P-Charts:
Attribute Based Control Charts
WASEEM UL HASAN ( 036)
WAQAS AHMED BAIG ( 043)
UZMA ZAHEER KHAN ( 035)
CONTENTS
What is a Control Chart?
What is a p-Chart?
What information does a
p-Chart convey?
When to use p-chart ?
How are p-Charts developed?
An example of p-chart
OVERVIEW
QUALITY TOOLS
PARETO FISH BONE HISTOGRAM CONTROL CHARTS FLOW CHART CHECK SHEET
VARIABLE ATTRIBUTE
¯X/R
S CHART
P/NP
P CHART C CHART U CHART
CHART CHART
What is a Control Chart?
A Control Chart is a graphical display of
process information which compares
item attributes or quantitative values
against a standard or reference value,
within a series of upper and lower
constraint values
Why are control charts used?
To determine if the rate of production of
nonconforming products is stable
To detect when a deviation from process stability
has occurred
Control charts are good for:
Improving Productivity
Preventing Defects
Provide Diagnostic Information
Provide Information About Process
Capability
TYPES OF CONTROL CHART
VARIABLE ATTRIBUTE
Variables control charts are Attribute control charts are
applied to numerical applied to discrete and
continuous measurements categorical measurements –
e.g., each item is classified as
Variables generate data that conforming (nondefective) or
are measured. nonconforming (defective).
Attributes generate
data that are counted.
What is a p-Chart?
A process control chart that measures a
proportion of defective or nonconforming
items within a sample or population
What information does a p-
Chart convey?
If any one of the inspected attributes is
nonconforming, the entire item is counted as
nonconforming
The number of items in the sample that are
determined to be nonconforming are summed and a
proportion of the total is evaluated
What information does a p-
Chart convey?
The p-Chart is a graph of the proportion of
nonconforming items in each sample or population
The graph is then used to determine whether or not
a process is stable
Use of p-Chart
When observations can be placed into
two categories.
Good or bad
Pass or fail
Operate or don’t operate
When the data consists of multiple
samples of several observations each
STEPS TO CREATE P CHART
Monitor proportion in noncompliance
1. Decide on the quality to be measured.
2. Determine a sample size.
3. Gather samples.
4. Compute the sample proportion for
each sample.
5. Determine the average sample
proportion for all samples.
6. Compute the UCL and the LCL
Example dataset for a
p-Chart (Equal Sample Sizes)
Nonconforming Sample Size Proportion
Sample # of defective”x “ “n” “x / n”
The proportion of 1 10 50 0.200
defective or 2 11 50 0.220
3 10 50 0.200
nonconforming items 4 9 50 0.180
in each sample is 5 8 50 0.160
calculated by dividing 6 11 50 0.220
the number defective 7 10 50 0.200
8 9 50 0.180
by the sample size 9 10 50 0.200
10 9 50 0.180
11 11 50 0.220
12 13 50 0.260
13 9 50 0.180
14 8 50 0.160
15 9 50 0.180
Σx 147 750 Σn
Required inputs for P chart
p
Defects
x
# samples sample _ size n
p 1 p
n
UCL / LCL p z
Creating a p-Chart with
equal sample sizes
With the Mean Subgroup Proportion, standard error,
and upper / lower control limits determined, fill out the
table with the calculated data:
Sample Nonconforming Sample Size Proportion UCL (0.359) p-bar (0.192) LCL (0.025)
1 10 50 0.200 0.359 0.192 0.025
2 11 50 0.220 0.359 0.192 0.025
3 10 50 0.200 0.359 0.192 0.025
4 9 50 0.180 0.359 0.192 0.025
5 8 50 0.160 0.359 0.192 0.025
6 11 50 0.220 0.359 0.192 0.025
Creating a p-Chart with
equal sample sizes
p-Chart
0.400
UCL (0.359)
0.350
0.300
Proportion
0.250
0.200
0.150
0.100
0.050 LCL (0.025)
0.000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Obse rv ations (Sample Numbe r)
Proportion UCL (0.359) p-bar (0.192) LCL (0.025)
Example dataset for a
p-Chart (Unequal Sample Sizes)
Subgroup
Sample Nonconforming Sample Size Proportion
The proportion of 1 10 50 0.200
defective or 2 11 51 0.216
nonconforming 3 10 48 0.208
items in each 4 9 47 0.191
5 8 50 0.160
sample is
6 11 55 0.200
calculated by 7 10 54 0.185
dividing the 8 9 51 0.176
number defective 9 10 56 0.179
by the sample 10 9 43 0.209
11 11 44 0.250
size 12 13 51 0.255
13 9 49 0.184
14 8 49 0.163
15 7 53 0.132
Creating a p-Chart with
unequal sample sizes
With the Mean Subgroup Proportion, and upper / lower
control limits determined, fill out the table with the
calculated data (note the UCL / LCL will not graph as
straight lines):
Sample Nonconforming Sample Size Proportion p-bar UCL LCL
1 10 50 0.200 0.192 0.362 0.022
2 11 51 0.216 0.192 0.365 0.019
3 10 48 0.208 0.192 0.368 0.016
4 9 47 0.191 0.192 0.364 0.020
5 8 50 0.160 0.192 0.347 0.036
6 11 52 0.212 0.192 0.362 0.022
7 10 51 0.196 0.192 0.359 0.025
8 9 50 0.180 0.192 0.355 0.029
9 10 49 0.204 0.192 0.365 0.019
Creating a p-Chart with
unequal sample sizes
p-Chart
0.450
0.400
UCL
0.350
0.300
Proportion
0.250
0.200
0.150
0.100
0.050 LCL
0.000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Obse rv ations (Sample Numbe r)
Proportion p-bar UCL LCL
Evaluating the p-Chart
Three conditions or trends which warrant immediate attention:
Five sample means in a row above or below the target or reference line
Six sample means in a row that are steadily increasing or decreasing (trending in one direction)
Fourteen sample means in a row alternating above and below the target or reference line
P-Chart Example
Quality assurance manager for a vendor is notified by an
OEM quality department that several recent orders have
been rejected due to nonconforming defects that were
unacceptable. The OEM identified three separate defect
categories; however, any one defect would cause the
whole part to be rejected.
QA manager decided to evaluate the process by running
several batches through production and then counting the
number of parts that fail inspection for any reason.
P-Chart Example
P-Chart Example
Mean Subgroup Proportion: 0.049
Standard Error: 0.021
Upper Control Limit: 0.115
Lower Control Limit: 0.000*
*Actually calculated -0.016, but a negative number
is not a legitimate number of defects, therefore
0.000 is used as a realistic substitute
P-Chart Example
P-Chart Example
Process Control Chart (Proportion Defective)
0.14
0.12
0.1
Proportion
0.08
0.06
0.04
0.02
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Observations
Proportion Mean Subgroup Proportion UCL (0.115) LCL (0.000)
P-Chart Example
Conclusion:
The process is trending out of control:
- Five sample means in a row, above the reference line
- More than six sample means on an increasing trend, with
some alternation; however, the trend is clearly increasing at
the end
Recommend: Shut down the production line and evaluate
THANK YOU
