Sample Standard Deviation Control Chart

SAMPLE STANDARD DEVIATION CONTROL CHART…
The reader should note that the standard deviation, σ0, is the same as the value obtained
from the range in the preceding section.
Using the standard values of X0 bar = 6.40 and σ0 = 0.038, the revised control limits are
computed.
Chapter 2: STATISTICAL PROCESS CONTROL 1

STATE OF CONTROL
(1) PROCESS IN CONTROL

When the assignable 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. No higher degree of uniformity can be attained with the existing process. Greater
uniformity can, however, be attained through a change in the basic process through quality
improvement ideas.
When a process is in control, there occurs a natural pattern of variation, which is illustrated
by the control chart in Figure 6-9.

FIGURE 6-9 Natural Pattern of Variation of a Control Chart

STATE OF CONTROL…
This natural pattern of variation has

(1) about 34% of the plotted points in an imaginary band between 1 standard deviation on
both sides of the central line,
(2) about 13.5% of the plotted points in an imaginary band between 1 and 2 standard
deviations on both sides of the central line, and
(3) about 2.5% of the plotted points in an imaginary band between 2 and 3 standard
deviations on both sides of the central line.
The points are located back and forth across the central line in a random manner, with no
points beyond the control limits. The natural pattern of the points or subgroup average values
forms its own frequency distribution. If all of the points were stacked up at one end, they
would form a normal curve (Figure 6-11 ).

STATE OF CONTROL…
 Control limits are usually established at 3 standard deviations from the central line. They
are used as a basis to judge whether there is evidence of lack of control.
 The choice of 3σ limits is an economic one with respect to two types of errors that can
occur.
 One error, called Type I by statisticians, occurs when looking for an assignable cause of
variation, when in reality a chance cause is present.
 When the limits are set at 3 standard deviations, a Type I error will occur 0.27% (3 out of
1000) of the time.
 In other words, when a point is outside the control limits, it is assumed to be due to an
assignable cause, even though it would be due to a chance cause 0.27% of the time. We
might think of this situation as “guilty until proven innocent.”

STATE OF CONTROL…
 The other type error, called Type II, occurs when assuming that a chance cause of
variation is present, when in reality there is an assignable cause.
 In other words, when a point is inside the control limits, it is assumed to be due to a chance
cause, even though it might be due to an assignable cause.
 We might think of this situation as “innocent until proven guilty.” Table 6-4 illustrates the
difference between the Type I and Type II errors.

STATE OF CONTROL…
 If control limits are established at, say, ±2.5 standard deviations, Type I errors would
increase and Type II decrease.
 Abundant experience since 1930 in all types of industry indicates that 3σ limits provide an
economic balance between the costs resulting from the two types of errors.
 Unless there are strong practical reasons for doing otherwise, the ±3 standard deviation
limits should be used.
 When a process is in control, only chance causes of variation are present. Small variations
in machine performance, operator performance, and material characteristics are expected
and are considered to be part of a stable process.

STATE OF CONTROL…
When a process is in control, certain practical advantages accrue to the producer and
consumer.
(1) Individual units of the product or service will be more uniform - or, stated another way,
there will be less variation and fewer rejections.
(2) Because the product or service is more uniform, fewer samples are needed to judge the
quality. Therefore, the cost of inspection can be reduced to a minimum. This advantage is
extremely important when 100% conformance to specifications is not essential.

STATE OF CONTROL…
(3) The process capability or spread of the process is easily attained from 6σ. With a knowledge
of process capability, a number of reliable decisions relative to specifications can be made, such
as (a) to decide the product or service specifications or requirements, (b) to decide the amount
of rework or scrap when there is insufficient tolerance, and (c) to decide whether to produce the
product to tight specifications and permit interchangeability of components or to produce the
product to loose specifications and use selective matching of components.
(4) Trouble can be anticipated before it occurs, thereby speeding up production by avoiding
rejections and interruptions.
(5) The percentage of product that falls within any pair of values may be predicted with the
highest degree of assurance. For example, this advantage can be very important when adjusting
filling machines to obtain different percentages of items below, between, or above particular
values.

STATE OF CONTROL…
(6) It permits the consumer to use the producer’s data and, therefore, to test only a few
subgroups as a check on the producer’s records. The X bar and R charts are used as
statistical evidence of process control.
(7) The operator is performing satisfactorily from a quality viewpoint. Further improvement in
the process can be achieved only by changing the input factors: materials, equipment,
environment, and operators. These changes require action by management.
When only chance causes of variation are present, the process is stable and predictable over
time, as shown in Figure 6-10 (a). We know that future variation as shown by the dotted
curve will be the same, unless there has been a change in the process due to an assignable
cause.

FIGURE 6-10
Stable and Unstable Variation

STATE OF CONTROL…
(2) PROCESS OUT OF CONTROL

The term out of control is usually thought of as being undesirable; however, there are
situations where this condition is desirable. It is best to think of the term out of control as a
change in the process due to an assignable cause.
When a point (subgroup value) falls outside its control limits, the process is out of control.
This means that an assignable cause of variation is present.
Another way of viewing the out-of-control point is to think of the subgroup value as coming
from a different population than the one from which the control limits were obtained.
Figure 6-10 (b) illustrates the effect of assignable causes of variation over time. The
unnatural, unstable nature of the variation makes it impossible to predict future variation. The
assignable causes must be found and corrected before a natural, stable process can
continue.
STATE OF CONTROL…
 A process can also be considered out of control even when the points fall inside the 3s
limits.
 This situation occurs when unnatural runs of variation are present in the process.
 First, let’s divide the control chart into 6 equal standard deviation bands in the same
manner as Figure 6-9.
 For identification purposes the bands are labeled A, B, and C zones, as shown in Figure 6-
12 .

FIGURE 6-12 Some Unnatural Runs - Process Out of Control

STATE OF CONTROL…
 It is not natural for seven or more consecutive points to be above or below the central line
as shown in Figure 6-12 (a).
 Also, when 10 out of 11 points or 12 out of 14 points, etc., are located on one side of the
central line, it is unnatural. Another unnatural run occurs at (b), where six points in a row
are steadily increasing or decreasing.
 In Figure 6-12 (c) we have two out of three points in a row in zone A and at (d) four out of
five points in a row in zone B and beyond. There are many statistical possibilities, with the
four common ones being shown in the figure.
 Actually, any significant divergence from the natural pattern as shown in Figure 6-9 would
be unnatural and would be classified as an out-of-control condition.

STATE OF CONTROL…
 Rather than divide the space into three equal zones of 1 standard deviation, a simplified
technique would divide the space into two equal zones of 1.5 standard deviations.
 The process is out of control when there are two successive points at 1.5 standard
deviations or beyond.
 The simplified rule makes for greater ease of implementation by operators without
drastically sacrificing power.
 10 It is shown in Figure 6-13 and replaces the information of Figures 6-12 (c) and (d).

FIGURE 6-13 Simplified Rule for Out-of-Control Process

STATE OF CONTROL…
ANALYSIS OF OUT-OF-CONTROL CONDITION

When a process is out of control, the assignable cause responsible for the condition must be
found. The detective work necessary to locate the cause of the out-of-control condition can
be minimized by knowledge of the types of out-of-control patterns and their assignable
causes.
Types of out-of-control X bar and R patterns are
(1) change or jump in level,
(2) trend or steady change in level,
(3) recurring cycles,
(4) two populations, and
(5) mistakes.

STATE OF CONTROL…
(1) CHANGE OR JUMP IN LEVEL: This type is concerned with a sudden change in level to
the X bar chart, to the R chart, or to both charts. Figure 6-14 illustrates the change in level.
For an X bar chart, the change in the process average can be due to
a. An intentional or unintentional change in the process setting
b. A new or inexperienced operator
c. A different material
d. A minor failure of a machine part
Some causes for a sudden change in the process spread or variability as shown on the R
chart are
a. Inexperienced operator
b. Sudden increase in gear play c. Greater variation in incoming material

FIGURE 6-14 Out-of-Control Pattern: Change or Jump in Level
Sudden changes in level can occur on both the X and the R charts.
This situation is common during the beginning of control chart activity, prior to the attainment
of a state of control.
There may be more than one assignable cause, or it may be a cause that could affect both
charts, such as an inexperienced operator.
STATE OF CONTROL…
(2) TREND OR STEADY CHANGE IN LEVEL:

Steady changes in control chart level are a very common industrial phenomena. Figure 6-15
illustrates a trend or steady change that is occurring in the upward direction; the trend could
have been illustrated in the downward direction.
Some causes of steady progressive changes on an X chart are
(a) Tool or die wear
(b) Gradual deterioration of equipment
(c) Gradual change in temperature or humidity
(d) Viscosity breakdown in a chemical process
(e) Buildup of chips in a work holding device

FIGURE 6-15 Out-of-Control Pattern: Trend or Steady Change in Level
A steady change in level or trend on the R chart is not as common as on the X chart.
It does, however, occur, and some possible causes are
(a) An improvement in operator skill (downward trend)
(b) A decrease in operator skill due to fatigue, boredom, inattention, etc. (upward trend)
(c) A gradual improvement in the homogeneity of incoming material

STATE OF CONTROL…
(3) RECURRING CYCLES: When the plotted points on an X or R chart show a wave or
periodic high and low points, it is called a cycle. A typical recurring out-of-control pattern is
shown in Figure 6-16.
For an X bar chart, some of the causes of recurring cycles are
(a) The seasonal effects of incoming material
(b) The recurring effects of temperature and humidity (cold morning start-up)
(c) Any daily or weekly chemical, mechanical, or psychological event
(d) The periodic rotation of operators

FIGURE 6-16 Out-of-Control Pattern: Recurring Cycles
Periodic cycles on an R chart are not as common as for an X chart.

Some affecting the R chart are due to
(a) Operator fatigue and rejuvenation resulting from morning, noon, and afternoon breaks
(b) Lubrication cycles

STATE OF CONTROL…
 The out-of-control pattern of a recurring cycle sometimes goes unreported because of the
inspection cycle.
 Thus, a cyclic pattern of a variation that occurs approximately every 2 h could coincide with
the inspection frequency.
 Therefore, only the low points on the cycle are reported, and there is no evidence that a
cyclic event is present.

STATE OF CONTROL…
(4)TWO POPULATIONS (ALSO CALLED MIXTURE): When there are a large number of
points near or outside the control limits, a two-population situation may be present. This type
of out-of-control pattern is illustrated in Figure 6-17.
For an X chart the out-of-control pattern can be due to
(a) Large differences in material quality
(b) Two or more machines on the same chart
(c) Large differences in test method or equipment
Some causes for an out-of-control pattern on an R chart are due to
(a) Different operators using the same chart
(b) Materials from different suppliers

FIGURE 6-17 Out-of-Control Pattern: Two Populations

STATE OF CONTROL…
(5) MISTAKES: Mistakes can be very embarrassing to quality assurance.

Some causes of out-of-control patterns resulting from mistakes are
(a) Measuring equipment out of calibration
(b) Errors in calculations
(c) Errors in using test equipment
(d) Taking samples from different populations
Many of the out-of-control patterns that have been described can also be attributed to
inspection error or mistakes.

STATE OF CONTROL…
 The causes given for the different types of out-of-control patterns are suggested
possibilities and are not meant to be all-inclusive.
 These causes will give production and quality personnel ideas for the solution of problems.
 They can be a start toward the development of an assignable cause checklist, which is
applicable to their particular organization.
 When out-of-control patterns occur in relation to the lower control limit of the R chart, it is
the result of outstanding performance.
 The cause should be determined so that the outstanding performance can continue.

SPECIFICATIONS
INDIVIDUAL VALUES COMPARED TO AVERAGES

Before discussing specifications and their relationship with control charts, it appears
desirable, at this time, to obtain a better understanding of individual values and average
values.
Figure 18 shows a tally of individual values (X’s) and a tally of the subgroup averages (X bar)
for the data on keyway depths given in Table 2. The four out-of-control subgroups were not
used in the two tallys; therefore, there are 84 individual values and 21 averages. It is
observed that the averages are grouped much closer to the center than the individual values.
When four values are averaged, the effect of an extreme value is minimized because the
chance of four extremely high or four extremely low values in one subgroup is slight.

Figure 18 Comparison of Individual Values and Averages Using the Same Data

SPECIFICATIONS…
Calculations of the average for both the individual values and for the subgroup averages are
the same, X bar = 38.9.
However, the sample standard deviation of the individual values (s) is 4.16, whereas the
sample standard deviation of the subgroup average (sXbar) is 2.77.
If there are a large number of individual values and subgroup averages, the smooth polygons
of Figure 18 would represent their frequency distributions if the distribution is normal.
The curve for the frequency distribution of the averages is a dashed line, whereas the curve
for the frequency distribution of individual values is a solid line.
In comparing the two distributions, it is observed that both distributions are normal in shape;
in fact, even if the curve for individual values were not quite normal, the curve for averages
would be close to a normal shape.

SPECIFICATIONS…
The base of the curve for individual values is about twice as large as

SPECIFICATIONS…

SPECIFICATIONS…

SPECIFICATIONS…
 Because the height of the curve is a function of the frequency, the

curve for individual values is higher.
 This is easily verified by comparing the tally sheet in Figure 18.
 However, if the curves represent relative or percentage frequency
distributions, then the area under the curve must be equal to 100%.
 Therefore, the percentage frequency distribution curve for averages,
with its smaller base, would need to be much higher to enclose the
same area as the percentage frequency distribution curve for
individual values.

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.
 Furthermore, the standardized normal can be used for the distribution of averages with the
modification,

 This theorem was illustrated by Shewhart for a uniform population distribution

and a triangular population distribution of individual values as shown in Figure
19.
 Obviously, the distribution of X’s is considerably different than a normal
distribution; however, the distribution of X bar is approximately normal.

Figure 19 Illustration of Central Limit Theorem

 The central limit theorem is one of the reasons the X bar chart works,
in that we do not need to be concerned if the distribution of X’s is not
normal, provided that the sample size is 4 or more.
 Figure 20 shows the results of a dice experiment.
 First is a distribution of individual rolls of a six-sided die; second is a
distribution of the average of rolls of two dice.
 The distribution of the averages (X bar) is unimodal, symmetrical,
and tapers off at the tails.
 This experiment provides practical evidence of the validity of the
central limit theorem.

Figure 20 Dice illustration of Central Limit Theorem

Control Limits and Specifications
 Control limits are established as a function of the averages; in other words,

control limits are for averages.
 Specifications, on the other hand, are the permissible variation in the size of
the part and are, therefore, for individual values.
 The specification or tolerance limits are established by design engineers to
meet a particular function.
 Figure 21 shows that the location of the specifications is optional and is not
related to any of the other features in the figure.
 The control limits, process spread, distribution of averages, and distribution of
individual values are interdependent.
 They are determined by the process, whereas the specifications have an
optional location.
 Control charts cannot determine whether the process is meeting specifications.

Figure 21 Relationship of Limits, Specifications, and
Distributions

Process Capability and Tolerance
 Hereafter the process spread will be referred to as the process

capability and is equal to 6σ. Also, 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.
Three situations are possible:
(1)when the process capability is less than the tolerance,
(2)when the process capability is equal to the tolerance, and
(3)when the process capability is greater than the tolerance.

Process Capability and Tolerance…
Case I: 6σ ˂ USL - LSL: This situation, where the process capability

(6s) is less than the tolerance (USL – LSL), is the most desirable case.
Figure 22 illustrates this ideal relationship by showing the distribution
of individual values (X’s), the X bar control chart limits, and
distribution of averages (X bar).

Figure 22 Case I: 6σ ˂ USL - LSL

 The process is in control in (a).

 Because the tolerance is appreciably greater than the process
capability, no difficulty is encountered even when there is a
substantial shift in the process average, as shown in (b).
 This shift has resulted in an out-of-control condition as shown by the
plotted points.
 However, no waste is produced, because the distribution of
individual values (X’s) has not exceeded the upper specification.
 Corrective action is required to bring the process into control.

 Case II: 6σ = USL – LSL

Figure 23 illustrates this case, where the process capability is equal to
the tolerance. The frequency distribution of X’s in (a) represents a
natural pattern of variation.
However, when there is a shift in the process average, as indicated in
(b), the individual values (X’s) exceed the specifications.
As long as the process remains in control, no nonconforming product
is produced; however, when the process is out of control as
indicated in (b), nonconforming product is produced.
Therefore, assignable causes of variation must be corrected as soon
as they occur.

Figure 23 Case II: 6σ = USL - LSL

 When the process capability is greater than the tolerance, an

undesirable situation exists.
 Figure 24 illustrates this case. Even though a natural pattern of
variation is occurring, as shown by the frequency distribution of X’s
in (a), some of the individual values are greater than the upper
specification and are less than the lower specification.
 This case presents the unique situation where the process is in
control as shown by the control limits and frequency distribution of X
bar, but nonconforming product is produced.
 In other words, the process is not capable of manufacturing a
product that will meet the specifications. When the process changes
as shown in (b), the problem is much worse.
Figure 24 Case III: 6σ ˃ USL - LSL

 When this situation occurs, 100% inspection is necessary to eliminate

the nonconforming product.
 One solution is to discuss with the design engineer the possibility of
increasing the tolerance.
 This solution may require reliability studies with mating parts to
determine if the product can function with an increased tolerance.
Selective assembly might also be considered by the engineer.
 A second possibility is to change the process dispersion so that a
more peaked distribution occurs.
 To obtain a substantial reduction in the standard deviation might
require new material, a more experienced operator, retraining, a new or
overhauled machine, or possibly automatic in-process control.
 Another solution is to shift the process average so that all of the

nonconforming product occurs at one tail of the frequency distribution, as
indicated in Figure 24(b).
 To illustrate this solution, assume that a shaft is being ground to tight
specifications. If too much metal is removed, the part is scrapped; if too
little is removed, the part must be reworked.
 By shifting the process average, the amount of scrap is eliminated and the
amount of rework is increased. A similar situation exists for an internal
member such as a hole or keyway, except that scrap occurs above the
upper specification and rework occurs below the lower specification.
 This type of solution is feasible when the cost of the part is sufficient to
economically justify the reworking operation.

Example Problem
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 (µ) and the dispersion is 0.02
mm (µ), 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?

PROCESS CAPABILITY
 The true process capability cannot be determined until the X bar and
R charts have achieved the optimal quality improvement without a
substantial investment for new equipment or equipment modification.
 Process capability is equal to 6σ0 when the process is in statistical
control.
 In Example Problem 1 for the X bar and R charts, the quality
improvement process began in January with σ0 = 0.038.
 The process capability is 6σ = (6)(0.038) = 0.228 mm.
 By July, σ0 = 0.030, which gives a process capability of 0.180 mm.
This is a 20% improvement in the process capability, which in most
situations will be sufficient to solve a quality problem.
PROCESS CAPABILITY…


 Either the range or the standard deviation method can be used,

although, as previously stated, the standard deviation method is
more accurate.
 A histogram should be constructed to graphically present the
process capability.
 Actually, a minimum of 50 measurements is required for a histogram.
Therefore, the histograms made from the same data that were used
to calculate the process capability should adequately represent the
process at that time.

 Process capability and the tolerance are combined to form a

capability index, defined as
 If the capability index is 1.00, we have the Case II situation discussed

in the preceding section; if the ratio is greater than 1.00, we have the
Case I situation, which is desirable; and if the ratio is less than 1.00,
we have the Case III situation, which is undesirable.
 Figure 25 shows these three cases.
Figure 25 Capability Index and Three Cases



Figure 26 illustrates Cp and Cpk
values for a process that is
centered and one that is off-
center by 1σ for the three cases.
Comments concerning Cp and
Cpk are as follows.
Figure 26 Cp and
Cpk Values for the
Three Cases

OTHER CONTROL CHARTS (SPECIAL CONTROL PROCEDURES)
 The basic control charts for variables were discussed in previous

sections.
 Although most quality control activity for variables is concerned with
the X bar and R chart or the X bar and s chart, there are other charts
which find application in some situations.
 These charts are discussed briefly in this section.

Charts for Better Operator Understanding
 Because production personnel have difficulty understanding the

relationships between averages, individual values, control limits,
and specifications, various charts have been developed to overcome
this difficulty.

1. PLACING INDIVIDUAL VALUES ON THE CHART:

This technique plots both the individual values and the subgroup
average and is illustrated in Figure 27.
A small dot represents an individual value and a larger circle
represents the subgroup average. In some cases, an individual value
and a subgroup average are identical, in which case the small dot is
located inside the circle. When two individual values are identical, the
two dots are placed side by side. A further refinement of the chart can
be made by the addition of upper and lower specification lines;
however, this practice is not recommended.
In fact, the plotting of individual values is an unnecessary activity that
can be overcome by proper operator training.
Figure 27 Chart Showing a Technique for Plotting Individual
Values and Subgroup Averages

2. CHART FOR SUBGROUP SUMS:

This technique plots the subgroup sum, ∑X, rather than the subgroup
average, X bar. Because the values on the chart are of a different
magnitude than the specifications, there is no chance for confusion.
Figure 28 shows a subgroup sum chart, which is an X bar chart with the
scale magnified by the subgroup size, n. The central line is nX0 bar and
the control limits are obtained by the formulas
This chart is mathematically equal to the X bar chart and has the added
advantage of simpler calculations.
Only addition and subtraction are required.
Figure 28 Subgroup Sum Chart

Chart for Variable Subgroup Size
 Every effort should be made to keep the subgroup size constant.

Occasionally, however, because of lost material, laboratory tests,
production problems, or inspection mistakes, the subgroup size
varies.
 When this situation occurs, the control limits will vary with the
subgroup size.
 As the subgroup size, n, increases, the control limits become
narrower; as the subgroup size decreases, the control limits become
wider apart (Figure 29).
 This fact is confirmed by an analysis of the control limit factors A2,
D1, and D2, which are functions of the subgroup size and which are
part of the control limit formulas.

Chart for Variable Subgroup Size…
 Control limits will also vary for the R chart.

 One of the difficulties associated with a chart for variable subgroup
size is the need to make a number of control limit calculations.
 A more serious difficulty involves the task of explaining to
production people the reason for the different control limits.
Therefore, this type of chart should be avoided.

Figure 29 Chart for Variable Subgroup Size

Chart for Trends
 When the plotted points of a chart have an upward or downward

trend, it can be attributed to an unnatural pattern of variation or to a
natural pattern of variation such as tool wear.
 In other words, as the tool wears, a gradual change in the average is
expected and considered to be normal.
 Figure 30 illustrates a chart for a trend that reflects die wear.
 As the die wears, the measurement gradually increases until it
reaches the upper reject limit. The die is then replaced or reworked.

FIGURE 30 Chart for Trend

Chart for Trends…

TABLE 5 Least-Squares Calculations for Trend Line

Chart for Trends…
 Once the trend-line equation is known, it can be plotted on the chart

by assuming values of G and calculating X.
 When two points are plotted, the trend line is drawn between them.
 The control limits are drawn on each side of the trend line, a distance
(in the perpendicular direction) equal to A2R bar or Aσ0.
 The R chart will generally have the typical appearance shown in
Figure 7. However, the dispersion may also be increasing.

Chart for Moving Average and Moving Range
In some situations a chart is used to combine a number of individual

values and plot them on a control chart.
This type is referred to as a moving-average and moving-range chart
and is quite common in the chemical industry, where only one reading
is possible at a time.
Table 6 illustrates the technique. In the development of Table 6, no
calculations are made until the third period, when the sum of the three
values is posted to the three-period moving-sum column (35 + 26 + 28
= 89).
The average and range are calculated (X bar = 89 ÷ 3 = 29.6)(R = 35 - 26
= 9) and posted to the X bar and R columns. Subsequent calculations
are accomplished
Chart for Moving Average and Moving Range…
Subsequent calculations are accomplished by adding a new value and

dropping the earliest one; therefore, 32 is added and 35 is dropped,
making the sum 26 + 28 + 32 = 86. The average and range calculations
are X bar = 86 ÷ 3 = 28.6 and R = 32 - 26 = 6.
Once the columns for X and R are completed, the charts are developed
and used in the same manner as regular X bar and R charts.
The preceding discussion used a time period of 3 h; the time period
could have been 2 h, 5 days, 3 shifts, and so on.

Chart for Moving Average and Moving Range…
 In comparing the moving-average and moving-range charts with

conventional charts, it is observed that an extreme reading has a
greater effect on the former charts.
 This is true because an extreme value is used a number of times in
the calculations, and therefore will detect small changes much faster.

Chart for Median and Range
A simplified variable control chart that minimizes calculations is the

median and range.
The data are collected in the conventional manner and the median, Md,
and range, R, of each subgroup are found.
When using manual methods, these values are arranged in ascending
or descending order.
The median of the subgroup medians or grand median, MdMd, and the
median of the subgroup range, RMd, are found by counting to the
midpoint value.

Chart for Median and Range
The median control limits are

determined from the formulas
where D5 and D6 are factors for determining the 3σ control limits based on
RMd and are found in Table 7.
An estimate of the population standard deviation can be obtained from σ =
RMd ÷ D3.
TABLE 7 Factors for Computing 3σ Control Limits for
Median and Range Charts from the Median Range

Chart for Median and Range…
The principal benefits of the median chart are as follows: (1) there is
less arithmetic, (2) it is easier to understand, and (3) it can be easily
maintained by operators.
However, the median chart fails to grant any weight to extreme values
in a subgroup.
When these charts are maintained by operating personnel, a subgroup
size of 3 is recommended. For example, consider the three values 36,
39, and 35.
The Md is 36 and R is 4 - all three values are used. Figure 31 is an
example of a median chart.

FIGURE 31 Control Charts for Median and Range

Chart for Median and Range….
 Subgroup sizes of 5 give a better chart; however, using the manual

method, the operator will have to order the data before determining
the median.
 Although these charts are not as sensitive to variation as the X bar
and R charts, they can be quite effective, especially after quality
improvement has been obtained and the process is in a monitoring
phase.
 An unpublished master’s thesis study showed little difference in the
effectiveness of the Md and R charts when compared to the X bar and
R charts.

Chart for Individual Values
 In many situations, only one measurement is taken on a quality

characteristic.
 This may be due to the fact that it is too expensive or too time-
consuming, there are too few items to inspect, or it may not be
possible.
 In such cases an X chart will provide some information from limited
data, whereas an X chart would provide no information or information
only after considerable delay to obtain sufficient data.
 Figure 32 illustrates an X chart.

FIGURE 32 Control Charts for Individual Values and Moving
Range

Chart for Individual Values…
These formulas require the moving-range technique with a subgroup

size of 2.
To obtain the first range point, the value of X1 is subtracted from X2 ; to
obtain the second point, X2 is subtracted from X3; and so forth.
Each individual value is used for two different points except for the
first and last - therefore, the name “moving” range.
The range points should be placed between the subgroup number on
the R chart because they are obtained from both values, or they can be
placed at the second point.


The X chart has the advantage of being easier for production

personnel to understand and of providing a direct comparison with
specifications.
It does have the disadvantages of (1) requiring too many subgroups to
indicate an out-of-control condition, (2) not summarizing the data as
well as X bar, and (3) distorting the control limits when the distribution
is not normal.
To correct for the last disadvantage, tests for normality should be
used.
Unless there is an insufficient amount of data, the X bar chart is
recommended.

Chart for Exponential Weighted Moving-Average
The exponential weighted moving-average (EWMA) chart, also called

the geometric moving-average chart, gives the greatest weight to the
most recent data and less weight to all previous data.
Its primary advantage is the ability to detect small shifts in the process
average; however, it does not react as quickly to large shifts as the X
chart.
The advantages of both techniques can be achieved by plotting them
on the same chart using different colors.

Chart for Exponential Weighted Moving-Average...
Another chart that has the ability to detect small shifts in the process
average is called the cusum chart.
It is more difficult to understand and calculate, and it does not react as
well as the EWMA chart to large shifts. Details are given in Juran’s
Quality Control Handbook.

Chart for Exponential Weighted Moving-Average…

Chart for Exponential Weighted Moving-Average…
Actually, the control limits for the first few samples use different equations; however,
the control limit values increase rapidly to their limiting values as determined by the
equations given above.

Example Problem 10
Using the information in Table 2 with the three out-of-control

subgroups removed, determine the control limits and plot the points
for an EWMA control chart using a λ value of 0.15.
Table 8 gives an EXCEL spreadsheet for the calculations, and Figure
34 shows the actual control chart.

FIGURE 34 EWMA Control Chart for Data in Table 8

Sample Standard Deviation Control Chart

Uploaded by

Copyright:

Available Formats

Sample Standard Deviation Control Chart

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

Sample Standard Deviation Control Chart

Uploaded by

Copyright:

Available Formats

SAMPLE STANDARD DEVIATION CONTROL CHART…

Chapter 2: STATISTICAL PROCESS CONTROL 1

(1) PROCESS IN CONTROL

Chapter 2: STATISTICAL PROCESS CONTROL 2

Chapter 2: STATISTICAL PROCESS CONTROL 3

This natural pattern of variation has

Chapter 2: STATISTICAL PROCESS CONTROL 4

Chapter 2: STATISTICAL PROCESS CONTROL 5

Chapter 2: STATISTICAL PROCESS CONTROL 6

Chapter 2: STATISTICAL PROCESS CONTROL 7

Chapter 2: STATISTICAL PROCESS CONTROL 8

Chapter 2: STATISTICAL PROCESS CONTROL 9

Chapter 2: STATISTICAL PROCESS CONTROL 10

Chapter 2: STATISTICAL PROCESS CONTROL 11

(2) PROCESS OUT OF CONTROL

Chapter 2: STATISTICAL PROCESS CONTROL 13

Chapter 2: STATISTICAL PROCESS CONTROL 14

Chapter 2: STATISTICAL PROCESS CONTROL 15

Chapter 2: STATISTICAL PROCESS CONTROL 16

Chapter 2: STATISTICAL PROCESS CONTROL 17

ANALYSIS OF OUT-OF-CONTROL CONDITION

Chapter 2: STATISTICAL PROCESS CONTROL 18

Chapter 2: STATISTICAL PROCESS CONTROL 19

(2) TREND OR STEADY CHANGE IN LEVEL:

Chapter 2: STATISTICAL PROCESS CONTROL 21

Chapter 2: STATISTICAL PROCESS CONTROL 22

Chapter 2: STATISTICAL PROCESS CONTROL 23

Periodic cycles on an R chart are not as common as for an X chart.

Chapter 2: STATISTICAL PROCESS CONTROL 24

Chapter 2: STATISTICAL PROCESS CONTROL 25

Chapter 2: STATISTICAL PROCESS CONTROL 26

Chapter 2: STATISTICAL PROCESS CONTROL 27

(5) MISTAKES: Mistakes can be very embarrassing to quality assurance.

Chapter 2: STATISTICAL PROCESS CONTROL 28

Chapter 2: STATISTICAL PROCESS CONTROL 29

INDIVIDUAL VALUES COMPARED TO AVERAGES

Chapter 2: STATISTICAL PROCESS CONTROL 30

Chapter 2: STATISTICAL PROCESS CONTROL 31

Chapter 2: STATISTICAL PROCESS CONTROL 32

Chapter 2: STATISTICAL PROCESS CONTROL 33

Chapter 2: STATISTICAL PROCESS CONTROL 34

Chapter 2: STATISTICAL PROCESS CONTROL 35

 Because the height of the curve is a function of the frequency, the

Chapter 2: STATISTICAL PROCESS CONTROL 36

Chapter 2: STATISTICAL PROCESS CONTROL 37

 This theorem was illustrated by Shewhart for a uniform population distribution

Chapter 2: STATISTICAL PROCESS CONTROL 38

Chapter 2: STATISTICAL PROCESS CONTROL 39

Chapter 2: STATISTICAL PROCESS CONTROL 40

Chapter 2: STATISTICAL PROCESS CONTROL 41

 Control limits are established as a function of the averages; in other words,

Chapter 2: STATISTICAL PROCESS CONTROL 42

Chapter 2: STATISTICAL PROCESS CONTROL 43

 Hereafter the process spread will be referred to as the process

Chapter 2: STATISTICAL PROCESS CONTROL 44

Case I: 6σ ˂ USL - LSL: This situation, where the process capability

Chapter 2: STATISTICAL PROCESS CONTROL 45

Chapter 2: STATISTICAL PROCESS CONTROL 46

 The process is in control in (a).