Business Statistics: A Decision-Making Approach: Introduction To Quality and Statistical Process Control

Business Statistics:
A Decision-Making Approach
6th Edition
Chapter 17
Introduction to Quality and
Statistical Process Control
Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-1

Chapter Goals
After completing this chapter, you should be

able to:
 Use the seven basic tools of quality
 Construct and interpret x-bar and R-charts
 Construct and interpret p-charts
 Construct and interpret c-charts

Chapter Overview
Quality Management and
Tools for Improvement
Philosophy of Tools for Quality

Quality Improvement
Deming’s 14 The Basic Control

Points 7 Tools Charts
Juran’s 10 X-bar/R-charts
Steps to p-charts
Quality
c-charts
Improvement
Themes of Quality Management
 Primary focus is on process improvement

 Most variations in process are due to systems
 Teamwork is integral to quality management
 Customer satisfaction is a primary goal
 Organization transformation is necessary
 It is important to remove fear
 Higher quality costs less

Deming’s 14 Points
 1. Create a constancy of purpose toward
improvement
 become more competitive, stay in business, and provide jobs
 2. Adopt the new philosophy
 Better to improve now than to react to problems later
 3. Stop depending on inspection to achieve
quality -- build in quality from the start
 Inspection to find defects at the end of production is too late
 4. Stop awarding contracts on the basis of
low bids
 Better to build long-run purchaser/supplier relationships
(continued)
 5. Improve the system continuously to improve
quality and thus constantly reduce costs
 6. Institute training on the job
 Workers and managers must know the difference between
common cause and special cause variation
 7. Institute leadership
 Know the difference between leadership and supervision
 8. Drive out fear so that everyone may work
effectively.
 9. Break down barriers between departments
so that people can work as a team.
(continued)
 10. Eliminate slogans and targets for the
workforce
 They can create adversarial relationships
 11. Eliminate quotas and management by
objectives
 12. Remove barriers to pride of workmanship
 13. Institute a vigorous program of education
and self-improvement
 14. Make the transformation everyone’s job

Juran’s 10 Steps to Quality
Improvement
 1. Build awareness of both the need for
improvement and the opportunity for
improvement
 2. Set goals for improvement
 3. Organize to meet the goals that have been
set
 4. Provide training
 5. Implement projects aimed at solving
problems

Juran’s 10 Steps to Quality
Improvement
(continued)
 6. Report progress
 7. Give recognition
 8. Communicate the results
 9. Keep score
 10. Maintain momentum by building
improvement into the company’s regular
systems

The Deming Cycle
Plan
The
Act Deming Do
Cycle
The key is a
continuous
cycle of
Study improvement
The Basic 7 Tools
1. Process Flowcharts
2. Brainstorming
3. Fishbone Diagram
4. Histogram
5. Trend Charts
6. Scatter Plots
7. Statistical Process
Control Charts

The Basic 7 Tools
(continued)
1. Process Flowcharts Map out the process to better

2. Brainstorming visualize and understand
3. Fishbone Diagram opportunities for improvement
4. Histogram
5. Trend Charts
6. Scatter Plots
Control Charts

The Basic 7 Tools
(continued)
1. Process Flowcharts Fishbone (cause-and-effect) diagram:

2. Brainstorming
Cause 1 Cause 2
3. Fishbone Diagram
4. Histogram Sub-causes
5. Trend Charts
6. Scatter Plots Problem
7. Statistical Process Sub-causes
Control Charts
Show patterns of variation Cause 3 Cause 4

The Basic 7 Tools
(continued)
1. Process Flowcharts Identify trend

2. Brainstorming y
3. Fishbone Diagram
4. Histogram
5. Trend Charts
6. Scatter Plots
7. Statistical Process time
Control Charts
Examine relationships
y
x
The Basic 7 Tools
(continued)
1. Process Flowcharts Examine the performance

2. Brainstorming of a process over time
3. Fishbone Diagram
4. Histogram X
5. Trend Charts
6. Scatter Plots
Control Charts
time

Introduction to Control Charts
 Control Charts are used to monitor variation in

a measured value from a process
 Exhibits trend
 Can make correction before process is out of control
 A process is a repeatable series of steps
leading to a specific goal
 Inherent variation refers to process variation
that exists naturally. This variation can be
reduced but not eliminated

Process Variation
Total Process Common Cause Special Cause

Variation = Variation + Variation
 Variation is natural; inherent in the world

around us
 No two products or service experiences
are exactly the same
 With a fine enough gauge, all things can
be seen to differ

Sources of Variation

Variation is often due to differences in:

 People
 Machines
 Materials
 Methods
 Measurement
 Environment
Common Cause Variation

Common cause variation

 naturally occurring and expected
 the result of normal variation in
materials, tools, machines, operators,
and the environment

Special Cause Variation

Special cause variation

 abnormal or unexpected variation
 has an assignable cause
 variation beyond what is considered
inherent to the process

Statistical Process Control Charts
 Show when changes in data are due to:

 Special or assignable causes
 Fluctuations not inherent to a process
 Represents problems to be corrected
 Data outside control limits or trend
 Common causes or chance

 Inherent random variations
 Consist of numerous small causes of random
variability

Control Chart Basics
Special Cause Variation:

Range of unexpected variability
UCL
Common Cause +3σ
Variation: range of Process Average
expected variability - 3σ
LCL
time
UCL = Process Average + 3 Standard Deviations
LCL = Process Average – 3 Standard Deviations
Process Variability
Special Cause of Variation:
A measurement this far from the process average
is very unlikely if only expected variation is present
UCL
±3σ → 99.7% of
process values
Process Average
should be in this
range
LCL
time
Statistical Process Control Charts
Statistical
Process Control
Charts
X-bar charts
p-charts c-charts
and R-charts
Used for Used for Used for

measured proportions number of
numeric data (attribute data) attributes per
sampling unit

x-bar chart and R-chart
 Used for measured numeric data from a

process
 Start with at least 20 subgroups of observed
values
 Subgroups usually contain 3 to 6 observations
each

Steps to create an x-chart
and an R-chart
 Calculate subgroup means and ranges

 Compute the average of the subgroup
means and the average range value
 Prepare graphs of the subgroup means
and ranges as a line chart

Steps to create an x-chart
and an R-chart
(continued)
 Compute the upper and lower control limits

for the x-bar chart
for the R-chart
 Use lines to show the control limits on the
x-bar and R-charts

Example: x-chart
 Process measurements:
Subgroup measures
Subgroup
number
Individual measurements Mean, x Range, R
1 15 17 15 11 14.5 6
2 12 16 9 15 13.0 7
3 17 21 18 20 19.0 4
… … … … … … …
Average Average
subgroup subgroup
mean = x range = R
Average of Subgroup
Means and Ranges
Average of Average of
subgroup means: subgroup ranges:
x
 x i
R
 R i
k k
where: where:
xi = ith subgroup average Ri = ith subgroup range
k = number of subgroups k = number of subgroups

Computing Control Limits
 The upper and lower control limits for an x-chart
are generally defined as
or
UCL  x  3

LCL  x  3

(continued)
 Since control charts were developed before it
was easy to calculate σ, the interval was
formed using R instead
 The value A2R is used to estimate 3σ , where
A2 is from Appendix Q
 The upper and lower control limits are
UCL  x  A 2 (R ) where A2 = Shewhart
factor for subgroup size
LCL  x  A 2 (R ) n from appendix Q

Example: R-chart
 The upper and lower control limits for an

R-chart are
UCL  D4 (R )
LCL  D3 (R )
where:
D4 and D3 are taken from the Shewhart table
(appendix Q) for subgroup size = n

x-chart and R-chart
UCL
x-chart x
LCL
time
UCL
R-chart R
LCL
time
Using Control Charts
 Control Charts are used to check for process
control
H0: The process is in control
i.e., variation is only due to common causes
HA: The process is out of control
i.e., special cause variation exists
 If the process is found to be out of control,
steps should be taken to find and eliminate the
special causes of variation

Process In Control
 Process in control: points are randomly
distributed around the center line and all
points are within the control limits
x
UCL
x
LCL
time

Process Not in Control
Out of control conditions:

 One or more points outside control limits
 Nine or more points in a row on one side of
the center line
 Six or more points moving in the same
direction
 14 or more points alternating above and
below the center line
Process Not in Control
 One or more points outside  Nine or more points in a row
control limits on one side of the center line
UCL UCL
x x
LCL LCL
 Six or more points moving in  14 or more points alternating

the same direction above and below the center line
UCL UCL
x x
LCL LCL
Out-of-control Processes
 When the control chart indicates an out-of-
control condition (a point outside the control
limits or exhibiting trend, for example)
 Contains both common causes of variation and
assignable causes of variation
 The assignable causes of variation must be identified
 If detrimental to the quality, assignable causes of
variation must be removed

 If increases quality, assignable causes must be
incorporated into the process design

p-Chart
 Control chart for proportions

 Is an attribute chart
 Shows proportion of nonconforming items
 Example -- Computer chips: Count the number of
defective chips and divide by total chips inspected
 Chip is either defective or not defective
 Finding a defective chip can be classified a
“success”

p-Chart
(continued)
 Used with equal or unequal sample sizes

(subgroups) over time
 Unequal sizes should not differ by more than ±25%
from average sample sizes
 Easier to develop with equal sample sizes
 Should have np > 5 and n(1-p) > 5

Creating a p-Chart
 Calculate subgroup proportions
 Compute the average of the subgroup
proportions
 Prepare graphs of the subgroup
proportions as a line chart
 Use lines to show the control limits on the
p-chart

p-Chart Example
Subgroup Sample Number of

Proportion, p
number size successes
1 150 15 10.00
2 150 12 8.00
3 150 17 11.33
… … …
Average
subgroup
proportion = p

Average of Subgroup Proportions
The average of subgroup proportions = p

If equal sample sizes: If unequal sample sizes:
p
 p i
p
 np i i
k n i
where: where:
pi = sample proportion for subgroup i ni = number of items in sample i
k = number of subgroups of size n ni = total number of items
sampled in k samples
 The upper and lower control limits for an p-chart
are
UCL = Average Proportion + 3 Standard Deviations
LCL = Average Proportion – 3 Standard Deviations
 or UCL  p  3
LCL  p  3

Standard Deviation of
Subgroup Proportions
 The estimate of the standard deviation for
the subgroup proportions is
If equal sample sizes: If unequal sample sizes:
Generally, s p is
( p )(1  p )
sp  computed separately
n for each different
sample size
where:
p = mean subgroup proportion
n = common sample size
(continued)
 The upper and lower control limits for the

p-chart are
UCL  p  3(s p ) Proportions are
never negative, so
if the calculated
LCL  p  3(s p ) lower control limit
is negative, set
LCL = 0
 If sample sizes
( p )(1  p )
are equal, this UCL  p  3
becomes n
( p )(1  p )
LCL  p  3
n
p-Chart Examples
 For equal  For unequal
sample sizes sample sizes
UCL UCL
p p
LCL LCL
s p is constant since s p varies for each

n is the same for subgroup since
all subgroups ni varies
c-Chart
 Control chart for number of nonconformities
(occurrences) per sampling unit (an area of
opportunity)
 Also a type of attribute chart
 Shows total number of nonconforming items
per unit
 examples: number of flaws per pane of glass
number of errors per page of code
 Assume that the size of each sampling unit
remains constant
Mean and Standard Deviation
for a c-Chart
 The mean for a  The standard deviation

c-chart is for a c-chart is
c
 x i
s c
k
where:
xi = number of successes per sampling unit
k = number of sampling units

c-Chart Control Limits
The control limits for a c-chart are
UCL  c  3 c
LCL  c  3 c

Process Control
Determine process control for p-chars and c-charts

using the same rules as for x-bar and R-charts
Out of control conditions:
 One or more points outside control limits
 Nine or more points in a row on one side of the center line
 Six or more points moving in the same direction
 14 or more points alternating above and below the center
line

c-Chart Example
 A weaving machine makes

cloth in a standard width.
Random samples of 10 meters
of cloth are examined for flaws.
Is the process in control?
Sample number 1 2 3 4 5 6 7
Flaws found 2 1 3 0 5 1 0

Constructing the c-Chart
 The mean and standard deviation are:
c
 x i

2  1 3  0  5  1 0
 1.7143
k 7
s  c  1.7143  1.3093
 The control limits are:

UCL  c  3 c  1.7143  3(1.3093)  5.642
LCL  c  3 c  1.7143  3(1.3093)  2.214
Note: LCL < 0 so set LCL = 0
The completed c-Chart
6 UCL = 5.642
5
4
3
2
c = 1.714
1
0 LCL = 0
1 2 3 4 5 6 7
Sample number
The process is in control. Individual points are distributed around

the center line without any pattern. Any improvement in the
process must come from reduction in common-cause variation
Chapter Summary
 Reviewed the philosophy of quality management
 Demings 14 points
 Juran’s 10 steps
 Described the seven basic tools of quality
 Discussed the theory of control charts
 Common cause variation vs. special cause variation
 Constructed and interpreted x-bar and R-charts
 Constructed and interpreted p-charts
 Constructed and interpreted c-charts

Business Statistics: A Decision-Making Approach: Introduction To Quality and Statistical Process Control

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Business Statistics: A Decision-Making Approach: Introduction To Quality and Statistical Process Control

Uploaded by

Copyright:

Available Formats

Business Statistics:

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-1

After completing this chapter, you should be

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-2

Philosophy of Tools for Quality

Deming’s 14 The Basic Control

 Primary focus is on process improvement

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-4

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-7

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-8

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-9

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-11

1. Process Flowcharts Map out the process to better

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-12

1. Process Flowcharts Fishbone (cause-and-effect) diagram:

Show patterns of variation Cause 3 Cause 4

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-13

1. Process Flowcharts Identify trend

1. Process Flowcharts Examine the performance

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-15

 Control Charts are used to monitor variation in

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-16

Total Process Common Cause Special Cause

 Variation is natural; inherent in the world

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-17

Total Process Common Cause Special Cause

Variation is often due to differences in:

Total Process Common Cause Special Cause

Common cause variation

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-19

Total Process Common Cause Special Cause

Special cause variation

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-20

 Show when changes in data are due to:

 Represents problems to be corrected

 Data outside control limits or trend

 Common causes or chance

 Consist of numerous small causes of random

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-21

Special Cause Variation:

Used for Used for Used for

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-24

 Used for measured numeric data from a

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-25

 Calculate subgroup means and ranges

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-26

 Compute the upper and lower control limits

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-27

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-29

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-30

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-31

 The upper and lower control limits for an

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-32

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-34

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-35

Out of control conditions:

 Six or more points moving in  14 or more points alternating

variation must be removed

incorporated into the process design

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-38

 Control chart for proportions

Business Statistics: A Decision-Making Approach, 6e © 2005 Prentice-Hall, Inc. Chap 17-39

 Used with equal or unequal sample sizes