Quality

Improvement
Chapter 6- Control
Charts for Variables

PowerPoint presentation to accompany

Besterfield, Quality Improvement, 9e
 Outline
❑ The Control Chart Techniques
❑ State of Introduction
❑ Control
❑ Specifications
❑ Process Capability
❑ Different Control Charts

2
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Learning Objectives
When you have completed this chapter you should:
❑ Know the three categories of variation and their
sources.
❑ Understand the concept of the control chart method.
❑ Know the purpose of variable control charts.
❑ Know how to select the quality characteristics, the
rational subgroup and the method of taking samples

3
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Learning Objectives
When you have completed this chapter you should:
❑ Be able to calculate the central value, trial control
limits and the revised control limits for Xbar and
R chart.
❑ Be able to explain what is meant by a process in
control and the various out-of-control patterns.
❑ Know the difference between individual measurements
and averages; control limits and specifications.

4
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Learning Objectives
When you have completed this chapter you should:
❑ Know the different situations between the process
spread and specifications and what can be done to
correct the undesirable situation.
❑ Be able to calculate process capability.

5
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 variation
 The variation concept is a law of nature in
that no two natural items are the same.
❑ The variation may be quite large and easily
noticeable
❑ The variation may be very small. It may
appear that items are identical; however,
precision instruments will show difference
❑ The ability to measure variation is necessary
before it can be controlled
6
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Variation
There are three categories of variation in piece
part production:
1. Within-piece variation: Surface
2. Piece-to-piece variation: Among pieces
produced at the same time
3. Time-to-time variation: Difference in product
produced at different times of the day

7
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Variation
Sources of Variation in production processes:
Measurement
Operators Methods
Materials Instruments

INPUTS PROCESS OUTPUTS

Tools Human
Machines Environment Inspection
Performance

8
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Variation
Sources of variation are:
1. Equipment:

1. Toolwear

2. Machine vibration

3. Electrical fluctuations etc.

2. Material

1. Tensile strength

2. Ductility

3. Thickness

4. Porosity etc.

9
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Variation
Sources of variation are:
3. Environment
1. Temperature
2. Light
3. Radiation
4. Humidity etc.
4. Operator
1. Personal problem
2. Physical problem etc.

10
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Charts
❑ Variable data
❑ x-bar and R-charts
❑ x-bar and s-charts
❑ Charts for individuals (x-charts)
❑ Attribute data
❑ For “defectives” (p-chart, np-chart)
❑ For “defects” (c-chart, u-chart)

12
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Charts

Continuous Categorical or Discrete

Numerical Data Control Numerical Data
Charts

Variables Attributes
Charts Charts

R X P C
Chart Chart Chart Chart

13
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Charts for Variables
The control chart for variables is a means of
visualizing the variations that occur in the
central tendency and the mean of a set of
observations. It shows whether or not a
process is in a stable state.

14
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  Each small solid circle represents the average value
within a subgroup. Thus, subgroup number 5
consists of, say, four observations, 3.46, 3.49,
3.45, and 3.44, and their average is 3.46 kg. This
value is the one posted on the chart
for subgroup number 5. As shown in Figure 6-1

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Charts

Figure 6-1 Example of a control chart

16
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  The solid line in the center of the chart can have
three different interpretations, depending on the
available data.
First, and most commonly, it can be the average of
the plotted points, which in the case of an X chart
is the average of the averages or “X-double bar,” X.
Second, it can be a standard or reference value,
X0, based on representative prior data, an
economic value based on production costs
or service needs
 Third, it can be the population mean, µ, if that
value is known.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Charts

Figure 6-1 Example of a method of reporting

Quality Improvement, 9e
inspection
© 2013, results
2008 by Pearson
18
Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Variable Control Charts
The objectives of the variable control charts are:
1. For quality improvement
2. To determine the process capability
3. For decisions regarding product specifications
4. For current decisions on the production process
5. For current decisions on recently produced items

19
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Chart Techniques
Procedure for establishing a pair of control charts for
the average Xbar and the range R:
1. Select the quality characteristic
2. Choose the rational subgroup
3. Collect the data
4. Determine the trial center line and control limits
5. Establish the revised central line and control limits
6. Achieve the objective

20
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Quality Characteristic
The Quality characteristic must be measurable.
It can expressed in terms of the seven basic units:
1. Length

2. Mass

3. Time

4. Electrical current

5. Temperature

6. Subatance

7. Luminosity

21
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Rational Subgroup
A rational subgroup is one in which the variation
within a group is due only to chance causes.
Within-subgroup variation is used to determine the
control limits.
Variation between subgroups is used to evaluate
long-term stability.

22
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Rational Subgroup
There are two schemes for selecting the subgroup
samples:
1. Select subgroup samples from product or
service produced at one instant of time or as
close to that instant as possible
2. Select from product or service produced over a
period of time that is representative of all the
products or services

23
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Rational Subgroup
The first scheme will have a minimum variation
within a subgroup.
The second scheme will have a minimum variation
among subgroups.
The first scheme is the most commonly used since
it provides a particular time reference for
determining assignable causes.
The second scheme provides better overall results
and will provide a more accurate picture of the
quality.

24
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Subgroup Size
❑ As the subgroup size increases, the control
limits become closer to the central value,
which make the control chart more sensitive
to small variations in the process average
❑ As the subgroup size increases, the inspection
cost per subgroup increases
❑ When destructive testing is used and the item
is expensive, a small subgroup size is required

25
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Subgroup Size
❑ From a statistical basis a distribution of
subgroup averages are nearly normal for
groups of 4 or more even when samples are
taken from a non-normal distribution
❑ When a subgroup size of 10 or more is used,
the s chart should be used instead of the R
chart.
❑ See Table 6-1 for (total) sample sizes

26
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Data Collection
Data collection can be accomplished using the
type of figure shown in Figure 6-2.
It can also be collected using the method in
Table 6-2.
It is necessary to collect a minimum of 25
subgroups of data.
A run chart can be used to analyze the data in
the development stage of a product or prior to
a state of statistical control
28
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Run Chart

Figure 6-4 Run Chart for data of Table 6-2

29
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Trial Central Lines
Central Lines are obtained using:
g g

X i R i
X = i =1
and R = i =1
g g
where
X = average of subgroup averages
X i = average of the ith subgroup
g = number of subgroups
R = average of subgroup ranges
Ri = range of the ith subgroup

30
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Trial Control Limits
Trial control limits are established at ±3 standard
deviatons from the central value
UCLX = X + 3 X UCLR = R + 3 R
LCLX = X − 3 X LCLR = R − 3 R
where
UCL=upper control limit
LCL=lower control limit
 X = population standard deviation of the subgroup averages
 R = population standard deviation of the range

33
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Trial Control Limits
In practice calculations are simplified by using
the following equations where A2,D3 and D4 are
factors that vary with the subgroup size and
are found in Table B of the Appendix.

UCLX = X + A2 R UCLR = D4 R
LCLX = X − A2 R LCLR = D3 R

34
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Trial Control Limits

Figure 6-5 Xbar and R chart for preliminary data with trial control limits
35
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Revised Central Lines

X new =  X−X d
and R new =  R−R d

g − gd g − gd
where
X d = discarded subgroup averages
g d = number of discarded subgroups
Rd = discarded subgroup ranges

36
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Standard Values
R0
X 0 = X new R0 = R new and  0 =
d2
UCLX = X 0 + A 0 UCLR = D2 0
LCLX = X 0 − A 0 LCLR = D1 0

37
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Figure 6-6 Trial control limits and revised control limits for Xbar and R charts 38
© 2013, 2008 by Pearson Higher Education, Inc
Quality Improvement, 9e
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Achieve the Objective

Figure 5-7 Continuing use of control charts, showing improved quality

39
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Revised Central Lines

X new =  X−X d
and R new =  R−R d

g − gd g − gd
where
X d = discarded subgroup averages
g d = number of discarded subgroups
Rd = discarded subgroup ranges

40
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 The Sample Standard Deviation Control Chart

 Some organizations prefer the sample

standard deviation, s, as the measure of
the subgroup dispersion.
 R chart is easier to compute and easier
to explain. On the other hand, the
subgroup sample standard deviation for
the s chart is calculated using all the
data rather than just the high and the
low value, as done for the R chart.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 The Sample Standard Deviation Control Chart
 An s chart is therefore more accurate than an
R chart.
 When subgroup sizes are less than
10, both charts will graphically portray the
same variation;
 as subgroup sizes increase to 10 or more,
extreme values have an improper influence on
the R chart. Therefore, at larger subgroup
sizes, the s chart must be used.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Sample Standard Deviation
Control Chart
For subgroup sizes >=10, an s chart is more
accurate than an R Chart.Trial control limits are
given by:

 
g g
s
i =1 i
Xi
s= X = i =1
g g
UCLX = X + A3 s UCLs = B4 s
LCLX = X − A3 s LCLs = B3 s
43
 Revised Limits for s chart

X0 = X new =
X − Xd
g − gd

s0 = snew =
s − s d
0 =
s0
g − gd c4
UCLX = X 0 + A 0 UCLs = B6 0
LCLX = X 0 − A 0 LCLs = B5 0
where
sd = discarded subgroup averages
c4 , A, B5 , B6 = factors found in Table B

44
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  The first step is to determine the standard
deviation for each subgroup from the preliminary
data. For subgroup 1, with values of 6.35, 6.40,
6.32, and 6.37, the standard deviation is

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  Example Problem 6-4
Using the data of Table 6-3, determine the revised
central line and control limits for X and s charts.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  The next step is to plot the subgroup (X bar )and s
on graph paper with the central lines and control
limits.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 State of Control
Process in Control
❑ When special causes have been eliminated
from the process to the extent that the points
plotted on the control chart remain within the
control limits, the process is in a state of
control
❑ When a process is in control, there occurs a
natural pattern of variation

53
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 State of Control

Figure 6-9 Natural pattern of variation of a control chart

54
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 State of Control
Types of errors:
❑ Type I, occurs when looking for a special
cause of variation when in reality a common
cause is present
❑ Type II, occurs when assuming that a common
cause of variation is present when in reality
there is a special cause

55
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 State of Control
When the process is in control:
1. Individual units of the product or service will
be more uniform
2. Since the product is more uniform, fewer
samples are needed to judge the quality
3. The process capability or spread of the
process is easily attained from 6ơ
4. Trouble can be predicted before it occurs

56
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 State of Control
When the process is in control:
5. The % of product that falls within any pair of
values is more predictable
6. It allows the consumer to use the producer’s
data
7. It is an indication that the operator is
performing satisfactorily

57
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Common
Causes

Special
Causes

45
 State of Control

Figure 6-11 Frequency Distribution of subgroup averages with control limits

59
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 State of Control
When a point (subgroup value) falls outside its
control limits, the process is out of control.
Out of control means a change in the process due
to a special or assignable cause.A process can
also be considered out of control even when the
points fall inside the 3ơ limits

60
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 State of Control
❑ It is not natural for seven or more consecutive
points to be above or below the central line.
❑ Also when 10 out of 11 points or 12 out of 14
points are located on one side of the central
line, it is unnatural.
❑ Six points in a row are steadily increasing or
decreasing indicate an out of control situation

61
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Patterns in Control Charts

Figure 6-12 Some unnatural runs-process out of control 62

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Patterns in Control Charts

Figure 6-13 Simplified rule for out-of-control pattern

64
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Out-of-Control Condition
1. Change or jump in level.
2. Trend or steady change in level
3. Recurring cycles
4. Two populations (also called mixture)
5. Mistakes

65
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Out-of-Control Patterns

Change or jump inlevel Trend or steady change in level

Recurring cycles Two populations

66
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Specifications

Figure 5-18 Comparison of individual values compared to averages

67
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Specifications
Calculations of the average for both the individual
values and for the subgroup avergaes are the
same. However the sample standard deviation is
different.

X =
n
where
 X = population standard deviation of subgroup averages
 = population standard deviation of individual values
n=subgroup size
If we assume normality, then the population standard deviation
s
can be estimated from  =
c4
68
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Central Limit Theorem
If the population from which samples are taken
is not normal, the distribution of sample
averages will tend toward normality provided
that the sample size, n, is at least 4. This
tendency gets better and better as the sample
size gets larger. The standardized normal can
be used for the distribution averages with the
modification.
X − X −
Z= =
X  n
69
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Central Limit Theorem

Figure 6-19 Illustration of central limit theorem

70
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  Example illustrates the Central Limit Theorem:

 : Exam Scores

 Imagine you're a teacher and you're collecting

exam scores from your students. The exam scores
come from various students with different levels of
preparation, understanding, and performance. The
scores might be influenced by a variety of factors,
such as study habits, test difficulty, and external
distractions.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  Example illustrates the Central Limit Theorem:

 : Exam Scores
 Let's say you're interested in the average exam
score of all students in your school. You collect
data from different classes, each with its own set of
scores. The scores might not follow a perfectly
normal distribution, as some students might
perform exceptionally well or poorly due to
individual circumstances.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  Example illustrates the Central Limit Theorem:
 : Exam Scores

 However, when you calculate the average exam

score for each class and then calculate the average
of those class averages, you start to see a pattern
emerge. As you collect more and more class
averages, you notice that their distribution starts to
resemble a normal distribution, even if the
individual scores within each class did not follow a
normal distribution.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
  Example illustrates the Central Limit Theorem:
 : Exam Scores

 This phenomenon is a result of the Central Limit

Theorem. The theorem states that regardless of
the shape of the original distribution (in this case,
individual student scores), as long as you have a
sufficiently large number of classes (sample size)
and the classes are independent and drawn from
the same population, the distribution of the class
averages will tend to become more normal.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Central Limit Theorem

Figure 6-20 Dice illustration of central limit theorem

75
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Limits & Specifications
❑ The control limits are established as a function
of the average
❑ Specifications are the permissible variation in
the size of the part and are, therefore, for
individual values
❑ The specifications or tolerance limits are
established by design engineers to meet a
particular function

76
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Control Limits & Specifications

Figure 6-21 Relationship of limits, specifications, and distributions

77
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability & Tolerance
❑ The process spread will be referred to as the
process capability and is equal to 6σ
❑ The difference between specifications is called
the tolerance
❑ When the tolerance is established by the design
engineer without regard to the spread of the
process, undesirable situations can result

78
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability & Tolerance
Three situations are possible:
❑ Case I: When the process capability is less
than the tolerance 6σ<USL-LSL
❑ Case II: When the process capability is equal
to the tolerance 6σ=USL-LSL
❑ Case III: When the process capability is
greater than the tolerance 6σ >USL-LSL

79
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability & Tolerance
Case I: When the process capability is less than
the tolerance 6σ<USL-LSL

Figure 6-24 Case I 6σ<USL-LSL

80
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability & Tolerance
Case II: When the process capability is equal to
the tolerance 6σ=USL-LSL

Figure 6-24 Case I 6σ=USL-LSL

81
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability & Tolerance
Case III: When the process capability is less than
the tolerance 6σ>USL-LSL

Figure 6-24 Case I 6σ>USL-LSL

82
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability
❑ The range over which the natural variation of
a process occurs as determined by the
system of common or random causes
❑ Measured by the proportion of output that
can be produced within design specifications

83
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability
This following method of calculating the process
capability assumes that the process is stable or in
statistical control:
❑ Take 25 (g) subgroups of size 4 for a total
of 100 measurements
❑ Calculate the range, R, for each subgroup
❑ Calculate the average range, RBar= ΣR/g
❑ Calculate the estimate of the population
R
standard deviation  = 0
d 2

❑ Process capability will equal 6σ0

84
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Process Capability
The process capability can also be obtained by using
the standard deviation:
❑ Take 25 (g) subgroups of size 4 for a total of
100 measurements
❑ Calculate the sample standard deviation, s, for
each subgroup
❑ Calculate the average sample standard
deviation, sbar = Σs/g
❑ Calculate the estimate of the population
standard deviation 0 =
s

❑Process capability will equal 6σo

c4

85
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Capability Index
Process capability (6 ơ) and tolerance are
combined to form the capability index.

USL − LSL
Cp =
6 0
where C p = capabilityindex
USL − LSL = tolerance
6 0 = process capability

86
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Capability Index
The capability index does not measure process
performance in terms of the nominal or target
value. This measure is accomplished by Cpk.

Min{(USL − X ) or ( X − LSL)
C pk =
3
Cpk: (Process
where C p = capabilityindex Capability
USL − LSL = tolerance Index - with
6 0 = process capability Centering
87
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Capability Index

Cp = USL - LSL
6 ơo

The Capability Index does not measure

process performance in terms of the
nominal or target

Min{(USL − X ) or ( X − LSL)
C pk =
3
88

Dale H. Besterfield where C p = capabilityindex

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Capability Index
1. The Cp value does not change as the process
center changes
2. Cp=Cpk when the process is centered
3. Cpk is always equal to or less than Cp
4. A Cpk = 1 indicates that the process is
producing product that conforms to
specifications
5. A Cpk < 1 indicates that the process is
producing product that does not conform to
specifications
89
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Capability Index

6. A Cp < 1 indicates that the process is not

capable
7. A Cpk=0 indicates the average is equal
to one of the specification limits
8. A negative Cpk value indicates that the
average is outside the specifications

90
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Cpk Measures

Cpk = negative number

Cpk = zero

Cpk = between 0 and 1

Cpk = 1

Cpk > 1

91
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Example Problem 6-5

Location pins for workholding devices are ground to a

diameter of 12.50 mm (approximately 1/2 in.), with a
tolerance of ;0.05 mm. If the process is centered at 12.50
mm (m) and the dispersion is 0.02 mm (s), what percent
of the product must be scrapped and what percent can
be reworked? How can the process center be changed to
eliminate the scrap? What is the rework percentage?

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 what percent
of the
product must
be scrapped
and what
percent can
be reworked?

From Table A of the Appendix, for a Z value of -2.50:

Area1 = 0.0062 or 0.62% scrap
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 How can the process center be changed
to eliminate the scrap?

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 What is the
rework
percentage?

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 In Example Problem 6-1 for the X and R charts, the
quality improvement process began
in January with s0 = 0.038. The process capability
is 6s = (6)(0.038) = 0.228 mm.
By July, s0 = 0.030, which gives a process capability
of 0.180 mm.
This is a 20% improvement in the process
capability,
improvement= 0.18/0.228= 80%

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Ex. 11
Plastic strips that are used in a sensitive electronic
device are manufactured to a maximum specification
of 305.70 mm (approximately 12 in.) and a minimum
specification of 304.55 mm. If the strips are less than
the minimum specification, they are scrapped; if they
are greater than the maximum specification, they are
reworked. The part dimensions are normally distributed
with a population mean of 305.20 mm and a standard
deviation of 0.25 mm. What percentage of the product is
scrap? What percentage is rework? How can the process be
centered to eliminate all but 0.1% of the scrap? What is the
rework percentage now?

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Solution
-USL: Upper Specification Limit (305.70 mm)
-LSL: Lower Specification Limit (304.55 mm)
-σ: Standard Deviation of the process (0.25 mm)
-μ: Process Mean (305.20 mm)

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Solution
Cp = (USL - LSL) / (6 * σ)
Cp = (305.70 - 304.55) / (6 * 0.25) = 1.5

Cpk =
min{(305.70 - 305.20) , (305.20 - 304.55) } / (3 * 0.25) = 0.67

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Step 2: Determine Percentage of Scrap and
Rework
-Scrap Percentage: This is the percentage
of products less than the lower
specification limit (LSL).
-Rework Percentage: This is the
percentage of products greater than the
upper specification limit (USL).

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 -Scrap Percentage:
For LSL (304.55 mm):
Z= (304.55 - 305.20) / 0.25 = -2.60
At Z= -2.6 Area 0.0047 (for scrap)

-Rework Percentage:
Z = (305.70 - 305.20) / 0.25 = 2.00

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 -Rework Percentage:
Z = (305.70 - 305.20) / 0.25 = 2.00
Z= 0.9966
0.0228 = 2.28% (for rework)

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Reducing Scrap to 0.1% = 0.001
Z= -3 at (0.001)
-3 = (304.55-µ)/0.25
µ = 305.3

Rework
Z= (305.7-305.3)/0.25 = 1.6
From Table A : Area = 1 – 0.9545= 4.55%

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Different Control Charts
Charts for Better Operator Understanding:
1. Placing individual values on the chart: This
technique plots both the individual values and
the subgroup average. Not recommended since
it does not provide much information.
2. Chart for subgroup sums: This technique plots
the subgroup sum, ΣX, rather than the group
average, Xbar. UCL = n(UCL )
X X

UCL X = n( LCLX )
107
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 FIGURE 6-27 Chart for FIGURE 6-28 Subgroup Sum
Individual Values & Subgroup Chart
Averages

108
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Different Control Charts
Charts for Variable Subgroup Size:
Used when the sample size is not the same
❑ Different control limits for each subgroup
❑ As n increases, limits become narrower
❑ As n decreases, limits become wider apart
❑ Difficult to interpret and explain
❑ To be avoided

109
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 FIGURE 6-29 Chart for Variable
Subgroup Size

110
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Different Control Charts
Chart for Trends:
Used when the plotted points have an upward
or downward trend that can be attributed to
an unnatural pattern of variation or a natural
pattern such as tool wear.
The central line is on a slope, therefore its
equation must be determined.

111
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Trends
Use Least Square Calculations
X = a + bG
(  X )(  G 2 ) − (  G )(  G X )
a=
g  G 2 − ( G )2
g  G X − (  G )(  X )
b=
g  G 2 − ( G )2
where
X = subgroup average and represents the vertical axis
a= point of intercept on the vertical axis
b=slope of the line
G=subgroup number and represents the horizontal axis
g=number of subgroups

112
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Least Squares
is a mathematical and statistical method used
for finding the best-fitting curve or line to a
set of data points.
The goal of the Least Squares method is to
minimize the sum of the squared differences
between the observed data points and the
values predicted by a mathematical model.
It's widely used in various fields, including
statistics, mathematics, engineering, and data
analysis

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Calculate the Trend Line
Determining the Trend Line: To establish the
trend line, you can employ various
methodologies based on the suspected trend:

1. Linear Regression: When a linear trend is

anticipated (either upward or downward), linear
regression is a valuable tool. It derives the
equation of the best-fitting line to the data. The
slope of this line reveals the trend's direction. A
positive slope implies an upward trend, whereas
a negative slope indicates a downward trend.
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Calculate the Trend Line
2. Moving Averages: In cases where the data
exhibits fluctuations alongside an underlying
trend, calculating moving averages (e.g., simple
moving averages or exponential moving
averages) proves effective. These averages help
smoothen noise within the data, making the
trend more discernible.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Calculate the Trend Line
3. Alternative Regression Models: Nonlinear
trends may necessitate the use of more
intricate regression models, such as polynomial
regression or exponential regression. These
models are better suited to accurately capture
complex trends that don't follow a linear
pattern.

Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc

Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Trends

Figure 6-32 Chart for Trend

117
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Moving Average and
Moving Range
Value Xbar R
44
Used when we cannot have 46

multiple observations per 54 48.00 10

38 46.00 16
time period 49 47.00 16
46 44.33 11

X=
X 45
31
46.67
40.67
4
15

n 55 43.67 24

R
37 41.00 24
42 44.67 18
R= 43 40.67 6
n 47 44.00 5
51 47.00 8
NOTE: n here is equal to 12, NOT 14
An example 118
Chart for Moving Average and
Moving Range
Extreme readings have a greater effect than in
conventional charts. An extreme value is used
several times in the calculations, the number of
times depends on the averaging period.

119
 Chart for Median and Range
This is a simplified variable control chart.
❑ Minimizes calculations
❑ Easier to understand
❑ Can be easily maintained by operators
❑ Recommended to use a subgroup of 3, then
all data is used.

120
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Median and Range

UCLMD = Md Md + A5 RMd
LCLMD = Md Md − A5 RMd
UCLR = D6 RMd
LCLR = D5 RMd
For Table for A5, D5 and D6 see page 230

121
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Median and Range

FIGURE 6-31 Control

Charts for Median and
Range

FIGURE 6-31 Control Charts for Median and Range

122
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Individual values
Used when only one measurement is taken on
quality characteristic
❑ Too expensive
❑ Time consuming
❑ Destructive
❑ Very few items

123
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Individual Values

X =
 X
R=
 R
g g
UCLx = X + 2.660 R
LCLx = X − 2.660 R
UCLR = 3.267 R
LCLR = (0) R
To use those equations, you have to use a moving range with n=2

124
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Individual Values
Revised Limits:

X 0 = X new R0 = R new
UCLx = X 0 + 3 0
LCLx = X 0 − 3 0
UCLR = 3.686 R0
LCLR = (0) 0
125
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Chart for Individual Values

FIGURE 6-32 Control Charts for Individual Values and

Moving Range

126
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Charts with Non-Acceptance
Limits
Non-Acceptance limits have the same
Relationship to averages as specifications
have to individual values. Control Limits tell
what the process is capable of doing, and reject
limits tell when the product is conforming to
specifications.

127
Charts with Non-Acceptance
Limits

Figure 6-35 Relationship of non-acceptance limits, control limits

and specifications. 128
 Exponential Weighted Average
 Gives greatest weight to most recent values
 The EWMA is defined by the euqation
 Vt = lXt + 11 - l2Vt-1
 where V t = the EWMA of the most recent
plotted point
 V t− 1 = the EWMA of the previous plotted
point
 l = the weight given to the subgroup average
or
 individual value
 Xt = the subgroup average or individual value
129
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Exponential Weighted Average
 Gives greatest weight to most recent values
 The EWMA is defined by the euqation
Vt = λ Xbart + (1 – λ) Vt-1

 where Vt = most recent plotted point

 Vt−1 = previous plotted point
 λ= weight given to subgroup average
or individual value
 Xbar = the subgroup average or individual
value
130
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Exponential Weighted Average

 UCL = Xdbar + A2Rbar(((SqRt(λ/(2 – λ)))

 LCL = Xdbar - A2Rbar(((SqRt(λ/(2 – λ)))

131
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Exponential Weighted Average

132
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved
 Computer Program
Computer Program file names are:
Xbar and R
Md and R
X and MR
EWMA
Process Capability

133
Quality Improvement, 9e © 2013, 2008 by Pearson Higher Education, Inc
Dale H. Besterfield Upper Saddle River, New Jersey 07458 • All Rights Reserved