Quality Control and Assurance
Statistical Process Control-
Attribute Control Chart
Prepared by Aditya T. Pratama
2023
Reference :
Introduction to Statistical Quality Control, Douglas Montgomery
Managing for Quality and Performance Excellence, James Evans & William Lindsay
Fundamentals of Quality Control and Improvement, Amita
Quality Control and Assurance
Introduction
• Variable Data is quantitative data that can be
measured in units of measure.
• Examples of variable data, for example weight (kg),
length (m), width (m), height (m), diameter (m), volume
(m3), time (s), electric current (A), and so on.
• Attribute Data is qualitative data that can be
calculated as a recording of identified data.
• Attribute data is usually obtained in units of the number
of defects in products that do not comply with the
specified attribute specifications.
Quality Control and Assurance
Quality Control and Assurance
Attribute Control Chart
• P Control Chart
• NP Control Chart
• C Control Chart
• U Control Chart
Quality Control and Assurance
P-Control Chart
• The P control chart is used to control the
proportion of items that do not meet the specified
specification requirements (defective).
• For this reason, a precise operational definition of
non-conformity or defect is very important and
must be understood by every user of the P control
chart.
Quality Control and Assurance
P-Control Chart
Steps to develop the P Control Chart (proportion of defective
unit):
• Determine a sample or subgroup size that is large enough (n
> 30)
• Collect the number of subgroups (k) for 20-25 subgroups.
• Calculate for each subgroup the value of the proportion of
defective units:
p=number of defective units/subgroup size
• Calculate P Control Limit
LCL
• Plot data on the proportion or percentage of defective units
and observe whether the data is within or not within
control.
Quality Control and Assurance
NP-Control Chart
• Sometimes it is easier to present information
through an NP Control Chart which is based on the
number of defective units, compared to a P Control
Chart which is based on the proportion of defective
units.
• NP Control Charts can be used if the sample size (n)
is constant, and the proportion of defective units is
relatively small.
Quality Control and Assurance
NP-Control Chart
Steps to develop the NP-Control Chart:
• Determine the sample or subgroup size that is large enough
(n > 30).
• Collect the number of subgroups (k) around 20-25
subgroups.
• Calculate total defects (NP) for each subgroup.
• Calculate the average value of NP.
• Determine the control limit
• Plot the data from each subgroup examined and observe
whether the data is within control or not.
Quality Control and Assurance
C-Control Chart
• P and NP-Control Charts relate to the terms defective units or
products. In contrast to C and U-Control Charts, the size of the
number of defects or the number of types of defects contained in
a product also determines the quality of the product.
• C-Control Chart is used to determine the number of defects
arising from the products produced, not the number of defective
products.
• For example, let's say we will observe defects in the manufacture
of shirts, defects are defined as an insufficient number of buttons,
loose stitching and so on.
• In this case, C-Control Chart is used to determine the number of
defects in 1 shirt. So, for example, there are 10 defects or only 1
defect, this defect is defined as the number of defects in 1
product being 10 or only 1, not 10 defective shirt products.
• C-Control Chart is not only used for the number of defects in a
product but can be used for measurements that use the Poisson
Distribution (the number of events referred to in the
specifications for place or time.
Quality Control and Assurance
C-Control Chart
Steps to develop C-Control Chart:
• Collect k= the number of subgroups to be inspected.
Make sure that k is sufficient in number, such as k=20-25
subgroups.
• Count the number of defects for each subgroup.
• Calculate the average value of the number of defects (C):
• Calculate the control limits for C-Control Chart:
• Plot data on the number of defects for each subgroup
examined and observe whether the data is within
control or not.
Quality Control and Assurance
U-Control Chart
• U-Control Chart has similarities with C-Control
Chart.
• A C-Control chart plots the number of defects. The
U-Control chart helps us check the stability of a
single unit, which might have one or more defects.
• For example, the number of defects in one unit. A
U-chart plots the number of defects per “inspection
unit.”
Quality Control and Assurance
U-Control Chart
Steps for developing the U-Control Chart:
• Determine the sample or subgroup size.
• Collect the number of subgroups (k).
• Calculate for each subgroup the u value (the number of
defects per inspection unit) using the formula:
(ci = Number of defects in the ith subgroup, ni = Sample
size in the ith subgroup)
• Calculate the control limits for the U-control chart using the
formula:
Quality Control and Assurance
Summary
Quality Control and Assurance
DPMO Calculation
• After ensuring that the data is within control limits,
the DPMO value is calculated to determine the
probability of a defective product occurring per one
million possibilities.
• The DPMO value can also be used to calculate the
sigma value before improvement.
Quality Control and Assurance
DPMO Calculation
Example:
• Number of Production Per Shift: Based on the
production data, the number of production per
shift varies between 1200 to 1500. The total
production is 42,192 bottles.
• Number of Defective Units: the number of
defective units in the bottle defect data is 6930
bottles.
• Calculate the total opportunity value (TOP): the
total opportunity value is
TPO = 42192 × 4 = 168768
Quality Control and Assurance
DPMO Calculation
Example:
• Defect value per opportunity (DPO), the value of
defects per opportunity is
DPO = 6930/168768 = 0.04106
• Calculate Defect per Million Opportunity (DPMO):
• The results of the DPMO calculation show that the
possibility of defects occurring in a production volume
of 1,000,000 is 41062.
• This DPMO value will then be used to obtain the sigma
level by converting the DPMO value to the sigma level.
Quality Control and Assurance
Sigma Level
• After the DPMO calculation is carried out, the
Sigma level can be known.
• The formula for calculating the sigma level is:
Sigma Level = NORMSINV(1 – DPO) + 1.5
• By using previously existing data, the sigma level is
as follows:
Sigma level = NORMSINV (1 – 0.041) + 1.5 = 3.238
• The results of the sigma level calculation show that
the sigma level is 3,238 which can still be increased
by making improvements.
Quality Control and Assurance
Exercise P-Control Chart
Day Sample Defect
1 50 12
• PT. XYZ checks the 2
3
40
50
8
10
number of failures 4 50 7
generated after the 5 45 9
process of creating a form 6 44 11
7 50 10
for a newly introduced 8 50 12
administrative purpose. 9 52 13
The sample size was set at 10 60 15
a different size each day, 11 55 11
then the number of forms 12
13
56
50
14
10
that did not meet the 14 45 9
requirements were 15 42 7
recorded for observation 16 35 8
for 20 days as follows: 17
18
45
50
9
10
19 40 8
20 60 9
Quality Control and Assurance
Exercise P-Control Chart
Quality Control and Assurance
Exercise NP-Control Chart
Day Inspection Defective
1 50 4
• To monitor the 2 50 8
manufacturing process 3 50 6
of laptops, a quality 4 50 10
control engineer 5 50 4
randomly selects 50 6 50 3
laptops from the 7 50 4
production line, each 8 50 7
day over a period of 20 9 50 8
days. The laptops are 10 50 4
inspected for certain 11 50 6
defects and the 12 50 1
number of defective 13 50 5
laptops found each day 14 50 3
is recorded in the 15 50 2
following table: 16 50 3
17 50 7
18 50 9
19 50 2
20 50 4
Quality Control and Assurance
Exercise NP-Control Chart
Quality Control and Assurance
Exercise C-Control Chart
Day Accident Day Accident
• PX is conducting 1 5 16 2
research regarding
the number of 2 1 17 1
accidents that 3 0 18 0
occurred during the 4 6 19 0
last month. This
research is used to 5 3 20 1
record the causes of 6 2 21 2
accidents so that 7 3 22 4
more accidents can
be prevented in the 8 4 23 1
future. For this 9 5 24 3
reason, data on 10 1 25 2
accidents that 11 2 26 0
occurred during the
last 30 days were 12 2 27 1
collected as follows: 13 3 28 2
14 0 29 3
15 5 30 1
Quality Control and Assurance
Exercise C-Control Chart
Quality Control and Assurance
Exercise U-Control Chart
• A QC unit from the Observation Length Sampling Size Non-conformance
P-Steel company 1 500 10 14
2 400 8 12
wants to inspect 3 650 13 20
the steel sheets. 4 500 10 11
Because the sheets 5 475 9.5 7
are long, an 6 500 10 10
7 600 12 21
inspection is 8 525 10.5 15
determined every 9 600 12 19
50 m2 of steel 10 625 12.5 23
sheet. Inspection is
carried out for 10
steel coils:
Quality Control and Assurance
Exercise U-Control Chart
Quality Control and Assurance
….Thank you
