Control Prevention
TM 4-3
11
Disadvantages of Inspection

• Wasteful
• Sampling and inspection add cost and decrease
  value
• Inaccurate
• Even 100 inspection is only 80 effective
  because of the possibility of human errors
• Impractical(Costly)
• Inspection may involve destructive testing

12
Disadvantages of Inspection

• Wrong message
• Inspection communicates to people and suppliers
  that bad parts will still be tolerated.
• Risks
• In sampling and inspection there is a risk of
  accepting bad lots and rejecting good lots
• No continuous improvement
• Sampling is still inspection, not prevention, so
  that quality is not typically continuously
  improved.

13
Advantages of a Stable Process

• Management and workers know the process
  capability and can predict performance, costs and
  quality levels.
• Productivity will be at a maximum and costs will
  be minimized.
• Management will be able to measure the effects of
  changes in the system with greater speed and
  reliability.

14
Advantages of a Stable Process

• If management wants to alter specification
  limits, it will have the data to back up its
  decision.
• (A stable process does not necessarily meet specs
  nor exhibit minimal variation - its just
  predictable)

15
Acceptance Sampling

• Acceptance sampling has three basic decisions
  accept, reject, or resample.
• Reason for using acceptance sampling
• Cost of passing defects is low
• Destructive testing is required
• Cost of inspection high relative to cost of loss
• Assumes stable process
• Large number of items must be processed in a
  short time

16
Acceptance Sampling

• Terms
• Producers Risk (?) Risk of rejecting a lot
  with acceptable quality level. (type I error)
• Consumers Risk (?) Risk of accepting a lot
  with unacceptable quality level. (type II error)
• Acceptable Quality Level (AQL) The maximum
  percentage defective that can be considered
  satisfactory.
• Lot Tolerance Percent Defective (LTPD) The
  percent defective where the consumer desires the
  probability of acceptance to be at a low level.

17
Acceptance Sampling - Attributes

• Types of plans
• Single
• N, n, c
• (1000, 50, 1)
• Double
• N, n1, n2, c1, c2 , c3
• (3000, 50, 80, 1, 3, 5)
• Sequential
• n, ca, cr
• (50, 0, 4) (50, 1, 5)

18
Acceptance Sampling - Attributes

• Measures
• Average Outgoing Quality (AOQ)
• Average Total Inspection
• Average Sample Number
• Standard Sampling Plans
• MIL-STD-105E
• Dodge-Romig
• Chain Sampling
• Skip-Lot
• Deming kp

19
Acceptance Sampling - Variables

• Advantages
• Smaller sample than equivalent attribute plan
• Provides more information
• Provides insight into quality improvements
• Disadvantages
• Separate plan for each variable
• Inspection costs are higher
• Distribution estimate required

20
Acceptance Sampling - Variables

• Process Parameter
• Average quality of the product/process or
  variability of the quality is known
• Single Specification
• n Xa
• Double Specification
• n, XLa, XUa
• Lot Proportion Nonconforming
• Form 1 (k-method)
• Form 2 (M-method)

21
Control Charts (Attributes)
22
Advantages Disadvantages of Attribute Charts

• Advantages
• Some quality characteristics can only be viewed
  as a attribute.
• Quality characteristic may be measurable as a
  variable but an attribute is used for time, cost
  or convenience.
• Combination of variables can be measured as an
  attribute rather than use a multivariate chart.

23
Advantages Disadvantages of Attribute Charts

• Disadvantages
• Attributes dont measure the degree to which
  specifications are met or not met.
• Doesnt provide much information on cause.
• Variable chart can indicate potential changes
  which allow preventive actions.
• Larger sample size required.

24
Types of Attribute Charts

• p-Chart - Fraction Nonconforming
• Can have constant or variable sample size.
• Good tool for relating information about average
  quality level.
• np-Chart - Number of Nonconforming
• Number of nonconforming items may be easier for
  user to understand.

25
Types of Attribute Charts

• c-Chart - Number of Nonconformities
• Used when desire is to control the number of
  defects where one defect may not cause the entire
  product to be defective.
• Often used where area of opportunity is
  continuous and a constant size

26
Types of Attribute Charts

• u-Chart - Number of Nonconformities/unit
• Area of opportunity is of variable size.
• U-Chart - Number of Demerits/unit
• Allows the use of variable weights for different
  classes of defects.

27
p Chart
TM 4-12
28
p Chart
TM 4-13
29
p Chart
TM 4-14
30
p-Chart
Exhibit 4-26
31
Hotel Suite Inspection - Defects Discovered
Exhibit 4-27
32
C-Chart Calculations

• Centerline c-bar (S c)/m sub-groups
• UCLc c-bar 3sqrt(c-bar)
• LCLc c-bar - 3sqrt(c-bar)

33
c Chart for Hotel Suite Inspection Exhibit
4-28
34
Process Capability
35
Process Capability Analysis

• Creates uniformity of output
• Level of quality is maintained or improved
• Facilitates product and process design
• Assists in supplier selection and control
• Reduces total costs

36
Process Capability Normal Curve TM 4-15
37
Process Capability TM 4-17
38
Capability Indexes

• Cp
• Ability to meet two-sided specification limits
• Cp (USL-LSL)/(6? )
• Assumes
• Stable process
• Normal distribution
• Variables data
• Centered process
• Goal Cpgt1.0

39
Capability Indexes

• Capability Ratio
• CR (6?)/(USL-LSL)
• Poor if CRgt1

40
Capability Indexes

• CPU CPL
• Ability to meet one-sided specification limit
• CPU (USL-X)/(3?)
• CPL (X-LSL)/(3?)
• Assumes
• Stable process
• Normal distribution
• Variables data

41
Process Capability Chart Exhibit
4-20
42
Capability Indexes

• Cpk
• Ability to meet two sided specification but the
  process does not have to be centered
• Cpk Cp - m-X/(3?) where mnominal
  centerline

43
Process Capability Index TM
4-18
44
Process Capability Varieties
TM 4-19
45
(No Transcript)