Presentation Based On Chapter 6 of The Text Book

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery.
1
Copyright (c) 2012 John Wiley & Sons, Inc.
Learning Objectives
Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 2

Subgroup Data with Unknown  and 

The development of the equations for computing the control limits on the x-bar
and R control charts is relatively easy. In Chapter 4 (Section 4.2) we observed
that there is a well-known relationship between the range of a sample from a
normal distribution and the standard deviation of that distribution.
The random variable W = R/σ is called the relative range. The parameters of
the distribution of W are a function of the sample size n. The mean of W is d2.
Consequently, an estimator of σ is σˆ R/d2. Values of d2 for various sample
sizes are given in Appendix Table VI. Therefore, if is the average range of the m
preliminary samples, we may use:

Phase I Application of x and R Charts
• Eqns 6.4 and 6.5 are trial control limits
– Determined from m initial samples
• Typically 20-25 subgroups of size n between 3 and 5
– Any out-of-control points should be examined for assignable causes
• If assignable causes are found, discard points from calculations and
revise the trial control limits
• Continue examination until all points plot in control
• Adopt resulting trial control limits for use
• If no assignable cause is found, there are two options
1. Eliminate point as if an assignable cause were found and revise limits
2. Retain point and consider limits appropriate for control
– If there are many out-of-control points they should be examined for
patterns that may identify underlying process problems

Example 6.1 The Hard Bake Process

Revision of Control Limits
and Center Lines
• Effective use of control charts requires periodic
review and revision of control limits and center lines
• Sometimes users replace the center line on the x chart
with a target value
• When R chart is out of control, out-of-control points
are often eliminated to recompute a revised value of R
which is used to determine new limits and center line
on R chart and new limits on x chart

Phase II Operation of Charts
• Use of control chart for monitoring future
production, once a set of reliable limits are
established, is called phase II of control chart
usage (Figure 6.4)
• A run chart showing individuals observations
in each sample, called a tolerance chart or
tier diagram (Figure 6.5), may reveal patterns
or unusual observations in the data

• In examining control chart data, it is
sometimes helpful to construct a run chart of
the individual observations in each sample.
This chart is sometimes called a tier chart or
tolerance diagram. This may reveal some
pattern in the data, or it may show that a
particular value of x-bar or R was produced by
one or two unusual observations in the sample.
A box plot is usually a very simple way to
construct the tier diagram.

• A tier chart of the flow width data observations is shown in Fig. 6.5. This
chart does not indicate that the out-of-control signals were generated by
unusual individual observations, but instead, they probably resulted from a
shift in the mean around the time that sample 38 was taken. The average of
the averages of samples 38 through 45 is 1.6633 microns. The specification
limits of (1.50–0.50) and (1.50+0.50) microns are plotted in Fig. 6.5, along
with a sketch of the normal distribution that represents process output
when the process mean equals the in-control value 1.5056 microns. A
sketch of the normal distribution representing process output at the new
apparent mean diameter of 1.6633 microns is also shown in Fig. 6.5.
• It is obvious that a much higher percentage of nonconforming wafers will

be produced at this new mean flow rate. Since the process is out of control,
a search for the cause of this shift in the mean must be conducted.

Control vs. Specification Limits
• Control limits are derived
from natural process
variability, or the natural
tolerance limits of a process
• Specification limits are

determined externally, for
example by customers or
designers
• There is no mathematical or
statistical relationship
between the control limits
and the specification limits
Rational Subgroups
• x charts monitor between-sample variability
• R charts measure within-sample variability
• Standard deviation estimate of  used to construct
control limits is calculated from within-sample
variability
• It is not correct* to estimate  using (see the next
display)

*The estimate of the process standard deviation σ used in constructing
the control limits is calculated from the variability within each sample
(i.e., from the individual sample ranges). Consequently, the estimate of σ
reflects only within-sample variability. It is not correct to estimate σ
based on the usual quadratic estimator s (see the previous display), where
where xij is the jth observation in the ith sample, because if the
sample means differ, then this will cause s to be too large.
Consequently, σ will be overestimated.
Pooling all of the preliminary data in this manner to estimate σ is

not a good practice because it potentially combines both between-
sample and within-sample variability. The control limits must be
based on only within-sample variability.

Guidelines for Control Chart Design
• Control chart design requires specification of sample size,
control limit width, and sampling frequency
– Exact solution requires detailed information on statistical
characteristics as well as economic factors
– The problem of choosing sample size and sampling frequency is one of
allocating sampling effort
• For x chart, choose as small a sample size is consistent with
magnitude of process shift one is trying to detect. For
moderate to large shifts, relatively small samples are effective.
For small shifts, larger samples are needed.
• For small samples, R chart is relatively insensitive to changes
in process standard deviation. For larger samples (n > 10 or
12), s or s2 charts are better choices (see Sections 6.3.1 and 6.3.2).

When it is possible to specify standard values for the process mean and
standard deviation, we may use these standards to establish the control
charts for x-bar and R without analysis of past data.
Suppose that the standards given are μ and σ. Then the parameters of
the x-bar chart are in Equations 6.14 and 6.15 (A can be read from Appendix
Table VI).
To construct the R chart with a standard value of σ, recall that σ = R/d2, where
d2 is the mean of the distribution of the relative range. Furthermore, the
standard deviation of R is σR =d3σ, where d3 is the standard deviation of the
distribution of the relative range. Therefore, the parameters of the control chart
are in Equation 6.16.

6.2.4 Interpretation of Control Charts

Cyclic patterns occasionally appear on the control chart. A
typical example is shown in Fig. 6.8. Such a pattern on the x-
bar chart may result from systematic environmental changes
such as temperature, operator fatigue, regular rotation of
operators and/or machines, or fluctuation in voltage or
pressure or some other variable in the production equipment.
R charts will sometimes reveal cycles because of
maintenance schedules, operator fatigue, or tool wear
resulting in excessive variability.

A mixture is indicated when the plotted points tend to fall near or
slightly outside the control limits, with relatively few points near
the center line, as shown in Fig. 6.9. A mixture pattern is
generated by two (or more) overlapping distributions generating
the process output. The probability distributions that could be
associated with the mixture pattern in Fig. 6.9 are shown on the
right-hand side of that figure. The severity of the mixture pattern
depends on
the extent to which the distributions overlap. Sometimes mixtures
result from “overcontrol,” where the operators make process
adjustments too often, responding to random variation in the
output rather than systematic causes. A mixture pattern can also
occur when output product from several sources (such as parallel
machines) is fed into a common stream which is then sampled for
process monitoring purposes.

A shift in process level is illustrated in Fig. 6.10. These shifts may result
from the introduction of new workers; changes in methods, raw materials, or
machines; a change in the inspection method or standards; or a change in
either the skill, attentiveness, or motivation of the operators. Sometimes an
improvement in process performance is noted following introduction of a
control chart program, simply because of motivational factors influencing the
workers.

A trend, or continuous movement in one direction, is shown on the
control chart in Fig. 6.11. Trends are usually due to a gradual wearing
out or deterioration of a tool or some
other critical process component. In chemical processes they often
occur because of settling or separation of the components of a
mixture. They can also result from human causes, such
as operator fatigue or the presence of supervision. Finally,
trends can result from seasonal influences, such as
temperature. When trends are due to tool wear or other
systematic causes of deterioration, this may be directly
incorporated into the control chart model. (e.g. Regression Control
Chart, Modified Control Chart (see Chapter 9))

Stratification, or a tendency for the points to cluster artificially
around the center line, is illustrated in Fig. 6.12. We note that
there is a marked lack of natural variability in the observed
pattern. One potential cause of stratification is incorrect
calculation of control limits. This pattern may also result when
the sampling process collects one or more units from several
different underlying distributions within each subgroup.
•For example, suppose that a sample of size 5 is obtained by taking one

observation from each of five parallel processes. If the largest and smallest
units in each sample are relatively far apart because they come from two
different distributions, then R will be incorrectly inflated, causing the limits on
the x-bar chart to be too wide. In this case R incorrectly measures the
variability between the different underlying distributions, in addition to the to
the chance cause variation that it is intended to measure.

6.2.5 The Effect of Non-normality
• An assumption in performance properties is that the
underlying distribution of quality characteristic is normal
– If underlying distribution is not normal, sampling distributions can be
derived and exact probability limits obtained
• Burr (1967) notes the usual normal theory control limits are
very robust to normality assumption
• Schilling and Nelson (1976) indicate that in most cases,
samples of size 4 or 5 are sufficient to ensure reasonable
robustness to normality assumption for x-bar chart
• Sampling distribution of R is not symmetric, thus symmetric
3-sigma
x limits are an approximation and -risk is not 0.0027.
R chart is more sensitive to departures from normality than
chart.
• Assumptions of normality xand independence are not a primary
concern in phase I

6.2.6 The Operating Characteristic Function

If the shift is 1.0σ and the sample
size is n = 5, then β = 0.75.

Development of the control limits:
Thius produces the control limits in equation (6.27)

This produces the control limits in equation (6.28)

Average Run Lengths
• Crowder (1987b) showed that
ARL0 of combined individuals
and moving-range chart with
conventional 3-sigma limits is
generally much less than ARL0
(= 370) of standard Shewhart
control chart
• Ability of individuals chart to

detect small shifts is very poor
– Rather than narrowing the 3-
sigma limits, correct approach to
detecting small shifts is a
cumulative-sum or exponentially
weighted moving-average control
chart (Chapter 9)

Normality
• Borror, Montgomery, and Runger (1999) report that the in-control ARL is dramatically
affected by nonnormal data
• One approach for nonnormal data is to determine control limits for individuals control chart
based on percentiles of correct underlying distribution
– Requires at least 100 and preferably 200 observations
– Transformations can also be useful

Learning Objectives


Presentation Based On Chapter 6 of The Text Book

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

Presentation Based On Chapter 6 of The Text Book

Uploaded by

Copyright:

Available Formats

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery.

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 2

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 5

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 8

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 11

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 12

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 18

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 19

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 22

• It is obvious that a much higher percentage of nonconforming wafers will

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 24

• Specification limits are

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 26

Pooling all of the preliminary data in this manner to estimate σ is

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 27

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 28

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 30

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 32

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 33

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 34

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 35

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 36

•For example, suppose that a sample of size 5 is obtained by taking one

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 38

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 39

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 40

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 41

Thius produces the control limits in equation (6.27)

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 48

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 49

• Ability of individuals chart to

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 66

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 67

Chapter 6 Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. 75

You might also like