Overview for Individuals Chart
Learn more about Minitab 19
Use Individuals Chart to monitor the mean of your process when you have
continuous data that are individual observations that are not in
subgroups. Use this control chart to monitor process stability over time so
that you can identify and correct instabilities in a process.
For example, a hospital administrator wants to determine whether the time
to perform outpatient hernia surgery is stable. Because the data are not
collected in subgroups, the administrator uses an individuals chart (I chart) to
monitor the mean of the surgery times.
The points vary randomly around the center line and are within the control limits. No
trends or patterns are present. The amount of time to perform hernia surgery in times
are stable.
NOTE
The process variation must be in control before you assess the process
center using an individuals chart because the control limits for the individuals
chart are calculated using the within-subgroup variation. If the process
variation is not stable, then the individuals chart control limits will be
questionable and the chart might show a false out-of-control signal. Use a
moving range chart to determine whether the process variation is in control.
Where to find this control chart
To create an individuals chart, choose Stat > Control Charts > Variables
Charts for Individuals > Individuals.
When to use an alternate control chart
If you can collect data in subgroups, use Xbar Chart.
If your data are counts of defectives or defects, use an attribute control
chart, such as P Chart or U Chart.
Data considerations for Individuals
Chart
Learn more about Minitab 19
To ensure that your results are valid, consider the following guidelines when
you collect data, perform the analysis, and interpret your results.
The data should be continuous
If your data are counts of defectives or defects, use an attribute control
chart, such as P Chart or U Chart.
The data should be in time order
Because control charts detect changes over time, the order of the data
is important. You should enter the data in the order it was collected,
with the oldest data at the top of the worksheet.
The data should be collected at appropriate time intervals
Collect data at equally spaced time intervals, such as every hour, every
shift, or every day. Select a time interval that is short enough that you
can identify changes to the process soon after the changes occur.
The data should be individual observations that are not
collected in subgroups
If you can collect data in subgroups, use Xbar Chart.
The data should include at least 100 total observations
If you have fewer than the recommended number of observations, then
you can still use the control chart, but the results are preliminary
because the control limits may not be precise. If you use the chart
regularly, re-estimate the standard deviation and the control limits
after you collect the recommended number of observations.
The process variation should be stable
If the process variation is not stable, the control limits may not
adequately reflect the control status of the process.
The data should be moderately normal
If the data are very skewed, you could try a Box-Cox transformation to
see if that corrects the nonnormal condition. If your process naturally
produces nonnormal data and the transformation is effective, you can
use the chart of the transformed data to assess the stability of your
process.
The observations should not be correlated with each
other
If consecutive data points are correlated, the control limits will be too
narrow and you may see a large number of false out-of-control signals.
Example of Individuals Chart
A quality engineer monitors the manufacture of liquid detergent and wants to
assess whether the process is in control. The engineer measures the pH of 25
consecutive batches of detergent.
The engineer creates an Individuals chart to monitor the detergent process.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Variables Charts for
Individuals > Individuals.
3. In Variables, enter pH.
4. Click I Chart Options.
5. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line (Test 2).
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2.
6. Click OK in each dialog box.
Interpret the results
One observation, failed Test 1 on the individuals chart (I chart) because the
observation is more than 3 standard deviations above the center line. The
engineer concludes that the pH level of the detergent is not stable.
I Chart of pH
Test Results for I Chart of pH
TEST 1. One point more than 3.00 standard deviations from center
line.
Test Failed at points: 8
* WARNING * If graph is updated with new data, the results above
may no longer
be correct.
Overview for P Chart
Learn more about Minitab 19
Use P Chart to monitor the proportion of defective items where each item can
be classified into one of two categories, such as pass or fail. Use this control
chart to monitor process stability over time so that you can identify and
correct instabilities in a process.
For example, a delivery service manager uses a P chart to monitor the
proportion of delivery vehicles that are out of service each day for 2 months.
A vehicle that is out of service is considered a defective item.
The chart shows that, on average, 8% of the delivery vehicles are out of
service on any particular day. The proportion of defective units for day 19 is
out of control. The manager should identify any special causes that
contribute to the unusually high rate of defectives.
Where to find this control chart
To create a P chart, choose Stat > Control Charts > Attributes
Charts > P.
When to use an alternate control chart
If you can count the number of defects on each item, use U
Chart, Laney U' Chart, or C Chart to plot the number of defects per unit.
If your data exhibit overdispersion or underdispersion, Laney P'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause a
traditional P chart to show an increased number of points outside the
control limits. Underdispersion can cause a traditional P chart to show
too few points outside of the control limits. The Laney P' chart adjusts
for these conditions. You can test your data for overdispersion and
underdispersion with P Chart Diagnostic. For more information, go
to Overdispersion and underdispersion.
NOTE
The NP chart also plots defective items. However, whereas the P chart plots
the proportion of defective items, the NP chart plots the number of defective
items.
Data considerations for P Chart
Learn more about Minitab 19
To ensure that your results are valid, consider the following guidelines when
you collect data, perform the analysis, and interpret your results.
Items must be classified into one of two categories, such as pass or
fail
A defective item has one or more defects that make it not
acceptable. If you can determine only whether an item is defective or
nondefective, use this chart. If you can count the number of defects on
each item, use U Chart, Laney U' Chart, or C Chart to plot the number
of defects per unit.
If your data exhibit overdispersion or underdispersion, a
traditional attributes chart may be misleading
If your data exhibit overdispersion or underdispersion, Laney P'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause a
traditional P chart to show an increased number of points outside the
control limits. Underdispersion can cause a traditional P chart to show
too few points outside of the control limits. The Laney P' chart adjusts
for these conditions. You can test your data for overdispersion and
underdispersion with P Chart Diagnostic. For more information, go
to Overdispersion and underdispersion.
The data should be in time order
Because control charts detect changes over time, the order of the data
is important. You should enter the data in the order it was collected,
with the oldest data at the top of the worksheet.
The data should be collected at appropriate time intervals
Collect data at equally spaced time intervals, such as every hour, every
shift, or every day. Select a time interval that is short enough that you
can identify changes to the process soon after the changes occur.
Collect data in subgroups
A subgroup is a sample of similar items from the process that you want
to evaluate. The items in each subgroup should be collected under the
same process conditions, such as personnel, equipment, suppliers, or
environment.
If the subgroup is a collection of units, they should be collected under
the same process conditions, such as personnel, equipment, suppliers,
or environment. If you don't collect data in subgroups where items are
subject to the same process conditions, you may not be able to
distinguish between common-cause and special-cause variation.
The subgroups must be large enough
If the subgroup sizes are not large enough, the control limits may not
be accurate when they are estimated from the data. The required
subgroup size ( ) depends on the average proportion of defective items
( ). Use the following formula to determine whether your subgroups
are large enough, . For example, if the average proportion of
defective items is 0.06, then all subgroups must have at least 9
items: , rounded up to the nearest whole number = 9.
The data must include enough subgroups to obtain
precise control limits
If you don't have enough subgroups, you can still use the control chart,
but you should consider the results preliminary because the control
limits may not be precise. If you are using the chart on an ongoing
basis, re-estimate the control limits after you have collected enough
subgroups.
Example of P Chart
Learn more about Minitab 19
The supervisor for a call center wants to evaluate the process for answering
customer phone calls. The supervisor records the total number of incoming
calls and the number of unanswered calls for 21 days. The supervisor creates
a P chart to monitor the proportion of unanswered calls.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Attributes Charts > P.
3. In Variables, enter Unanswered Calls.
4. In Subgroup sizes, enter Total Calls.
5. Click P Chart Options.
6. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line (Test 2).
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2 when you first establish the control limits based on your
data.
7. Click OK in each dialog box.
Interpret the results
The chart shows that, on average, 9.57% of calls are unanswered. None of
the subgroup proportions are outside of the control limits. Furthermore, the
points inside the limits display a random pattern. This P chart does not
provide any evidence for lack of control. Thus, the process is in control.
Overview for U Chart
Learn more about Minitab 19
Use U Chart to monitor the number of defects per unit, where each item can
have multiple defects. Use this control chart to monitor process stability over
time so that you can identify and correct instabilities in a process.
For example, an LCD manufacturer wants to monitor the number of dead
pixels on 17-inch LCD screens. Technicians record the number of dead pixels
for each screen. Each subgroup has a different number of screens. The
manufacturer uses a U chart to monitor the average number of dead pixels
per screen.
Where to find this control chart
To create a U chart, choose Stat > Control Charts > Attributes
Charts > U.
When to use an alternate control chart
If you can determine only whether each item is defective or
nondefective, use P Chart or Laney P' Chart to plot the proportion of
defective items, or use NP Chart to plot the number of defective items.
If your data exhibit overdispersion or underdispersion, Laney U'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause a
traditional U chart to show an increased number of points outside the
control limits. Underdispersion can cause a traditional U chart to show
too few points outside of the control limits. The Laney U' chart adjusts
for these conditions. You can test your data for overdispersion and
underdispersion with U Chart Diagnostic. For more information, go
to Overdispersion and underdispersion.
Data considerations for U Chart
Learn more about Minitab 19
To ensure that your results are valid, consider the following guidelines when
you collect data, perform the analysis, and interpret your results.
You must be able to count the number of defects on each item or
unit
If you can determine only whether each item is defective or
nondefective, use P Chart or Laney P' Chart to plot the proportion of
defective items, or use NP Chart to plot the number of defective items.
If your data exhibit overdispersion or underdispersion, a
traditional attributes chart may be misleading
If your data exhibit overdispersion or underdispersion, Laney U'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause a
traditional U chart to show an increased number of points outside the
control limits. Underdispersion can cause a traditional U chart to show
too few points outside of the control limits. The Laney U' chart adjusts
for these conditions. You can test your data for overdispersion and
underdispersion with U Chart Diagnostic. For more information, go
to Overdispersion and underdispersion
The data should be in time order
Because control charts detect changes over time, the order of the data
is important. You should enter the data in the order it was collected,
with the oldest data at the top of the worksheet.
The data should be collected at appropriate time intervals
Collect data at equally spaced time intervals, such as every hour, every
shift, or every day. Select a time interval that is short enough that you
can identify changes to the process soon after the changes occur.
Collect data in subgroups
A subgroup is a collection of similar items that are representative of
the output from the process you want to evaluate. A subgroup can be
either a single unit or a collection of similarly sized units. For example,
you could record the number of surface imperfections for one LCD
panel (a single unit) or for a collection of LCD panels that are all the
same size.
If the subgroup is a collection of units, they should be collected under
the same process conditions, such as personnel, equipment, suppliers,
or environment. If you don't collect data in subgroups where items are
subject to the same process conditions, you may not be able to
distinguish between common-cause and special-cause variation.
The subgroups must be large enough
If the subgroups are not large enough, the control limits may not be
accurate when they are estimated from the data. The required
subgroup size ( ) depends on the average number of defects per unit (
). Use the following formula to determine whether your subgroups are
large enough, . For example, if the average number of defects
per unit is 0.06, then all subgroups must have at least 9
units: , rounded up to the nearest whole number = 9.
You must collect enough subgroups to obtain precise
control limits
If you don't have enough subgroups, you can still use the control chart,
but you should consider the results preliminary because the control
limits may not be precise. If you are using the chart on an ongoing
basis, re-estimate the control limits after you have collected enough
subgroups.
Example of U Chart
Learn more about Minitab 19
A manager for a transcription company wants to assess the quality of the
transcription service. The manager randomly selects 25 sets of pages from
consecutive orders and counts the number of typographical errors (defects).
Each set has a different number of pages.
The manager creates a U chart to monitor the number of errors.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Attributes Charts > U.
3. In Variables, enter Errors.
4. In Subgroup sizes, enter Number of Pages.
5. Click U Chart Options.
6. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line (Test 2).
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2 when you first establish the control limits based on your
data.
7. Click OK in each dialog box.
Interpret the results
Because the sample sizes are unequal, the control limits vary. The average
number of defects per set of pages is 0.238. Subgroups 6 and 18 failed Test 1
because they are outside of the control limits. Thus, the process is out of
control. The manager should identify and correct any factors that contribute
to the special-cause variation.
U Chart of Errors
Test Results for U Chart of Errors
TEST 1. One point more than 3.00 standard deviations from center
line.
Test Failed at points: 6, 18
* WARNING * If graph is updated with new data, the results above
may no longer
be correct.
Overview for C Chart
Learn more about Minitab 18
Use C Chart to monitor the number of defects where each item can have
multiple defects. You should use a C chart only when your subgroups are the
same size. Use this control chart to monitor process stability over time so
that you can identify and correct instabilities in a process.
For example, an LCD manufacturer wants to monitor defects on 17-inch LCD
screens. Technicians record the number of dead pixels for each subgroup of
10 screens per hour and use a C chart to monitor the number of dead pixels.
The chart shows that, on average, technicians find 10 dead pixels in each
sample. Sample 17 is out of control. The technicians should identify any
special causes that contribute to the unusually high number of dead pixels.
Where to find this control chart
To create a C chart, choose Stat > Control Charts > Attributes
Charts > C.
When to use an alternate control chart
If your subgroup sizes are not equal, use U Chart, or Laney U' Chart.
If you can determine only whether each item is defective or
nondefective, use P Chart or Laney P' Chart to plot the proportion of
defective items, or use NP Chart to plot the number of defective items.
If your data exhibit overdispersion or underdispersion, Laney U'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause a C
chart to show an increased number of points outside the control limits.
Underdispersion can cause a C chart to show too few points outside of
the control limits. The Laney U' chart adjusts for these conditions. You
can test your data for overdispersion and underdispersion with U Chart
Diagnostic. For more information, go to Overdispersion and
underdispersion.
Data considerations for C Chart
Learn more about Minitab 18
To ensure that your results are valid, consider the following guidelines when
you collect data, perform the analysis, and interpret your results.
You must be able to count the number of defects on each item or
unit
If you can determine only whether each item is defective or
nondefective, use P Chart or Laney P' Chart to plot the proportion of
defective items, or use NP Chart to plot the number of defective items.
If your data exhibit overdispersion or underdispersion, a
traditional attributes chart may be misleading
If your data exhibit overdispersion or underdispersion, Laney U'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause a C
chart to show an increased number of points outside the control limits.
Underdispersion can cause a C chart to show too few points outside of
the control limits. The Laney U' chart adjusts for these conditions. You
can test your data for overdispersion and underdispersion with U Chart
Diagnostic. For more information, go to Overdispersion and
underdispersion
The data should be in time order
Because control charts detect changes over time, the order of the data
is important. You should enter the data in the order it was collected,
with the oldest data at the top of the worksheet.
The data should be collected at appropriate time intervals
Collect data at equally spaced time intervals, such as every hour, every
shift, or every day. Select a time interval that is short enough that you
can identify changes to the process soon after the changes occur.
Collect data in subgroups
A subgroup is a collection of similar items that are representative of
the output from the process you want to evaluate. A subgroup can be
either a single unit or a collection of similarly sized units. For example,
you could record the number of surface imperfections for one LCD
panel (a single unit) or for a collection of LCD panels that are all the
same size.
If the subgroup is a collection of units, they should be collected under
the same process conditions, such as personnel, equipment, suppliers,
or environment. If you don't collect data in subgroups where items are
subject to the same process conditions, you may not be able to
distinguish between common-cause and special-cause variation.
Subgroup sizes should be equal, or nearly equal
For example, an LCD manufacturer wants to monitor defects on 17-inch
LCD screens. Technicians record the number of dead pixels for each
subgroup of 10 screens per hour and use a C chart to monitor the
number of dead pixels.
If your subgroup sizes are not equal, use U Chart, or Laney U' Chart.
The subgroups must be large enough
If the subgroups are not large enough, the control limits may not be
accurate when they are estimated from the data. The required
subgroup size ( ) depends on the average number of defects per unit (
). Use the following formula to determine whether your subgroups are
large enough, . For example, if the average number of defects
per unit is 0.06, then all subgroups must have at least 9
units: , rounded up to the nearest whole number = 9.
You must collect enough subgroups to obtain precise
control limits
If you don't have enough subgroups, you can still use the control chart,
but you should consider the results preliminary because the control
limits may not be precise. If you are using the chart on an ongoing
basis, re-estimate the control limits after you have collected enough
subgroups.
Example of C Chart
A quality engineer for a wallpaper manufacturer wants to assess the stability
of the printing process. Every hour, the engineer takes a sample of 100 feet
of wallpaper and counts the number of printing defects, which include print
smears, pattern distortions, and missing ink.
The engineer creates a C chart to monitor the number of defects.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Attributes Charts > C.
3. In Variables, enter Defects.
4. Click C Chart Options.
5. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line (Test 2).
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2 when you first establish the control limits based on your
data.
6. Click OK in each dialog box.
Interpret the results
The average number of defects per sample is 36.68. Samples 12 and 13
failed Test 1 because they are outside the control limits. Thus, the process is
out of control. The engineer should identify and correct any factors that
contribute to the special-cause variation.
TEST 1. One point more than 3.00 standard deviations from center line.
Test Failed at points: 12, 13
* WARNING * If graph is updated with new data, the results above may no
longer
be correct.
Overview for Moving Range Chart
Learn more about Minitab 19
Use Moving Range Chart to monitor the variation of your process when you
have continuous data that are individual observations not in subgroups. Use
this control chart to monitor process stability over time so that you can
identify and correct instabilities in a process.
A hospital administrator wants to monitor whether the variation in surgery
times is stable. Because the data are not collected in subgroups, the
administrator uses a moving range chart to monitor the variation of the
surgery times.
The points vary randomly around the center line and are within the control limits. No
trends or patterns are present. The variation in times are stable.
Where to find this control chart
To create a moving range chart, choose Stat > Control Charts > Variables
Charts for Individuals > Moving Range.
When to use an alternate control chart
If you can collect data in subgroups, use R Chart or S Chart.
If your data are counts of defectives or defects, use an attribute control
chart, such as P Chart or U Chart.
Data considerations for Moving
Range Chart
Learn more about Minitab 19
To ensure that your results are valid, consider the following guidelines when
you collect data, perform the analysis, and interpret your results.
The data should be continuous
If your data are counts of defectives or defects, use an attribute control
chart, such as P Chart or U Chart.
The data should be in time order
Because control charts detect changes over time, the order of the data
is important. You should enter the data in the order it was collected,
with the oldest data at the top of the worksheet.
The data should be collected at appropriate time intervals
Collect data at equally spaced time intervals, such as every hour, every
shift, or every day. Select a time interval that is short enough that you
can identify changes to the process soon after the changes occur.
The data should be individual observations that are not
collected in subgroups
If you can collect data in subgroups, use Xbar-R Chart or Xbar-S Chart.
The data should include at least 100 total observations
If you have fewer than the recommended number of observations, then
you can still use the control chart, but the results are preliminary
because the control limits may not be precise. If you use the chart
regularly, re-estimate the standard deviation and the control limits
after you collect the recommended number of observations.
The data should be moderately normal
If the data are very skewed, you could try a Box-Cox transformation to
see if that corrects the nonnormal condition. If your process naturally
produces nonnormal data and the transformation is effective, you can
use the chart of the transformed data to assess the stability of your
process.
The observations should not be correlated with each
other
If consecutive data points are correlated, the control limits will be too
narrow and you may see a large number of false out-of-control signals.
Example of Moving Range Chart
A quality engineer monitors the manufacture of liquid detergent and wants to
assess whether the process is in control. The engineer measures the pH of 25
consecutive batches of detergent.
The engineer creates a Moving Range chart to monitor the detergent process.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Variables Charts for
Individuals > Moving Range.
3. In Variables, enter pH.
4. Click MR Options.
5. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line (Test 2).
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2.
6. Click OK in each dialog box.
Interpret the results
None of the points are outside the control limits, and the points display a
random pattern on the moving range chart. Thus, the process variation is in
control, and the engineer can use the Individuals chart to examine the
process center.
Overview for S Chart
Learn more about Minitab Statistical Software
Use S Chart to monitor the variation (standard deviation) of your process
when you have continuous data and subgroups sizes of 9 or more. Use this
control chart to monitor process stability over time so that you can identify
and correct instabilities in a process.
For example, a paint manufacturer wants to determine whether the filling
process for cans of paints is stable. Operators collect subgroups of 10 cans
every hour for 30 hours and use an S chart to monitor the variation of the
filled paint cans.
The points vary randomly around the center line and are within the control limits. No
trends or patterns are present. The variation of the weights of cans are stable across the
30 subgroups.
Where to find this control chart
To create an S chart, choose Stat > Control Charts > Variables Charts
for Subgroups > S.
When to use an alternate control chart
For subgroups that have 2−8 observations, use R Chart.
If you do not have subgroups, use Moving Range Chart.
If your data are counts of defectives or defects, use an attribute control
chart, such as P Chart or U Chart.
Data considerations for S Chart
Learn more about Minitab Statistical Software
To ensure that your results are valid, consider the following guidelines when
you collect data, perform the analysis, and interpret your results.
The data should be continuous
If your data are counts of defectives or defects, use an attribute control
chart, such as P Chart or U Chart.
The data should be in time order
Because control charts detect changes over time, the order of the data
is important. You should enter the data in the order it was collected,
with the oldest data at the top of the worksheet.
The data should be collected at appropriate time intervals
Collect data at equally spaced time intervals, such as every hour, every
shift, or every day. Select a time interval that is short enough that you
can identify changes to the process soon after the changes occur.
The data should be in rational subgroups
A rational subgroup is a small sample of items that are similar, that are
produced in a short period of time under the same conditions (such as
operator, equipment, or supplier), and that are representative of the
output from a process. If the subgroups are not rational subgroups, the
estimated control limits might be too wide.
The subgroup size should be 9 or more observations
For subgroups that have 2−8 observations, use R Chart.
If you do not have subgroups, use Moving Range Chart.
The data should include at least 60 total observations
If you have fewer than the recommended number of observations, then
you can still use the control chart, but the results are preliminary
because the control limits may not be precise. If you use the chart
regularly, re-estimate the standard deviation and the control limits
after you collect the recommended number of observations.
The data do not need to be normally distributed
Although most control charts for variables data are formally based on
the assumption of normality, you can still obtain good results with
nonnormal data if you collect data in subgroups. The required subgroup
size depends on how nonnormal the data are.
The observations within each subgroup should not be
correlated with each other
If consecutive data points within each subgroup are correlated, the
control limits will be too narrow, and the control chart might incorrectly
show some in-control points as out of control.
Example of S Chart
A quality engineer at a canning company assesses whether the can-filling
process is in control. Each hour, the engineer collects a subgroup of 10 cans.
To minimize the within-subgroup (can-to-can) variation, the engineer collects
the cans for a given subgroup in a short period of time.
The quality engineer creates a S chart to monitor the variation in the weight
of the cans.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Variables Charts for
Subgroups > S.
3. From the drop-down list, select All observations for a chart are in
one column, enter Weight.
4. In Subgroup sizes, enter Subgroup ID.
5. Click S Options.
6. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line (Test 2).
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2.
7. Click OK in each dialog box.
Interpret the results
None of the points are outside the control limits, and the points display a
random pattern. Thus, the process variation is in control, and the engineer
can examine the process center on the Xbar chart.
Example of I-MR Chart
A quality engineer monitors the manufacture of liquid detergent and wants to
assess whether the process is in control. The engineer measures the pH of 25
consecutive batches of detergent.
The engineer creates an I-MR chart to monitor the detergent process.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Variables Charts for
Individuals > I-MR.
3. In Variables, enter pH.
4. Click I-MR Options.
5. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line.
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2.
6. Click OK in each dialog box.
Interpret the results
Interpret the Moving Range chart (MR chart) first to examine the process
variation. None of the points are outside the control limits and the points
display a random pattern. Thus, the process variation is in control, and the
quality engineer can examine the process center on the Individuals chart (I
chart).
One observation failed Test 1 on the I chart because the observation is more
than 3 standard deviations above the center line.
I-MR Chart of pH
Test Results for I Chart of pH
TEST 1. One point more than 3.00 standard deviations from center
line.
Test Failed at points: 8
* WARNING * If graph is updated with new data, the results above
may no longer
be correct.
Overview for NP Chart
Learn more about Minitab 18
Use NP Chart to monitor the number of defective items where each item can
be classified into one of two categories, such as pass or fail. Use this control
chart to monitor process stability over time so that you can identify and
correct instabilities in a process.
For example, a delivery service manager uses an NP chart to monitor the
number of delivery vehicles that are out of service each day for 2 months. A
vehicle that is out of service is considered a defective unit.
The chart shows that, on average, 25 of the delivery vehicles are out of
service each day. The number of defective units for day 19 is out of control.
The manager should identify any special causes that contribute to the
unusually high number of defectives.
Where to find this control chart
To create an NP chart, choose Stat > Control Charts > Attributes
Charts > NP.
When to use an alternate control chart
If you can count the number of defects on each item, use U
Chart, Laney U' Chart, or C Chart to plot the number of defects per unit.
If your subgroup sizes are not equal and you want the center line to be
straight, use P Chart or Laney P' Chart to plot the proportion of
defective items.
If your data exhibit overdispersion or underdispersion, Laney P'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause an NP
chart to show an increased number of points outside the control limits.
Underdispersion can cause an NP chart to show too few points outside
of the control limits. The Laney P' chart adjusts for these
conditions. You can test your data for overdispersion and
underdispersion with P Chart Diagnostic. For more information, go
to Overdispersion and underdispersion.
Data considerations for NP Chart
Learn more about Minitab 18
To ensure that your results are valid, consider the following guidelines when
you collect data, perform the analysis, and interpret your results.
Items must be classified into one of two categories, such as pass or
fail
A defective item has one or more defects that make it not
acceptable. If you can determine only whether an item is defective or
nondefective, use this chart. If you can count the number of defects on
each item, use U Chart, Laney U' Chart, or C Chart to plot the number
of defects per unit.
If your data exhibit overdispersion or underdispersion, a
traditional attributes chart may be misleading
If your data exhibit overdispersion or underdispersion, Laney P'
Chart may more accurately distinguish between common-cause
variation and special-cause variation. Overdispersion can cause an NP
chart to show an increased number of points outside the control limits.
Underdispersion can cause an NP chart to show too few points outside
of the control limits. The Laney P' chart adjusts for these
conditions. You can test your data for overdispersion and
underdispersion with P Chart Diagnostic. For more information, go
to Overdispersion and underdispersion.
The data should be in time order
Because control charts detect changes over time, the order of the data
is important. You should enter the data in the order it was collected,
with the oldest data at the top of the worksheet.
The data should be collected at appropriate time intervals
Collect data at equally spaced time intervals, such as every hour, every
shift, or every day. Select a time interval that is short enough that you
can identify changes to the process soon after the changes occur.
Collect data in subgroups
A subgroup is a sample of similar items from the process that you want
to evaluate. The items in each subgroup should be collected under the
same process conditions, such as personnel, equipment, suppliers, or
environment.
If the subgroup is a collection of units, they should be collected under
the same process conditions, such as personnel, equipment, suppliers,
or environment. If you don't collect data in subgroups where items are
subject to the same process conditions, you may not be able to
distinguish between common-cause and special-cause variation.
The subgroups must be large enough
If the subgroup sizes are not large enough, the control limits may not
be accurate when they are estimated from the data. The required
subgroup size ( ) depends on the average proportion of defective items
( ). Use the following formula to determine whether your subgroups
are large enough, . For example, if the average proportion of
defective items is 0.06, then all subgroups must have at least 9
items: , rounded up to the nearest whole number = 9.
The data must include enough subgroups to obtain
precise control limits
If you don't have enough subgroups, you can still use the control chart,
but you should consider the results preliminary because the control limits
may not be precise. If you are using the chart on an ongoing basis, re-
estimate the control limits after you have collected enough subgroups.
Example of NP Chart
A quality engineer assesses whether the process used to manufacture light
bulbs is in control. The engineer tests 500 light bulbs each hour for three 8-
hour shifts and records the number of bulbs that did not light (defectives).
The engineer creates an NP chart to monitor the number of defective light
bulbs.
1. Open the sample data, [Link].
2. Choose Stat > Control Charts > Attributes Charts > NP.
3. In Variables, enter Defectives.
4. In Subgroup sizes, enter 500.
5. Click NP Chart Options.
6. On the Tests tab, select 1 point > K standard deviations from
center line (Test 1) and K points in a row on same side of center
line (Test 2).
If you are not sure which tests apply in your specific situation, use
Tests 1 and 2 when you first establish the control limits based on your
data.
7. Click OK in each dialog box.
Interpret the results
One point is out of control on the NP chart. The engineer concludes that the
process is not stable and should be improved.
NP Chart of Defectives
Test Results for NP Chart of Defectives
TEST 1. One point more than 3.00 standard deviations from center
line.
Test Failed at points: 16
* WARNING * If graph is updated with new data, the results above
may no longer
be correct.