Control Chart for
Attributes
Bahagian 1
Introduction
Many quality characteristics cannot be
conveniently represented numerically.
In such cases, each item inspected is
classified as either conforming or
nonconforming to the specifications on
that quality characteristic.
Quality characteristics of this type are
called attributes.
Examples are nonfunctional semiconductor
chips, warped connecting rods, etc,.
Types of Control Charts
Control Charts for Variables Data
X and R charts: for sample averages and ranges.
X and s charts: for sample means and standard deviations.
Md and R charts: for sample medians and ranges.
X charts: for individual measures; uses moving ranges.
Control Charts for Attributes Data
p charts: proportion of units nonconforming.
np charts: number of units nonconforming.
c charts: count of nonconformities.
u charts: count of nonconformities per unit.
Type of Attribute Charts
p charts
This chart shows the fraction of nonconforming or defective
product produced by a manufacturing process.
It is also called the control chart for fraction nonconforming.
np charts
This chart shows the number of nonconforming. Almost the
same as the p chart.
c charts
This shows the number of defects or nonconformities
produced by a manufacturing process.
u charts
This chart shows the nonconformities per unit produced by a
manufacturing process.
p charts
In this chart, we plot the percent of
defectives (per batch, per day, per machine,
etc.).
However, the control limits in this chart are
not based on the distribution of rate events
but rather on the binomial distribution (of
proportions).
Formula
Fraction nonconforming:
p = (np)/n
where p = proportion or fraction nc in the
sample or subgroup, n = number in the
sample or subgroup, np = number nc in the
sample or subgroup.
Example
During the first shift, 450 inspection are made of
book-of the month shipments and 5 nc units are
found. Production during the shift was 15,000
units. What is the fraction nc?
p = (np)/n = 5/450 = 0.011
The p, is usually small, say 0.10 or less.
If p > 0.10, indicate that the organization is in
serious difficulty.
p-Chart contruction for
constant subgroup size
Select the quality characteristics.
Determine the subgroup size and method
Collect the data.
Calculate the trial central line and control
limits.
Establish the revised central line and
control limits.
Achieve the objective.
Select the quality
characteristics
The quality characteristic?
A single quality characteristic
A group of quality characteristics
A part
An entire product, or
A number of products.
Determine the subgroup size and
method
The size of subgroup is a function of the
proportion nonconforming.
If p = 0.001, and n = 1000, then the average
number nc, np = 1. Not good, since a large
number of values would be zero.
If p = 0.15, and n = 50, then np = 7.5, would
make a good chart.
Therefore, the selection subgroup size requires
some preliminary observations to obtain a rough
idea of the proportion nonconforming.
Collect the data
The quality technician will need to collect
sufficient data for at least 25 subgroups.
The data can be plotted as a run chart.
Since the run chart does not have limits, its is
not a control chart.
Calculate the trial central line and
control limits
The formula:
p (1 p )
UCL p 3
n
p (1 p )
LCL p 3
n
np = average of p for many subgroups
p
n=
number
inspected in a subgroup
n
np
138
p
0.018
n 7500
0.018(1 0.018)
UCL 0.018 3
300
0.041
0.018(1 0.018)
300
0.005 0.0
LCL 0.018 3
Subgroup
Number
Number
Inspected
n
np
300
12
0.040
300
0.010
300
0.030
300
0.013
300
0.0
300
0.020
300
0.020
300
0.003
19
300
16
0.053
25
300
0.007
7500
138
Total
Negative value of LCL is possible in a theoritical
result, but not in practical (proportion of nc never
negative).
p Chart
0.053
p
UCL
0.04
0.03
0.02
p-bar
0.01
LCL
5
10
15
20
Subgroup
25
Establish the revised central line
and control limits
Determine the standard or reference value for
the proportion nc, po.
pnew
np np
n n
where npd = number nc in the discarded
subgroups
nd = number inspected in the discarded
subgroups
Revised control limits
po pnew
po (1 po )
UCL po 3
n
po (1 po )
LCL po 3
n
where po is central line
pnew
138 16
0.017
7500 300
0.017(1 0.017)
UCL 0.017 3
300
0.039
0.017(1 0.017)
300
0.005 0.0
LCL 0.017 3
Achieve the objective
The first 5 steps are planning (p304-308).
The last step involves action and lead to the
achievement of the objective.
The revised control limits were based on data
collected in May.
For June, July, August? See Fig 8-3, p.310
Quality improved?
Analysis of June results shows the quality
improved.
Using June data, a better estimation of
proportion nc is obatained.
The new value: po = 0.014, UCL = 0.036
Data from July are used to determine CL &
UCL for August.
np Chart
The np chart is almost the same as the p chart.
Central line = npo
UCL npo 3 npo (1 po )
LCL
npo it3must
npbe
po ) by
If po
is unknown,
determined
o (1
collecting data, calculating UCL, LCL.
Example
c Chart
The procedures for c chart are the same a s
those for the p chart.
If count of nonconformities, co, is unknown, it
must be found by collecting data, calculating
UCL & LCL.
UCL c 3 c
LCL c 3 c
= average count of nonconformities
c
c
g
Example c
c 141
UCL 5.64 3 5.64 12.76
5.64
g
25
LCL 5.64 3 5.64
1.48 0
c-Chart
25
20
Count of Nonconformities
c
UCL
c-bar
15
LCL
10
0
1
10
11
12
13
14
15
Subgroup Num ber
16
17
18
19
20
21
22
23
24
25
Revised
Out-of-control: sample no. 5, 11, 23.
c c d 141 20 14 14
c new
4.23
g gd
25 3
UCL co 3 co 4.23 3 4.23 10.40
LCL co 3 co 4.23 3 4.23 1.94 0
u Chart
The u chart is mathematically equivalent
to the c chart.
c
u
n
u
n
u
UCL u 3
n
u
LCL u 3
n
Example
c 3389
1.20
n 2823
For January 30:
c 120
u Jan 30
1.09
n 110
1.20
UCL Jan 30 1.20 3
1.51
110
LCL Jan 30 1.20 3
1.20
0.89
110
Nonconformity Classification
Critical nonconformities
Major nonconformities
Indicate hazardous or unsafe conditions.
Failure
Minor nonconformities
Control Charts for Variables
vs. Charts for Attributes
Sometimes, the quality control engineer has
a choice between variable control charts and
attribute control charts.
Advantages of attribute
control charts
Allowing for quick summaries, that is, the engineer
may simply classify products as acceptable or
unacceptable, based on various quality criteria.
Thus, attribute charts sometimes bypass the need
for expensive, precise devices and time-consuming
measurement procedures.
More easily understood by managers unfamiliar with
quality control procedures.
Advantages of variable
control charts
More sensitive than attribute control charts.
Therefore, variable control charts may alert us to
quality problems before any actual "unacceptables"
(as detected by the attribute chart) will occur.
Montgomery (1985) calls the variable control charts
leading indicators of trouble that will sound an alarm
before the number of rejects (scrap) increases in the
production process.
