BITS Pilani

Pilani Campus

Course No: MATH F432

Applied Statistical Methods

1
Statistical Quality Control: Control Charts for
Process Variation and Attributes
(Lecture 31)
Control Charts for Process Variation
The control charts discussed in the previous section were
designed to control the location (equivalently, central
tendency; chart) of a process, with particular attention to
the mean as a measure of location.

It is equally important to ensure that a process is under

control with respect to variation.

In fact, most practitioners recommend that control be

established on variation prior to constructing an chart or
any other chart for controlling location.

3
Control Charts for Process Variation
In this section, we consider charts for variation based on
the sample standard deviation S (S-chart) and also charts
based on the sample range R (R-chart).

The former are generally preferred because the standard

deviation gives a more efficient assessment of variation
than does the range, but R charts were used first and easy
to understand / implement.

4
The S Chart

5
The S Chart
We again suppose that k independently selected samples
are available, each one consisting of n observations on a
normally distributed variable. .

Denote the sample standard deviations by s1, s2, … sk, with

The values s1, s2, s3 are plotted in sequence on
an S chart.

The center line of the chart will be at height , and the

3-sigma limits necessitate determining 3 S (just as 3-sigma
limits of an chart required with  then
estimated from the data).
6
Recall:

7
The S Chart
We know that for any rv Y, V(Y) = E(Y2) – [E(Y)]2, and that
a sample variance S2 is an unbiased estimator of  2, that
is, E(S2) =  2. Thus

V(S) = E(S2) – [E(S)]2 =  2 – (an)2 =  2(1 – )

where values of an for n = 3, . . . ,8 are tabulated in the

previous section. The standard deviation of S is then

8
The S Chart
It is natural to estimate  using s1,…, sk, as was done in the
previous section namely, . Substituting for s in the
expression for S gives the quantity used to calculate
3-sigma limits.

The 3-sigma control limits for an S control chart are

The expression for LCL will be negative if n  5, in which

case it is customary to use LCL = 0.
9
Example
Observations on stress resistance of plastic sheets (the
force, in psi, necessary to crack a sheet).

10
Example cont’d

There are k = 22 samples, obtained at equally spaced time

points, and n = 4 observations in each sample.

It is easily verified that si = 51.10 and s_bar = 2.32 so the

center of the S chart will be at 2.32 (though because n = 4,
LCL = 0 and the center line will not be midway between the
control limits).

From the previous section, a4 = .921 , from which the UCL is

11
Example cont’d

The resulting control chart is shown below. All plotted

points are well within the control limits, suggesting stable
process behavior with respect to variation.

S chart for stress-resistance data for Example 4

Figure 16.3
12
The R Chart

13
The R Chart
Let r1, r2, . . . , rk denote the k sample ranges and r = ri /k.
The center line of an R chart will be at height r.

Determination of the control limits requires R, where R

denotes the range (prior to making observations—as a
random variable) of a random sample of size n from a
normal distribution with mean value  and standard
deviation . Because

R = max(X1,. . . , Xn) – min(X1,. . . , Xn)

=  {max(Z1,. . . , Zn) – min(Z1, . . . , Zn)}

14
The R Chart
Where Zi = (Xi –  )/, and the Zi’s are standard normal rv’s,
it follows that

=   cn

The values of cn for n = 3, . . . , 8 appear in the

accompanying table.

15
Recall:

16
The R Chart
It is customary to estimate  by as discussed in the
previous section.

This gives as the estimated standard deviation of

R.

The 3-sigma limits for an R chart are

The expression for LCL will be negative if n  6, in which

case LCL = 0 should be used.
17
Example
In tissue engineering, cells are seeded onto a scaffold that
then guides the growth of new cells. The article “On the
Process Capability of the Solid Free-Form Fabrication:

A Case Study of Scaffold Moulds for Tissue Engineering”

(J. of Engr. in Med., 2008: 377–392) used various quality
control methods to study a method of producing such
scaffolds.

An unusual feature is that instead of subgroups being

observed over time, each subgroup resulted from a
different design dimension (m).
18
Example cont’d

Table 16.3 contains data from Table 2 of the cited article on

the deviation from target in the perpendicular orientation
(these deviations are indeed all positive—the printed
beams exhibit larger dimensions than those designed).

Deviation-from-Target Data for Example 5 (continued)

19
Example cont’d

Deviation-from-Target Data for Example 5 (continued)

Table 16.3

Table 16.3 yields rI = 124, from which = 7.29.

Since n = 3, LCL = 0.

With b3 = 1.693 and c3 = .888,

UCL = 7.29 + 3  (.888)(7.29)/1.693 = 18.76

20
Example cont’d

Figure 16.4 shows both an R chart and an chart from the

Minitab software package (the cited article also included
these charts).

Control charts for the deviation-from-target data of Example 5

Figure 16.4 21
Control Charts for Attributes

22
Control Charts for Attributes
The term attribute data is used in the quality control
literature to describe two situations:
1. Each item produced is either defective or nondefective
(conforms to specifications or does not).

2. A single item may have one or more defects, and the

number of defects is determined.

In the former case, a control chart is based on the binomial

distribution; in the latter case, the Poisson distribution is the
basis for a chart.

23
The p Chart for Fraction
Defective

24
The p Chart for Fraction Defective
Suppose that when a process is in control, the probability
that any particular item is defective is p (equivalently, p is
the long-run proportion of defective items for an in-control
process) and that different items are independent of one
another with respect to their conditions.

Consider a sample of n items obtained at a particular time,

and let X be the number of defectives and = X/n. Because
X has a binomial distribution, E(X) = np and
V(X) = np(1 – p), so

25
The p Chart for Fraction Defective
Also, if np  10 and n(1 – p)  10. has approximately a
normal distribution. In the case of known p (or a chart
based on target value), the control limits are

If each sample consists of n items, the number of defective

items in the ith sample is xi, and = xi /n, then
are plotted on the control chart.

26
The p Chart for Fraction Defective
Usually the value of p must be estimated from the data.
Suppose that k samples from what is believed to be an
in-control process are available, and let

The estimate p is then used in place of p in the

aforementioned control limits.

27
The p Chart for Fraction Defective
The p chart for the fraction of defective items has its center
line at height p and control limits

If LCL is negative, it is replaced by 0.

28
Example
A sample of 100 cups from a particular dinnerware pattern
was selected on each of 25 successive days, and each
was examined for defects.

The resulting numbers of unacceptable cups and

corresponding sample proportions are as follows:

29
Example cont’d

Assuming that the process was in control during this

period, let’s establish control limits and construct a p chart.
Since, = 1.52, p = 1.52/25 = .0608 and

30
Example cont’d

The LCL is therefore set at 0. The chart pictured in Figure

16.5 shows that all points are within the control limits. This
is consistent with an in-control process.

Control chart for fraction-defective data of Example 6

Figure 16.5
31
The c Chart for Number of
Defectives

(Self Study, in Syllabus)

32
The c Chart for Number of Defectives
We now consider situations in which the observation at
each time point is the number of defects in a unit of some
sort.

The unit may consist of a single item (e.g., one automobile)

or a group of items (e.g., blemishes on a set of four tires).
In the second case, the group size is assumed to be the
same at each time point.

The control chart for number of defectives is based on the

Poisson probability distribution.

33
The c Chart for Number of Defectives
Recall that if Y is a Poisson random variable with
parameter , then

E(Y) =  V(Y) =  Y =

Also, Y has approximately a normal distribution when  is

large (  10 will suffice for most purposes).

Furthermore, if Y1, Y1, . . . , Yn, are independent Poisson

variables with parameters 1, 2, . . . , n it can be shown
that Y1 + . . . + Yn has a Poisson distribution with parameter
1 + . . . + n.
34
The c Chart for Number of Defectives
In particular, if 1 = . . . = n =  (the distribution of the
number of defects per item is the same for each item), then
the Poisson parameter is n.

Let  denote the Poisson parameter for the number of

defects in a unit (it is the expected number of defects per
unit). In the case of known  (or a chart based on a target
value),

LCL =  – 3 UCL =  + 3

35
The c Chart for Number of Defectives
With xi denoting the total number of defects in the ith unit
(i = 1, 2, 3, . . .), then points at heights x1, x2, x3 , . . . are
plotted on the chart.

Usually the value of  must be estimated from the data.

Since E(Xi) = , it is natural to use the estimate  = x
(based on x1 , x2, . . . , , xk).

The c chart for the number of defectives in a unit has

center line at x and
LCL = x – 3
UCL = x + 3

If LCL is negative, it is replaced by 0. 36

Example
A company manufactures metal panels that are baked after
first being coated with a slurry of powdered ceramic. Flaws
sometimes appear in the finish of these panels, and the
company wishes to establish a control chart for the number
of flaws.

The number of flaws in each of the 24 panels sampled at

regular time intervals are as follows:

7 10 9 12 13 6 13 7 5 11 8 10
13 9 21 10 6 8 3 12 7 11 14 10

37
Example cont’d

with xi = 235 and = x = 235/24 = 9.79. The control limits

are

LCL = 9.79 – 3

= .40

UCL = 9.30 + 3

= 19.18

38
Example cont’d

The control chart is in Figure 16.6. The point corresponding

to the fifteenth panel lies above the UCL.

Control chart for number of flaws data of Example 7

Figure 16.6

39
Example cont’d

Upon investigation, the slurry used on that panel was

discovered to be of unusually low viscosity (an assignable
cause).

Eliminating that observation gives x = 214/23 = 9.30 and

new control limits

LCL = 9.30 – 3 = .15

UCL = 9.30 + 3 = 18.45

The remaining 23 observations all lie between these limits,

indicating an in-control process. 40