How do I compare the Cp/Pp and Cpk/Ppk?

Assume the original target is a Cpk or Ppk of 1.0. Capability Indices: Cpk

If Cpk or Ppk is greater than

If Cpk or Ppk is less than 1.0
1.0
If Cp or Pp is less than 1.0 Variation in the process should be Not mathematically possible.
reduced. Check for an error in calculations.
If Cp or Pp is greater than The process should be centered Fine tune and improve the process
1.0 within its specifications. continuously. Increase the Cpk
target.
Capability Indices: Cp
The Cp index is used to summarize a system's ability to meet two-sided specification limits (upper and
lower). Like Cpk, it uses estimated sigma and, therefore, shows the system's potential to meet the
specifications. However, it ignores the process average and focuses on the spread. If the system is not
centered within the specifications, Cp alone may be misleading.
The higher the Cp value...
..the smaller the spread of the system’s output. Cp is a measure of spread only. A process with a narrow
spread (a high Cp) may not meet customer needs if it is not centered within the specifications.
If the system is centered on its target value...
Cp should be used in conjunction with Cpk to account for both spread and centering. Cp and Cpk will be
equal when the process is centered on its target value. If they are not equal, the smaller the difference
between these indices, the more centered the process is.
Capability Indices: Cpk
Cpk is a capability index that tells how well as system can meet specification limits. Cpk calculations use
estimated sigma and, therefore, shows the system's "potential" to meet specifications. Since it takes the
location of the process average into account, the process does not need to be centered on the target value
for this index to be useful.
If Cpk is 1.0...
...the system is producing 99.73% of its output within specifications. The larger the Cpk, the less variation
you will find between the process output and specifications.
If Cpk is between 0 and 1.0...
...not all process output meets specifications.
If the system is centered on its target value...
...Cpk should be used in conjunction with the Cp index. Cpk and Cp will be equal when the process is
centered on its target value. If they are not equal, the smaller the difference between these indices, the
more centered the process is.
Capability Indices: Cr
The Cr capability ratio is used to summarize the estimated spread of the system compared to the spread of
the specification limits (upper and lower). The lower the Cr value, the smaller the output spread. Cr does
not consider process centering.
When the Cr value is multiplied by 100, the result shows the percent of the specifications that are being
used by the variation in the process. Cr is calculated using an estimated sigma and is the reciprocal of Cp.
In other words, Cr = 1/Cp.
Capability Indices: Cpm
The Cpm index indicates how well the system can produce within specifications. Its calculation is similar
to Cp, except that sigma is calculated using the target value instead of the mean. The larger the Cpm, the
more likely the process will produce output that meets specifications and the target value.
Capability Indices: Pp
The Pp index is used to summarize a system's performance in meeting two-sided specification limits
(upper and lower). Like Ppk, it uses actual sigma (sigma of the individuals), and shows how the system is
actually running when compared to the specifications. However, it ignores the process average and
focuses on the spread. If the system is not centered within the specifications, Pp alone may be misleading.
The higher the Pp value...
...the smaller the spread of the system’s output. Pp is a measure of spread only. A process with a narrow
spread (a high Pp) may not meet customer needs if it is not centered within the specifications.
If the system is centered on its target value...
...Pp should be used in conjunction with Ppk to account for both spread and centering. Pp and Ppk will be
equal when the process is centered on its target value. If they are not equal, the smaller the difference
between these indices, the more centered the process is.
Capability Indices: Ppk
Ppk is an index of process performance which tells how well a system is meeting specifications. Ppk
calculations use actual sigma (sigma of the individuals), and shows how the system is actually running
when compared to the specifications. This index also takes into account how well the process is centered
within the specification limits.
If Ppk is 1.0...
...the system is producing 99.73% of its output within specifications. The larger the Ppk, the less the
variation between process output and specifications.
If Ppk is between 0 and 1.0...
...not all process output meets specifications.
If the system is centered on its target value...
...Ppk should be used in conjunction with the Pp index. If the system is centered on its target value, Ppk
and Pp will be equal. If they are not equal, the smaller the difference between these indices, the more
centered the process is.
Capability Indices: Pr
The Pr performance ratio is used to summarize the actual spread of the system compared to the spread of
the specification limits (upper and lower). The lower the Pr value, the smaller the output spread. Pr does
not consider process centering.
When the Pr value is multiplied by 100, the result shows the percent of the specifications that are being
used by the variation in the process. Pr is calculated using the actual sigma (sigma of the individuals) and
is the reciprocal of Pp. In other words, Pr = 1/Pp.
Capability Analysis: Indices
What are the measures of process capability? What are the measures of process
These indices show what the process is performance?
"capable" of doing. This is the six sigma range These indices show what a process is actually
of the inherent variation of a statistically stable doing. This is the six sigma range of the total
process. Sigma is estimated using the /d2 process variation, where sigma is usually
formula. determined by the sample standard deviation.
Cpk Ppk
Cp Pp
Cr Pr
Cpm
Capabilty Analysis: Can a process produce output within spec?
A process that is in control is not necessarily producing an output that meets customer or engineering
requirements. To find out if your process is capable of producing outputs that are in spec, you should
perform capability analysis.
Capability analysis is a set of calculations used to assess whether a system is statistically
able to meet a set of specifications or requirements. To complete the calculations, a set of
data is required, usually generated by a control chart; however, data can be collected
specifically for this purpose.
Specifications or requirements are the numerical values within which the system is expected to operate,
that is, the minimum and maximum acceptable values. Occasionally there is only one limit, a maximum
or minimum. Customers, engineers, or managers usually set specifications. Specifications are numerical
requirements, goals, aims, or standards. It is important to remember that specifications are not the same
as control limits. Control limits come from control charts and are based on the data. Specifications are the
numerical requirements of the system.
All methods of capability analysis require that the data is statistically stable, with no special causes of
variation present. To assess whether the data is statistically stable, a control chart should be completed. If
special causes exist, data from the system will be changing. If capability analysis is performed, it will
show approximately what happened in the past, but cannot be used to predict capability in the future. It
will provide only a snapshot of the process at best. If, however, a system is stable, capability analysis
shows not only the ability of the system in the past, but also, if the system remains stable, predicts the
future performance of the system.
Capability analysis is summarized in indices; these indices show a system’s ability to meet its numerical
requirements. They can be monitored and reported over time to show how a system is changing. The
main indices used are Cp and Cpk. The indices are easy to interpret; for example, a Cpk of more than one
indicates that the system is producing within the specifications or requirements. If the Cpk is less than
one, the system is producing data outside the specifications or requirements. This section contains
detailed explanations of various capability indices and their interpretation.
Capability analysis is an excellent tool to demonstrate the extent of an improvement made to a process. It
can summarize a great deal of information simply, showing the capability of a process, the extent of
improvement needed, and later the extent of the improvement achieved.
Capability indices help to change the focus from only meeting requirements to continuous improvement
of the process. Traditionally, the focus has been to reduce the proportion of product or service that does
not meet specifications, using measures such as percentage of nonconforming product. Capability indices
help to reduce the variation relative to the specifications or requirements, achieving increasingly higher
Cp and Cpk values.
Capability vs control :
A process is said to be in control or stable, if it is in statistical control. A process is in statistical control
when all special causes of variation have been removed and only common cause variation remains.
Control charts are used to determine whether a process is in statistical control or not. If there are no points
beyond the control limits, no trends up, down, above, or below the centerline, and no patterns, the process
is said to be in statistical control.
Capability is the ability of the process to produce output that meets specifications. A process is said to be
capable if nearly 100% of the output from the process is within the specifications. A process can be in
control, yet fail to meet specification requirements. In this situation, you would need to take steps to
improve or redesign the process.
Capability Analysis for normal data :
The calculations for capability analysis are based on the following assumptions:
The data is normally distributed. In other words, the shape shown by the histogram follows the "normal"
bell curve.
The system being studied is stable and no assignable causes for variation are present. A control chart of
the system has been be made to determine stability before a capability analysis is done.
The mean of the system being studied falls between the upper and lower specification limits defined for
the process.
If these assumptions are not met, the results of a capability analysis will be misleading.
Capability Analysis for non-normal data :
Since a capability study makes the assumption that the data being analyzed is normally distributed, what
can be done if the data is not normally distributed?
Usually if the data is not normally distributed, the process is not in control and a capability study is
premature. However, in some cases the non-normal process is due to a measure that legitimately has only
a single-sided specification. For example, if you are measuring flatness, the measurements can never be
smaller than 0. In these cases, you will need to use Pearson curve fitting. Pearson curve fitting is a
technique in which the distribution is compared to one of many theoretical distributions. If the data
matches closely enough, it will pass a chi-square test and the capability indices will be useful. As with
normally distributed data, if the data does not match one of the theoretical distributions, then the
capability indices may be misleading and should not be used.
Capability Analysis :
1.0 What is it?
Capability analysis is a set of calculations used to assess whether a system is statistically able to meet a
set of specifications or requirements. To complete the calculations, a set of data is required, usually
generated by a control chart; however, data can be collected specifically for this purpose.
Specifications or requirements are the numerical values within which the system is expected to operate,
that is, the minimum and maximum acceptable values. Occasionally there is only one limit, a maximum
or minimum. Customers, engineers, or managers usually set specifications. Specifications are numerical
requirements, goals, aims, or standards. It is important to remember that specifications are not the same as
control limits. Control limits come from control charts and are based on the data. Specifications are the
numerical requirements of the system.
All methods of capability analysis require that the data is statistically stable, with no special causes of
variation present. To assess whether the data is statistically stable, a control chart should be completed. If
special causes exist, data from the system will be changing. If capability analysis is performed, it will
show approximately what happened in the past, but cannot be used to predict capability in the future. It
will provide only a snapshot of the process at best. If, however, a system is stable, capability analysis
shows not only the ability of the system in the past, but also, if the system remains stable, predicts the
future performance of the system.
Capability analysis is summarized in indices; these indices show a system’s ability to meet its numerical
requirements. They can be monitored and reported over time to show how a system is changing. Various
capability indices are presented in this section; however, the main indices used are Cp and Cpk. The
indices are easy to interpret; for example, a Cpk of more than one indicates that the system is producing
within the specifications or requirements. If the Cpk is less than one, the system is producing data outside
the specifications or requirements. This section contains detailed explanations of various capability
indices and their interpretation.
Capability analysis is an excellent tool to demonstrate the extent of an improvement made to a process. It
can summarize a great deal of information simply, showing the capability of a process, the extent of
improvement needed, and later the extent of the improvement achieved.
Capability indices help to change the focus from only meeting requirements to continuous improvement
of the process. Traditionally, the focus has been to reduce the proportion of product or service that does
not meet specifications, using measures such as percentage of nonconforming product. Capability indices
help to reduce the variation relative to the specifications or requirements, achieving increasingly higher
Cp and Cpk values.
Before capability analysis is completed, a histogram and control chart need to be completed.
2.0 What does it look like?
A courier company has set up a team to look at the actual arrival time at customers’ locations to pick up
packages, in comparison to the scheduled arrival time. The company guarantees pick up of packages
within 14 minutes of the scheduled time. It is unacceptable to customers for the courier to arrive early.
Therefore, there are two requirements: on time and up to 14 minutes late. The result of the capability
analysis for this example follows.
 = 2.00
Zupper = 2.00 Zlower = 5.00
Cpk = 0.67 Cp = 1.17
Cpu = 0.67 Cpl = 1.67
3.0 Capability Analysis: When is it used?
Use the standard method for calculating capability analysis when you can answer "yes" to all of the
following questions:
Is it necessary to understand how the system performs in comparison to specification limits?
Specifications or requirements must be available to complete capability analysis. The system must also be
measured in the same way as the specifications, so a direct comparison can be made.
Does the specification consist of an upper and lower requirement?
For processes with one-sided specifications, see the article Capability analysis for one-sided
specifications.
Are no special causes of variation present?
A system with special causes is unstable and constantly changing. If capability analysis is performed
under these circumstances, it will be unreliable. Always construct a control chart and check for special
causes before completing capability analysis.
Is the data in variables form?
In order to complete the standard method for capability analysis, the data must be in variables form, that
is measured data, such as time, length, weight, or distance.
Do the individual values form a normal distribution?
In order to complete capability analysis using the standard method, a normal distribution is required. Use
a histogram to check for normal distribution. If the distribution is not normal, non-normal capability
analysis can be used.
Has the data been collected over a period of time?
There are two ways to collect data for capability analysis. The standard method is from a control chart,
where the data is collected over a period of time. If data has been collected in this way, the standard
method for performing capability analysis is used.
Note that if you cannot answer these questions with a "yes," Practical Tools for Continuous Improvement
includes capability analysis for one-sided specifications, trial runs, attributes data, and nonnormal data.
4.0 How is it made?
These steps assume that variables data has been collected over time, and that a control chart and
histogram have been completed. The control chart should show no special causes, and the histogram
should reveal that the data is normally distributed.
In the example, the team examined the arrival time of couriers in comparison to the scheduled arrival time
over a month. Since time is being measured, the data is variables. The team completed a histogram and
found the data to be normal. An
X- R control chart was also completed, showing no special causes of variation.
Since a control chart must be completed before performing capability analysis, the calculations from the
control chart can be utilized. The information taken from the control chart follows:
The sample size used in the control chart, n
The overall average, (from an chart) or (from an X-MR chart.)
The average range,(from an chart) or (from an X-MR chart.)
The numerical specifications or requirements should also be known. Information for the example is
shown below:
n=5

= 10.00

= 4.653

USL = upper specification limit = 14 minutes

LSL = lower specification limit = 0 minutes (on time)

Steps:
4.1 Sketch the distribution
4.2 Calculate the estimated standard deviation.
4.3 Determine the location of the tails for the distribution
4.4 Draw the specification limits on the distribution
4.5 Calculate how much data is outside the specifications
4.6 Calculate and interpret the capability indices
4.7 Analyze the Results
1 Sketch the Distribution
Sketch a picture of a normal distribution. Begin by drawing a horizontal line (axis). Next, draw a normal
(bell-shaped) curve centered on the horizontal axis. Then draw a vertical line from the horizontal axis
through the center of the curve, cutting it in half. This line represents the overall average of the data and is
always located in the center of a normal distribution. Label the line with the value for the overall average
and its symbol. The value of the overall average in the example is 10.00 and the symbol for the overall
average from the chart is . The example completed through this step follows.

4.2. Calculate the estimated standard deviation.

The next stage is to calculate the position of the tails of the distribution that has just been drawn.
However, in order to calculate the position of the tails, the standard deviation is required. In this version
of capability analysis where data has been collected over a period of time, an estimated standard deviation
is used. The symbol for the estimated standard deviation is (read “sigma hat”). The formula for the
estimated standard deviation is:

is calculated when constructing a control chart. Substitute M for if an X-MR chart has been
completed. In the example, is 4.653. The denominator (d2) is a weighting factor whose value is based on
the subgroup size, n, from the control chart. The value for d2 in the example, based on a subgroup size of
5, is 2.326. A short listing of the d2 values for other subgroup sizes follows. The full table of values is
given in the appendix.
n 2 3 4 5 6 7 8 9 10
 d2 1.128 1.693 2.059 2.326 2.534 2.704 2.847 2.970 3.078
The estimated standard deviation for the example is:

The estimated standard deviation is calculated to one more decimal place than the original data.
4.3. Determine the location of the tails for the distribution
The next step is to determine where (at what value) the tails or ends of the curve are located. These values
can be estimated by adding and subtracting three times the estimated standard deviation from the overall
average. Remember, from the histogram section, that for a normal distribution, plus or minus three times
the standard deviation from the overall average includes 99.73 percent of the area under the curve. The
calculation for the location of the left tail is:

For the example the left tail is:

The right tail is calculated as follows:

For the example the right tail is:

Add the values to the distribution drawn earlier. The example completed through this step follows.

4.4.4 Draw the specification limits on the distribution

Draw vertical lines on the distribution to represent the lower and upper specification limits. In the
example, the lower specification limit (LSL) is 0 minutes (on time) and the upper specification limit
(USL) is 14 minutes. Estimate where the two lines should be located in reference to the overall average
and the tails of the curve. Label each specification with its abbreviation and value. The example
completed through this step follows.

The diagram shows whether any portion of the curve is beyond the specifications. In the example, some
of the distribution is beyond the upper specification. If the overall average of the distribution is outside
the specification, refer to “Variation – Capability analysis where the overall average is outside the
specification” later in this section.
4.5. Calculate how much data is outside the specifications
As indicated in the previous step, some of the distribution is outside the specification limit. The question
is, how much? To determine the percentage that falls outside the specification limits, it is necessary to
find how many estimated standard deviations exist between the overall average and each specification
limit. The number of standard deviations is known as the Z value. Z values are used to determine the
percentage of output that is outside the specification limits using the Standard normal distribution table.
4.5.1. Find the percentage above the upper specification.
The first step in determining the percentage above the upper specification is to calculate the Z value for
the upper specification. This is found by subtracting the overall average from the upper specification, and
then dividing by the estimated standard deviation. The Z value for the upper specification is denoted as
Zupper. The upper specification for the example is 14, the overall average is 10.00, and the estimated
standard deviation is 2.00. Thus, the value of Zupper for the example is:

This means that the upper specification is located 2.00 estimated standard deviations away from the
overall average. Look up the Z value in the Standard Normal Distribution Table to find the estimated
proportion of output that is outside the upper specification.
Z values are listed along the left and top of the table. The whole number (number to the left of the
decimal) and the tenths digit (first number to the right of the decimal) are listed on the left hand side of
the table, and the hundredths digit (second number to the right of the decimal) is along the top. The table
shows Z values only up to 4. If the Z value is greater than 4, the proportion outside the specification is
virtually 0. In the example, the Z value is 2.00. To find the percentage outside the specification, go down
the left hand side of the table to 2.0 and then across to the column marked x.x0. The number is 0.0228,
which is the proportion outside the specification. To convert the proportion to a percentage, multiply it by
100. The percentage outside the upper specification is 2.28 percent. Place this percentage on the diagram.
4.5.2. Find the percentage below the lower specification.
The Z value for the lower specification is found by subtracting the lower specification from the overall
average, and then dividing by the estimated standard deviation. The Z value for the lower specification is
denoted as Zlower. The lower specification for the example is 0, the overall average is 10.00, and the
estimated standard deviation is 2.00. Thus, the value for Zlower for the example is:

This means that the lower specification is located 5.00 estimated standard deviations away from the
overall average. Look up the Z value in the Standard normal distribution table as previously described.
Since the table shows Z values up to only 4, the proportion and percentage outside of the specification is
taken as 0. If any of the data is outside the specification, add the percentage to the diagram.
4.5.3. Determining the total percentage outside the specifications
The total percentage outside the specification limits or requirements is found by adding the percentage
outside the upper and lower specification limits. The total percent of output located outside the
specification limits for the example is:
2.28 + 0 = 2.28%
4.6. Calculate and interpret the capability indices
This step describes the key capability indices.
4.6.1. Calculate Cp.
4.6.2. Calculate Cpk.
4.6.3. Interpret Cp and Cpk.
4.6.4. Calculate Cpu and Cpl.
4.6.1. Calculate Cp.
Cp is an index used to assess the width of the process spread in comparison to the width of the
specification. It is calculated by dividing the allowable spread by the actual spread. The allowable spread
is the difference between the upper and lower specification limits. The actual spread is 6 times the
estimated standard deviation. Plus or minus 3 times the estimated standard deviation contains 99.73
percent of the data and is commonly used to describe actual spread.
Cp for the example is:

A Cp of one indicates that the width of the process and the width of the specification are the same. A Cp
of less than one indicates that the process spread is greater than the specification. This means that some of
the data lies outside the specification. A Cp of greater than one indicates that the process spread is less
than the width of the specification. Potentially this means that the process can fit inside the specification
limits. The following diagrams show this graphically.

In fact, the Cp states how many times the process can fit inside the specification. So a Cp of 1.5 means
the process can fit inside the specification 1.5 times. A Cp greater than one is obviously desirable.
However, the example has a Cp greater than one and yet it still has data outside the specification. This is
due to the position of the overall average relative to the specification. When the overall average is away
from the center of the specification, the system can still produce data outside the specification even
though the Cp is greater than one, as in the example below:
To overcome this problem, Cpk was created.
4.6.2. Calculate Cpk
Cpk takes into account the center of the data relative to the specifications, as well as the variation in the
process. Cpk is simple to calculate. The smaller of the two Z values is selected. This is known as Zmin .
When Zmin has been selected, it is divided by 3. The formula is:

The Z values for the example are Zupper of 2.00 and Zlower of 5.00, therefore Zmin is 2.00. Cpk for the
example is:

If the Cpk formula is written in full, it becomes more apparent how Cpk works.

This is the smaller of:

Graphically this can be drawn for the example as follows:

The diagram clearly shows that the overall average is too close to the upper specification. By taking the
smaller of the two Z values, Cpk is always looking at the worst side, where the specification is closest to
the overall average. Since it is looking only at half the picture, instead of dividing by 6 as in Cp, it is
divided by 3 .

A Cpk value of one indicates that the tail of the distribution and the specification are an equal distance
from the overall average, as shown below:
A Cpk of less than one, as in the example, means that some of the data is beyond the specification limit. A
Cpk greater than one indicates that the data is within the specification. The larger the Cpk, the more
central and within specification the data.
4.6.3. Interpret Cp and Cpk.
The Cp and Cpk indices are the primary capability indices. Cp shows whether the distribution can
potentially fit inside the specification, while Cpk shows whether the overall average is centrally located.
If the overall average is in the center of the specification, the Cp and Cpk values will be the same. If the
Cp and Cpk values are different, the overall average is not centrally located. The larger the difference in
the values, the more offset the overall average. This concept is shown graphically below.

Cpk can never exceed Cp, so Cp can be seen as the potential Cpk if the overall average is centrally set. In
the example, Cp is 1.17 and Cpk is 0.67. This shows that the distribution can potentially fit within the
specification. However, the overall average is currently off center. The Cpk value does not state whether
the overall average is offset on the upper or lower side. It is necessary to go to the Z values to discern this.
An alternative is to show the capability indices Cpu and Cpl.
4.6.4. Calculate Cpu and Cpl
Cpu and Cpl are the Cpk values calculated for both Z values.
Therefore, Cpu is:

For the example:

Cpl is:

For the example:

From Cpu and Cpl, it is evident that the smaller value for the example is Cpu, which is the same value as
Cpk. By comparing Cpu to Cpl, it is evident that the overall average is off center and closer to the upper
specification than the lower specification. The larger the difference between the Cpu and Cpl, the more
off center the process.
4.7. Analyze the Results
The completed analysis for the example is shown below.

Calculations:
= 2.00
Zupper = 2.00 Zlower = 5.00
Cpk = 0.67 Cp = 1.17
Cpu = 0.67 Cpl = 1.67
Examine the capability indices and the distribution. What do they show? Is the process capable? In the
example, the process is off center, reflecting a capability issue. The upper specification of 14 minutes
cannot be achieved consistently. The team must either improve the process or revise the specification. In
the example, the team chose to revise the specification, but this is not an option in many industries.
The aim of capability is to achieve improved Cpk values, resulting in a more capable system. For most
industries, the aim is to achieve a Cpk of at least one. Certainly this is the case for most service
organizations. Some manufacturing companies require Cpk values greater than one. For example, the
minimum is frequently 1.33, providing room for process drift, etc. When parts are being assembled,
reduced variation at the center of the specification gives considerable benefit, namely parts assemble
more quickly and more easily. Motorola, for example, constantly strives for higher and higher Cpk values.
The company’s 6-sigma program has received a great deal of attention, and translates to a Cpk of 2.0. By
pushing for improved Cpk values, the improvement effort is focused on shrinking variation around the
center of the specifications.
Caution.
If a process is unstable—that is, if special causes are evident in the control chart—capability analysis will
be unreliable. Every time the capability indices are calculated, they will be different. Special causes
should be removed from a process. While special causes are present, the process is unpredictable, causing
it to go in and out of specification. As soon as a special cause occurs, the Cpk is meaningless, since
special causes often result in unpredictable defects; even an apparently good Cpk can not be relied upon.
If the sampling selected in the control chart is not appropriate for the process, this can also affect the
Cp/Cpk values. For example, sampling too frequently will artificially reduce the range values and cause
the Cp/Cpk values to appear high. Using the wrong sample size can have a similar effect. Refer to the
sampling section for guidance with appropriate sample size and frequency.
If the data being analyzed is not normal, the estimated standard deviation () will not be accurate.
Nonnormal capability analysis must be used if the distribution is not normal, refer to the topic
"Nonnormal capability analysis," later in this section for more information. The method shown in this
section can be significantly affected by nonnormal data, giving inaccurate results.
Standard Normal Distribution Table
z x.x0 x.x1 x.x2 x.x3 x.x4 x.x5 x.x6 x.x7 x.x8 x.x9
4.0 .00003
3.9 .00005 .00005 .00004 .00004 .00004 .00004 .00004 .00004 .00003 .00003
3.8 .00007 .00007 .00007 .00006 .00006 .00006 .00006 .00005 .00005 .00005
3.7 .00011 .00010 .00010 .00010 .00009 .00009 .00008 .00008 .00008 .00008
3.6 .00016 .00015 .00015 .00014 .00014 .00013 .00013 .00012 .00012 .00011
3.5 .00023 .00022 .00022 .00021 .00020 .00019 .00019 .00018 .00017 .00017

3.4 .00034 .00032 .00031 .00030 .00029 .00028 .00027 .00026 .00025 .00024
3.3 .00048 .00047 .00045 .00043 .00042 .00040 .00039 .00038 .00036 .00035
3.2 .00069 .00066 .00064 .00062 .00060 .00058 .00056 .00054 .00052 .00050
3.1 .00097 .00094 .00090 .00087 .00084 .00082 .00079 .00076 .00074 .00071
3.0 .00135 .00131 .00126 .00122 .00118 .00114 .00111 .00107 .00104 .00100

2.9 .0019 .0018 .0018 .0017 .0016 .0016 .0015 .0015 .0014 .0014
2.8 .0026 .0025 .0024 .0023 .0023 .0022 .0021 .0021 .0020 .0019
2.7 .0035 .0034 .0033 .0032 .0031 .0030 .0029 .0028 .0027 .0026
2.6 .0047 .0045 .0044 .0043 .0041 .0040 .0039 .0038 .0037 .0036
2.5 .0062 .0060 .0059 .0057 .0055 .0054 .0052 .0051 .0049 .0048

2.4 .0082 .0080 .0078 .0075 .0073 .0071 .0069 .0068 .0066 .0064
2.3 .0107 .0104 .0102 .0099 .0096 .0094 .0091 .0089 .0087 .0084
2.2 .0139 .0136 .0132 .0129 .0125 .0122 .0119 .0116 .0113 .0110
2.1 .0179 .0174 .0170 .0166 .0162 .0158 .0154 .0150 .0146 .0143
2.0 .0228 .0222 .0217 .0212 .0207 .0202 .0197 .0192 .0188 .0183

z x.x0 x.x1 x.x2 x.x3 x.x4 x.x5 x.x6 x.x7 x.x8 x.x9
1.9 .0287 .0281 .0274 .0268 .0262 .0256 .0250 .0244 .0239 .0233
1.8 .0359 .0351 .0344 .0336 .0329 .0322 .0314 .0307 .0301 .0294
1.7 .0446 .0436 .0427 .0418 .0409 .0401 .0392 .0384 .0375 .0367
1.6 .0548 .0537 .0526 .0516 .0505 .0495 .0485 .0475 .0465 .0455
1.5 .0668 .0655 .0643 .0630 .0618 .0606 .0594 .0582 .0571 .0559

1.4 .0808 .0793 .0778 .0764 .0749 .0735 .0721 .0708 .0694 .0681
1.3 .0968 .0951 .0934 .0918 .0901 .0885 .0869 .0853 .0838 .0823
1.2 .1151 .1131 .1112 .1093 .1075 .1056 .1038 .1020 .1003 .0985
1.1 .1357 .1335 .1314 .1292 .1271 .1251 .1230 .1210 .1190 .1170
1.0 .1587 .1562 .1539 .1515 .1492 .1469 .1446 .1423 .1401 .1379
 0.9 .1841 .1814 .1788 .1762 .1736 .1711 .1685 .1660 .1635 .1611
0.8 .2119 .2090 .2061 .2033 .2005 .1977 .1949 .1922 .1894 .1867
0.7 .2420 .2389 .2358 .2327 .2297 .2266 .2236 .2206 .2177 .2148
0.6 .2743 .2709 .2676 .2643 .2611 .2578 .2546 .2514 .2483 .2451
0.5 .3085 .3050 .3015 .2981 .2946 .2912 .2877 .2843 .2810 .2776

0.4 .3446 .3409 .3372 .3336 .3300 .3264 .3228 .3192 .3156 .3121
0.3 .3281 .3783 .3745 .3707 .3669 .3632 .3594 .3557 .3520 .3483
0.2 .4207 .4168 .4129 .4090 .4052 .4013 .3974 .3936 .3897 .3859
0.1 .4602 .4562 .4522 .4483 .4443 .4404 .4364 .4325 .4286 .4247
0.0 .5000 .4960 .4920 .4880 .4840 .4801 .4761 .4721 .4681 .4641
Cpk or Ppk: Which should you use?
Your customer has asked you to report the Given that Ppk uses the calculated sigma, it is
Cpk of the product you are sending. You no longer necessary to use the calculated
know that to compute the Cpk, you need to sigma in Cpk. The only acceptable formula
have the product specifications, and that you for Cpk uses the estimated sigma.
need to have the mean and sigma. As you
gather the information, someone asks,
"which sigma do they want?"
You know that Cpk is calculated by dividing
by 3 sigma. But which sigma should you use,
estimated or calculated? Which is correct?
Which would you report? Naturally, most of
us would use the sigma that makes the Cpk
look the best. But the sigma that makes the
Cpk look best may not accurately reflect
what you or your customer need to know Estimated sigma:
about the process.
Confusion over calculating Cpk by two
different methods is one reason that a new
Given that Ppk uses the calculated sigma, it is
index, Ppk, was developed. Ppk uses the
no longer necessary to use the calculated
calculated sigma from the individual data.
sigma in Cpk. The only acceptable formula
for Cpk uses the estimated sigma.
In 1991, the ASQC/AIAG Task Force
published the "Fundamental Statistical
Process Control" reference manual, which
shows the calculations for Cpk as well as
Ppk. These should be used to eliminate
confusion about calculating Cpk.
So which value is best to report, Cpk or Ppk?
Although they show similar information, they
have slightly different uses.
Sigma of the individuals: Estimated sigma and the related capability
indices (Cp, Cpk, and Cr) are used to
measure the potential capability of a system
to meet customer needs. Use it when you
want to analyze a system's aptitude to
perform.
Actual or calculated sigma (sigma of the
individuals) and the related indices (pp, Ppk,
 and Pr) are used to measure the performance
of a system to meet customer needs. Use it
when you want to measure a system's actual
process performance.
How can Cpk be good with data outside the specifications?
A customer who called our technical support line recently could not understand why his Cpk was above
1.0 when his data was not centered between the specifications and some of the data was outside the
specification. How can you have a good Cpk when you have data outside the specification and/or data
which is not centered on the target/nominal value?
To calculate Cpk, you need to know only three pieces of information: the process average, the variation in
the process, and the specification(s). First, find out if the mean (average) is closest to the upper or lower
specification. If the process is centered, then either Zupper or Zlower can be used, as you will see below.
If you only have one specification, then the mean will be closest to that specification since the other one
does not exist.
To measure the variation in the process, use the estimated sigma (standard deviation). If you decide to use
the standard deviation from the individual data, you should use the Ppk calculation, since Ppk uses this
sigma. To calculate the estimated sigma, divide the average range, R-bar, by d2. The d2 value to use
depends on the subgroup size and will come from a table of constants shown below. If your subgroup size
is one, you will use the average moving range, MR-bar.
d2 values
Subgroup size d2
1 1.128
2 1.128
3 1.693
4 2.059
5 2.326
6 2.534
7 2.704
You, of course, provide the specifications. Now that you have these 3 pieces of information, the Cpk can
be easily calculated. For example, let’s say your process average is closer to the upper specification. Then
Cpk is calculated by the following:
Cpk = (USL - Mean) /( 3*Est. sigma). As you can see, the data is not directly used. The data is only
indirectly used. It is used to determine mean and average range, but the raw data is not used in the Cpk
calculation. Here is an example that might serve to clarify. Suppose you have the following example of 14
subgroups with a subgroup size of 2:
Averag
Sample No. Range
e
1 0.03 0.06 0.045 0.030
2 0.10 0.20 0.150 0.100
3 0.05 0.10 0.075 0.050
4 1.00 0.00 0.500 1.000
5 1.50 1.50 1.500 0.000
6 1.10 1.50 1.300 0.400
7 1.10 1.00 1.050 0.100
8 1.10 1.01 1.055 0.090
9 1.25 1.20 1.225 0.050
10 1.00 0.30 0.650 0.700
 11 0.75 0.76 0.755 0.010
12 0.75 0.50 0.625 0.250
13 1.00 1.10 1.050 0.100
14 1.20 1.40 1.300 0.200

Average 0.8057 0.220

The mean, X-bar, is 0.8057 and the average range, R-bar, is 0.220. For this example, the upper
specification is 2.12, the target value is 1.12, and the lower specification is 0.12. In the data shown above,
more than 21% of the data is outside the specification, so you would expect Cpk to be low, right? As it
turns out, Cpk is relatively healthy at 1.17. (Yes, for this example, we have ignored the first cardinal rule:
Before one looks at Cpk, the process must be in control.)
Before we go on, let’s check the math.
Mean = 0.8057
Average range = 0.2200
Est. sigma = R-bar / d2
= 0.2200/1.128 =0.1950
Cpk = smallest of (Zupper and Zlower) / 3
Zlower = (Mean - LSL) / Est. sigma
= (.8057 - 0.12) / .1950
= .6857 / .1950
Zlower = 3.516
Zupper is larger, so in this example,
Cpk = Zlower / 3
= 3.516 / 3
Cpk = 1.172
So what gives? Here is an example where Cpk is good, yet the process is not centered and data is outside
of at least one of the specifications. The reason Cpk is good is because the average range is understated
and thus when you divide by the estimated sigma (which uses the average range), it over-inflates Cpk.
The reason the average range is understated will be discussed in a future article. One last note, if you look
at this data on a control chart, you will quickly see that it is not in control. Therefore, the Cpk statistic
should be ignored when the process is not in control.
How do we determine process capability if the process isn't normal?
Cp, Cpk, Pp, and Ppk do not necessarily rely on the normal distribution. If any of these indices increases,
you know that the process capability has improved. What you do not know is how that improvement
translates into good product. This requires knowledge of the distribution of the individual units produced
by the process. The Central Limit Theorem refers to averages and this works for the control chart, but it
doesn't work for the histogram. Therefore, we generally make the assumption of the normal distribution in
order to estimate the percent out of specification (above, below, and total).
For non-normal distributions, we first estimate some parameters using the data. We then use these
parameters and follow a Pearson curve fitting procedure to select an appropriate distribution. Since the
relationship between the standard deviation and the percent within CAN vary differently from the normal
distribution for distributions that are not normal, e.g., plus and minus one sigma may not equal 68.26%,
plus and minus 2 sigma may not equal 95.44%, etc., we try to transform the capability indices into
something comparable. With this distribution equation, we integrate in from the tails to the upper and
lower specifications respectively. Once the percent out-of-spec above and below the respective spec limits
are estimated, the z values (for the normal with the same mean and standard deviation) associated with
those same percents are determined. Then, Cpk and Ppk are calculated using their respective estimates for
the standard deviation. This makes these values more comparable to those that people are used to seeing.
Is Cpk the best capability index?
Cpk has been a popular capability index for many years and perhaps because of its momentum it
continues to remain popular. But is it the best index to use?
Answering this question assumes that there is one best index, which is a different discussion altogether.
Let’s agree that there several other useful capability indices. Two other indices that can be beneficial are
Ppk and Cpm.
As mentioned in a previous article on Cpk, "Cpk or Ppk: Which should you use," Cpk uses only the
estimated sigma to measure variation. While this is acceptable, the estimated sigma can be artificially low
depending on the subgroup size, sample interval, or sampling plan. This in turn can lead to an over-
inflated Cpk
For a process that drifts, such as the process shown in the chart below, the estimated sigma will usually be
artifcially low. This is because the estimated sigma looks at only variation within subgroups.

Ppk, on the other hand, uses the standard deviation from all of the data. We can call this the sigma of the
individual values or sigmai. Sigma of the individual values looks at variation within and between
subgroups.
For a process that exhibits drifting, estimated sigma would not pick up the total variation in the process
and thus the Cpk becomes a cloudy statistic. In other words, one can not be sure it is a valid statistic.
In contrast to Cpk, Ppk, which uses the sigma of the individual values, would pick up all the variation in
the process. Again, sigmai uses between and within subgroup variation. So if there is drifting in the
process, sigmai would typically be larger than the estimated sigma, sigmae, and thus Ppk would, as it
should, be lower than Cpk.
Here is a quick review of the formulae for Cpk and Ppk:
Cpk = Zmin/3 where Zmin Ppk = Zmin/3 where Zmin

Zmin = (USL – Mean) / est.sigma Zmin = (USL – Mean) / sigmai

or = (Mean - LSL) / est. sigma = (Mean - LSL) / sigmai

We should be concerned with how well the process is behaving, therefore Ppk might be preferred over
Cpk. Ppk is a more conservative approach to answering the question, "How good is my process?"
Watch for a future article discussing the relatively new capability index, Cpm, and how it stacks up
against Cpk and Ppk.
Should you calculate Cpk when your process is not in control?
The AIAG Statistical Process Control reference manual (p. 13) states:
"The process must first be brought into statistical control by detecting and acting upon special causes of
variation. Then its performance is predictable, and its capability to meet customer expectations can be
assessed. This is the basis for continual improvement."
True, but to take it one step further, if the process is not in a state of statistical control then the validity of
a Cpk value is questionable.
Suppose your customer requires you to provide a Cpk value and does not require control charts. Or
perhaps the customer is willing to accept lack of control as long as the Cpk is acceptable. You provide a
"good" Cpk number and relax, knowing that your customer is satisfied. But have you really satisfied your
customer’s need, which is to ensure that your product or service is within an acceptable specification
region and consistent over time?
It is certainly possible to calculate Cpk even when a process is not in control, but one might ask what
value this calculation provides. Rather than state "You should never calculate Cpk when the process is out
of control," I prefer to say that the less predictable your process is, the less meaningful Cpk is or the less
value Cpk carries. While it is easy to say that one should never calculate Cpk when the process is out-of-
control, it is not always practical, since customers may dictate otherwise.
One of the reasons that minimal emphasis should be placed on Cpk when the process is not in control is
predictability. Customers want good Cpk values as well as some confidence that in the future, Cpk will be
consistent or improved over previous capability studies. [This topic will be addressed in a future article.]
Another reason that you should not put too much weight on Cpk when the process is not in control is due
to the underlying statistics that are used in calculating Cpk. Since Cpk uses the range, a process can
appear "better" simply because the range is not a fair representation of the process variability when the
process is not in control or predictable. If the process is in control, one could conclude that the range is
sufficient for calculating Cpk.

Chart produced using SQCpack

A hypothetical example might clarify the point:

Suppose I have 100 pieces of data that are grouped into units of 5 each. The chart above shows how the
control chart and histogram of the data might look. In this example, the process is in control, and the Cpk
= 1.252.
Suppose I am evaluating another process whose mean and specifications are the same, but the Cpk =
1.803. Most of us would want to have the second process with the higher Cpk, but is the process
necessarily better? Unless you determine if the process is in statistical control, you can not fairly answer
this question.
As it turns out, the data is exactly the same, but what has changed is the order in which the data was
grouped in the samples. This caused the range of the subgroups and R-bar (the average range) to be
different. In the second data set, the data was rearranged so that the data within the sample is similar. The
sigma of the individuals does not change, but the estimated sigma, which is used in the control limits and
Cpk calculations, changes between the two distributions.
With this example, determining if the process is in control before looking at Cpk pays off. Since the
control chart in the second example, shown below, is not in control, you cannot be sure that its Cpk is a
good representation of process capability. The first process, on the other hand, is in control and thus its
Cpk is a good predictor of process capability.
Chart produced using SQCpack

If you do not have the control chart to evaluate for process control, you might be tempted to select the
second process as being "better" on the basis of the higher Cpk. As this example illustrates, you cannot
fairly evaluate Cpk without first establishing process control.
The capability index dilemma: Cpk, Ppk, or Cpm
Lori, one of our customers, phoned the other day to ask if Cpk is the best statistic to use in a process that
slits metal to exacting widths. I too wondered what index would be best suited for her application.
Perhaps Cpk, Ppk, Cpm, or some other index offers the best means of reporting the capability of her
product or process.
Lori’s process capability has never dipped below 2 and averages above 3. Given this high degree of
capability, she might consider reducing variation about the target. While the Cpk and Ppk are well
accepted and commonly-used indices, they may not provide as much information as Lori needs to
continue to improve the process.
Cpm incorporates the target when calculating the
standard deviation. Like the sigma of the
individuals formula, compares each
observation to a reference value. However, instead
of comparing the data to the mean, the data is
compared to the target. These differences are
squared. Thus any observation which is different
from the target observation will increase the
standard deviation.
As this difference increases, so does the sigma. And as this sigma becomes larger, the Cpm index gets
smaller. If the difference between the data and the target is small, so too is the sigma. And as this sigma
gets smaller, the Cpm index becomes larger. The higher the Cpm index, the better the process, as shown
in the diagrams below.
In these 3 charts the This Cpm is
process is the same, good.
but as the process
becomes more
centered, the Cpm
gets better.
 This Cpm is
better.

This Cpm is best.

In these 3 charts, the

process stays
centered about the This Cpm is
target, but as the good.
variation is reduced,
the Cpm gets better.

This Cpm is
better.

This Cpm is best.

We can use Lori’s raw data to provide an example of how Cpm is calculated:
Sample Sample Sample Sample Sample Sample Sample Sample Sample
1 2 3 4 5 6 7 8 9
obs 1 90.741 102.711 104.066 106.602 100.904 104.922 112.738 102.388 97.825
obs 2 102.300 100.882 105.620 95.978 108.558 100.243 108.145 104.159 95.209
obs 3 98.642 103.314 96.165 96.265 94.882 97.053 98.679 100.204 91.273
obs 4 106.069 98.569 100.412 95.869 98.573 111.042 103.788 99.328 93.430
obs 5 97.635 96.639 96.316 84.872 108.588 99.068 105.664 94.157 98.263
And the specifications are: USL = 145, Target = 105, LSL = 60

Cpm = (145 – 60) / (6 * )

Cpm = 1.91
In a process with both upper and lower specifications, the target is typically the midpoint of these. When
such a high degree of capability exists, you may want to ask your customer if the target value is ideal. I
recommended to Lori that she check with her customer to determine if he or she wants a small shift
toward one of the specifications. Regardless of the target in relation to the specifications, the focus should
always be on making the product to target with minimum variation. Cpm is the capability index that most
accurately depicts this.
Calculating capability indices with one specification
The following formula for Cpk is easily found in most statistics books.
Cpk = Zmin / 3
Zmin = smaller of Zupper or Zlower
Zupper = [(USL – Mean) / Estimated sigma*]
Zlower = [(Mean – LSL) / Estimated sigma*]
Estimated sigma = average range / d2
And, we’ve all learned that generally speaking, the higher the Cpk, the better the product or process that
you are measuring. That is, as the process improves, Cpk climbs.
What is not apparent, however, is how to calculate Cpk when you have only one specification or
tolerance. For example, how do you calculate Cpk when you have an upper tolerance and no lower
tolerance?
When faced with a missing specification, you could consider:
Not calculating Cpk since you don’t have all of the variables.
Entering in an arbitrary specification.
Ignoring the missing specification and calculating Cpk on the only Z value.
An example may help to illustrate the outcome of each option. Let’s assume you are making plastic
pellets and your customer has specified that the pellets should have a low amount of moisture. The lower
the moisture content, the better. No more than .5 is allowed. If the product has too much moisture, it will
cause manufacturing problems. The process is in statistical control.
It is not likely your customer would be happy if you went with option A and decided not to calculate a
Cpk.
Going with option B, you might argue that the lower specification limit (LSL) is 0 since it is impossible
to have a moisture level below 0. So with USL at .5 and LSL at 0, Cpk is calculated as follows:
If USL = .5, X-bar = .0025, and estimated sigma = .15, then:
Zupper = [(.5 - .0025) / .15] = 3.316,
Zlower = [(.0025 – 0) / .15] = .01667 and
Zmin = .01667
Cpk = .01667 /3 = .005
Your customer will probably not be happy with a Cpk of .005 and this number is not representative of the
process.
Example C assumes that the lower specification is missing. Since you do not have a LSL, Zlower is
missing or non-existent. Zmin therefore becomes Zupper and Cpk is Zupper/3.
Zupper = 3.316 (from above)
Cpk = 3.316 / 3 = 1.10.
A Cpk of 1.10 is more realistic than .005 for the data given in this example and is representative of the
process.
As this example illustrates, setting the lower specification equal to 0 results in a lower Cpk. In fact, as the
process improves (moisture content decreases) the Cpk will decrease. When the process improves, Cpk
should increase. Therefore, when you only have one specification, you should enter only that
specification, and treat the other specification as missing.
An interesting debate (well, about as interesting as statistics gets) occurs with what to do with Cp (or Pp).
Most textbooks show Cp as the difference between both specifications (USL – LSL) divided by 6 sigma.
Because only one specification exists, some suggest that Cp can not be calculated. Another suggestion is
to look at ~ ½ of the Cp. For example, instead of evaluating [(USL – Mean) + (Mean – LSL)] / 6*sigma,
instead think of Cp as (USL – Mean) / 3*sigma or (Mean – LSL) / 3*sigma. You might note that when
you only have one specification, this becomes the same formula as Cpk.
Frequently-Asked Questions about Capability
Are Cpk & Ppk acronyms? If so, what do they actually mean or represent?
Since their introduction, there has been a lot of speculation as to meaning. Here is my two cents
worth.
Cp has always been know as capability of the process since I became aware of it and it has been around
for some time. My connection with Cpk came through the Ford "Continuous Process Control and Process
Capability Improvement Manual" probably more than 20 years ago. In the Ford manual, a k value was
used to represent the number of standard deviations between the Target and the . I would assume that the
Cpk came literally from Cp with a k factor adjustment. In reference to the Pp and Ppk, the reference from
the beginning has been to Process Performance as opposed to Process Capability.

What is the difference between Cp and Pp?

The technical difference is that the 6 sigma used for the Cp calculation (or the 3 sigma used for the Cpk
calculation) comes from the estimate of sigma based on the average range, and the 6 sigma used for Pp
calculation (or 3 sigma used for the Ppk calculation) comes from the estimate of sigma based on using all
the data and the classical formula for the standard deviation. The formulas for Cp and Cpk are here;
formulas for Pp and Ppk are here.
In general, if the process is in control and normally distributed (standard assumptions when doing
capability analysis), both values should be close. However, since most processes wander around a little bit
(and are in control), an intuitive interpretation is that the Cpk is what you could be doing and Ppk is what
you are doing.

What is the difference in the formulas for Cpk and Ppk?

The only difference I see is the i and r after the sigma symbol. What are these referring to?
The six sigma used for the Cpk calculation comes from the estimate of sigma based on the average range
(r). The six sigma used for the Ppk calculation comes from the estimate of sigma based on using all the
Individual data (i) and the classical formula.
In general, if the process is in control and normally distributed (standard assumptions when doing
capability analysis), both values should be close. However, since most processes wander around a little bit
(and are still in control), an intuitive interpretation is that the Cpk is what you could be doing and Ppk is
what you are doing.

Are there maximum values for Cp, Cpk, Pp and Ppk?

No. As long as the spec range does not change and you continually reduce the variation, you will increase
these indices. I have seen as high as 36 and have heard of higher.

How can I improve Cpk value, when it is less than 1.0?

First, compare Cpk to Cp. If Cpk is less than Cp and Cp is greater than one, center the process in the
specification. This should make Cpk comparable to Cp. If Cp and Cpk are less than one, there are two
actions you can take. The first (an unadvisable one) is to widen the specification particularly on the side
that has the spec limit closest to the center of the process ( ). The second and more advisable answer is to
improve the process by reducing variation in the process. If the process is off-center, it would be
advisable to try to center it as you try to improve it.

Is it possible to have a Ppk value of 10 and a Pp number of 5?

This should not occur. You might have a negative number for the Ppk that is larger in absolute value then
the Pp number. This implies that the process mean lies outside the specification limits.

What do the letters in Cp and Cpk stand for?

There is no authoritative answer. Cp has been around for a long time and many believe it stands for
Capability of the Process. Others say Process Capability, but that would reverse the letters.
As for Cpk, in the literature that I first saw about Cpk, k was the amount of the difference in the target
value and in standard deviations (the number of standard deviations that the process is off target). Before
you ask, Pp generally is said to be Process Performance.

Why do capability indices formulas divide by 3?

When calculating Cp you divide the specification range by six sigma. This is plus and minus three sigma
on each side of the mean of the process which would include about 99.7% of the distribution of output if
the process is normal. Cp considers only the spread
and not the centering of the process. Consequently,
you can have a capable process (Cp > 1) and not be making any good product. Cpk considers the mean of
the process and calculates two values ([Cp-usl = (USL - )/3] and [Cp-lsl = ( - LSL)/3]). Since the
What is an ideal Cpm value?
Generally there is no "ideal." Bigger is always better. The difference in Cpm as defined in SQCpack is in
the calculation of the stand deviation or variance term. The standard deviation for Cpm is based on the
target value rather than the which will make
sigma(pm) larger and Cpm smaller. You could say "ideally" the process should be centered in the
specification making Cpm = Cp. However, Cp might only be 0.80 which surely is not "ideal."
Can I compare two processes based on only the Cpk values of each of them? Is there any other tool
by which I can say that one process is better than other?
It depends on what you mean by better. If the processes are producing the same product dimension, then
you can compare them more or less directly.
Cpk includes a centering factor as well as the
variation factor. Unless you want to compare
Can the process performance index Ppk be applied on the ongoing process? If yes, how?
The capability indices are designed to be applied to on going processes. They are an indication of what a
customer can expect in terms of quality from a particular process.
If you have a control chart on a characteristic for a process, SQCpack or CHARTrunner will calculate
these values for you if you enter the specifications.
If you do not have either of these programs, the Capability Analysis article series provides information on
Why would I have Cp and Cpk indices well over 1 when some readings are outside the specification
limits?
My first guess would be that if you look at a control chart of the data, it is out of control. Before you can
do capability analysis, the process should be predictable and that requires that it be stable (in-control). For
a more detailed discussion, see How can Cpk be good with data outside the specification?

Glossary
Assignable cause : An assignable cause is a source of variation that is intermittent, not predictable. It is
sometimes called "special cause" variation. On a control chart, an assignable cause is signaled by points
beyond the control limits or nonrandom patterns within the control limits.
Attributes data : Attributes data is data that can be classified and counted. There are two types of
attributes data: counts of defects per item or group of items (nonconformities) and counts of defective
items (nonconforming). For example, yes/no, good/bad, pass/fail, and go/no go.
Average : Another term for a mean, it is an indicator of the center of a set of data points. It is found by
adding all the individual values and dividing by the number of values.
Bell curve : Another term for the shape formed by a normal distribution when drawn as a histogram.
Bias : Something that influences the selection of certain items when collecting a sample.
Bimodal distribution : A distribution that has two modes. Drawn as a histogram, this condition is
reflected by two peaks or high points.
Capability : The capability of a process is how the process performs when compared to specification
limits or requirements. It uses a series of indices: Cp, Cpk, Cr, and Cpm
Capability Analysis : A set of statistical calculations performed on a set of data to assess how the
distribution formed by the data compares to specifications or requirements.
Capable process : A process is said to be capable if nearly 100% of its output falls within specification
limits
c-chart : An attributes control chart that is used to monitor the number of nonconformities, such as
defects per subgroup. The subgroup size must remain constant for this type of chart.
Central location : Central location is the center of a set of data points. Mean, median, and mode are the
statistics used to describe it.
Central tendency : Statistics such as the mean, median, and mode are said to be measures of central
tendency
Characteristic : A distinguishing feature of a process or its output on which variables or attributes data
can be collected.
Chi-square : A goodness-of-fit-test statistic used to test the assumption that the distribution of a set of
data is similar to the expected distribution, such as a normal distribution.
Coefficient of variance : A ratio that measures the significance of the standard deviation in relation to the
mean.
Common cause : A source of variation that is inherent in a system and is predictable. A control chart
identifies a system with only common causes of variation. Common causes of variation affect all
individual values of a system, and can be eliminated only by a systemic change.
Control chart : A control chart is a graphical representation of a characteristic of a process, showing
plotted values of some statistic, a central line, and one or two control limits. It is used to determine
whether a process has been operating in statistical control and is an aid to maintaining statistical control.
Control limits : Lines on a control chart used as a basis for judging whether variation in data on a chart is
due to special or common causes. These limits are calculated from data collected from the system, they
are not specifications or limits set by customers or management.
Cp : A capability index that compares the width of a two-sided specification with the variation in the
process. Estimated standard deviation is used to calculate the process variation. A Cp larger than 1
indicates that the process variation is narrower than the specification.
Cpk : Cpk is a capability index that tells how well a system can meet two-sided specification limits.
Because it takes the target value into account, the system does not have to be centered on the target value
for this index to be useful. It is calculated with estimated standard deviation. A Cpk greater than 1
indicates that the process can meet the specification.
Cpl : A capability index that compares the variation in the process to the lower specification. Estimated
standard deviation is used to calculate the process variation. A Cpl greater than 1 indicates the process is
capable of meeting the lower specification.
Cpm : Cpm is a capability index that shows how well the system can produce output within
specifications while taking the target into account. Its calculation uses sigma calculated from the target
value instead of the mean.
Cpu : A capability index that compares the variation in the process to the upper specification. Estimated
standard deviation is used to calculate the process variation. A Cpu greater than 1 indicates the process is
capable of meeting the upper specification.
Cr : Capability ratio compares the variation in a process with the width of a two-sided specification.
Estimated standard deviation is used to calculate the process variation. It is the inverse of Cp.
Defect : An occurrence such as a blemish, scratch, burn, error, or omission that appears on an object. A
defect does not necessarily make the object unusable or unacceptable.
Defective : A product or service flawed beyond use or acceptability.
Discrimination : This refers to a description of the capability of a measurement system.
Dispersion : Statistics such as the range and standard deviation (sigma) are said to be measures of
dispersion.
Distribution : Distribution is a way of describing the output from a system of variation. The
distribution’s location, shape, and spread may be evaluated by statistics such as the mean, median, sigma,
and range.
Estimated sigma : This is an estimate of the standard deviation calculated by dividing the average range
by the tabular constant d2 (R-bar/d2).
Histogram : A histogram is a bar chart that represents the frequency distribution of data. The height of
each bar corresponds to the number of items in the class or cell. The width of each bar represents a
measurement interval. The histogram shows basic information such as central location, shape, and spread
of the data being examined.
In control : A process is said to be "in control" or "stable" if it is in statistical control. If a process is in
statistical control, a control chart will have no subgroups falling outside the control limits, no runs, and no
nonrandom patterns.
Individuals control chart : The individual portion of an X-MR control chart. The individual data points
are plotted onto the chart and compared with control limits.
Kurtosis : Kurtosis is a statistic that is used to measure the "flatness" or "peakedness" of a set a of data. It
represents a measure of the combined weight of the tails relative to the rest of a distribution. As the tails
of a distribution become heavier, the kurtosis will increase. As the tails become lighter, the kurtosis value
will decrease.
Lower control limit : A line on a control chart used as a basis for judging whether variation from the data
on the chart is due to special or common causes. Any point beyond the lower control limit is an indication
of a special cause occurring. This limit is calculated from data collected on the system, it is not a
specification or limit set by customers or management. The symbol is LCL.
Lower specification limit : The lower limit of a specification. This limit is set as an aim for a system or
process, it is usually set by the customer of the process, engineering, or management. The symbol for the
lower specification is LSL – lower specification limit.
Maximum acceptable subgroup size : When a varying sample size is being used in a p or u-control
chart, the maximum acceptable sample size is usually a sample size that is twenty-five percent larger than
the average sample size. Any subgroup with a sample size larger than the maximum acceptable subgroup
size has to have control limits calculated specifically for that subgroup.
Mean : Another term for average, it is an indicator of the central location of a set of data. It is found by
adding all the individual values and dividing by the number of values.
Measurement System : A measurement system consists of the people, procedures, systems, and devices
used to take measurements.
Median : The middle number in a set of data when it is ranked from lowest to highest, it is an indicator of
central location in a data set.
Minimum acceptable subgroup size : When a varying sample size is being used in a p or u-control
chart, the minimum acceptable sample size is usually a sample size that is twenty-five percent smaller
than the average sample size. Any subgroup with a sample size smaller than the minimum acceptable
subgroup size has to have control limits calculated specifically for that subgroup.
Mode : Is the number that occurs most frequently in a data set. It is usually an indicator of central
location.
Moving range : The difference between consecutive subgroup values on an X-MR control chart. The
moving range is used as a measure of variability.
Moving range chart : The moving range portion of an individuals and moving range control chart. The
moving ranges are plotted on the chart and compared with control limits.
Negatively skewed distribution : A distribution of data where most of the data appears on the right hand
side of the distribution and then tails off to the left. Also known as a skewed left distribution.
Nonconforming : Nonconforming data is a count of defective units. It is often described as go/no go,
pass/fail, or yes/no, since there are only two possible outcomes to any given check. You can track either
the number of defective units or the number of nondefective units.
Nonconformities : Nonconformities data is a count of defects per unit or group of units. It can refer to
defects or occurrences that should not be present but are, or any characteristic that should be present but is
not.
Nonnormal data distribution : Any data set that does not show a normal, bell-shaped distribution.
Nonnormal data : Data that does not form a normal distribution.
Nonrandom pattern : A pattern in data that is repeating, or is not due to normal variation.
Normal curve : This bell-shaped curve is used to illustrate the shape of a normal distribution.
Normal distribution : A data distribution that is bell shaped and symmetrical, the normal distribution is
the basis for control chart and capability analysis.
Normal probability plot : A normal probability plot is a graphical method for showing a frequency
distribution. The scaling is set up so that if the distribution is normal, a straight line will result.
np-chart : An attributes control chart that plots the number of items that are defective or possess a
characteristic of interest. The subgroup size must remain constant for this type of chart to be used.
Observation : An observation is a single piece of data, usually a count or a measurement. It is also known
as a reading.
Operational definition
When applied to data collection, it is a clear, concise, and detailed definition of a measure. It ensures that
those collecting data do so consistently.
Outlier : An outlier is a point on a chart that does not fall into the pattern of the rest of the data.
Out-of-control : When applied to a control chart, out of control means that at least one special cause of
variation is present.
Overcontrol : Over reaction to a set of data. For example, in a control chart, it would be reacting to a
common cause as if it were a special cause.
Pareto chart : A Pareto chart is a bar chart for ranking aspects of a problem. Typically, a few aspects
make up a significant portion of the problem while many trivial aspects exist.
p-chart : An attributes control chart that plots the number of items possessing a characteristic of interest.
The subgroup size may vary.
Positively skewed distribution : A distribution of data where most of the data appears on the left hand
side of the distribution and then tails off to the right. Also known as a skewed right distribution.
Pp : Pp is a capability index, similar to Cp, that is a measure of process performance. Pp tells how well a
system can meet two-sided specification limits, assuming that the average is centered on the target value.
It is calculated with the actual sigma (using the actual individual values) rather than the estimated sigma.
A Pp larger than 1 indicates that the process variation is narrower than the specification.
Ppk : Similar to Cpk, Ppk is a capability index that indicates whether a process is capable of meeting
two-sided specification limits. However, Ppk uses actual standard deviation to calculate the process
variation, whereas Cpk uses an estimated standard deviation. The target value is taken into account with
Ppk, so the system does not have be center on the target value to be useful. A Ppk greater than 1 indicates
that the process can meet the specification.
Ppl : A capability index similar to Cpl in that it compares the variation in the process to the lower
specification. However, Ppl uses standard deviation to calculate the process variation, whereas Cpl uses
an estimated standard deviation. A Ppl greater than 1 indicates the process is capable of meeting the lower
specification.
Ppu : A capability index similar to Cpu in that it compares the variation in the process to the upper
specification. However, Ppu uses standard deviation to calculate the process variation, whereas Cpu uses
an estimated standard deviation. A Ppu greater than 1 indicates the process is capable of meeting the
upper specification.
Pr : A capability ratio similar to Cr in that it compares the variation in a process with the width of a two-
sided specification. However, Pr uses standard deviation to calculate the process variation, whereas Cr
uses an estimated standard deviation. It is the inverse of Pp.
Process : A process is the combination of people, equipment, materials, methods, and environment that
produce output—a given product or service. The words process and system are often used
interchangeably.
Process capability : Process capability is the 6 sigma range of common cause variation for statistically
stable processes only. Sigma is usually estimated by R-bar/d2.
Process performance : The process performance is the 6 sigma range of inherent variation for
statistically stable processes only, where sigma is usually estimated by the sample standard deviation
Random distribution : A distribution that forms no particular shape.
Random sample : A sample that allows every item in a population to have an equal chance of being
selected, with no bias.
Range : Range is an estimate of spread in a set of data points; the difference between the highest and
lowest values in the data set.
Repeatability : Repeatability refers to variation in a series of measurements that have been taken with
one gage measuring one characteristic of the same item by the same person.
Reproducibility : Reproducibility refers to variation in a series of measurements that have been taken
with one gage measuring one characteristic of the same item by different people.
Run chart : A run chart is a simple line chart that plots one characteristic over time. It is used to plot
individual observations and detect patterns in the data.
Sample : A sample is a collection of one or more observations used to analyze the performance of a
process, as opposed to the total populations. It is intended to represent the characteristics of the
population. Sample is a synonym for "subgroup" in process control applications
Sample size : The number of pieces of data taken at one time. For example five boxes are checked for
stiffness every hour, the sample size in this case is five. If the temperature of a room is taken every hour,
only one number is collected every hour, so the sample size is one.
Sigma of the individuals : Sigma of the individuals is standard deviation calculated from the individual
data values in a data set. It is also known as actual or calculated sigma.
Sigma : Sigma is the Greek symbol, , used to denote standard deviation. It is a measure of the variation
or spread within a set of data.
Skewed distribution : A distribution that tails off to one side, either to the left or right.
Skewness : Skewness is a statistic that is used to measure the symmetry of the distribution for a set of
data. A process that is skewed tails off to the left or to the right.
Special cause : Special cause variation is a source of variation that is intermittent, not predictable.
Sometimes it is called "assignable cause" variation. On a control chart, a special cause is signaled by
points beyond the control limits, runs, or nonrandom patterns within the control limits. A process that has
special cause variation is said to be out-of-control, unstable, or unpredictable.
Specification limits : Specifications are boundaries, usually set by management, engineering, or
customers, within which a system must operate. They are sometimes called engineering tolerances
Spread : Spread is the range of data from the lowest value to the highest value.
Stable process : A system, analysed by a control chart, with no special causes of variation present, this
system is also said to be in control. Variation within a stable system is due to common causes, and is
predictable.
Standard deviation : A statistic that describes the variation or spread within a data set. It can be used to
indicate the variation in a process and to compare with specifications.
Statistical control : Statistical control is a condition describing a process from which all special causes of
variation have been removed and only common causes of variation remain. On a control chart, processes
that are in statistical control show no subgroups outside the control limits, no runs, and no nonrandom
patterns. This condition is also referred to as in control, stable, or predictable.
Subgroup : A subgroup is one or more occurrences or measurements taken at one time. Multiple
subgroups are used to analyze the performance of a process. Subgroup is used as a synonym for "sample."
Symmetrical distribution : A distribution that if cut in half, shows each side is the mirror of the other.
Target value : The exact value at which customers, engineering, or management want the system to
operate.
Trial limits : On a control chart, trial limits are calculated when there is insufficient data to calculate
control limits. These give a temporary guide until sufficient data has been collected
u-chart : An attributes control chart that is used to monitor the number of nonconformities per unit, such
as defects per item. The subgroup size may vary.
Undercontrol : Not reacting to a set of data when the data is showing an issue or problem. For example,
in a control chart, it would be ignoring a special cause of variation.
Uniform distribution : A distribution, when drawn as a histogram, has each bar at a similar frequency.
Unstable system : A system that contains special and common causes of variation; this system is also
said to be out of control. An unstable system is unpredictable.
Upper control limit : A line on a control chart used as a basis for judging whether variation from the data
on the chart is due to special or common causes. Any point beyond the upper control limit is an indication
of a special cause occurring. This limit is calculated from data collected on the system, it is not a
specification or limit set by customers or management. Its symbol is UCL.
Upper specification limit : The upper limit of a specification. This limit is set as an aim for a system or
process, it is usually set by the customer of the process, engineering, or management. The symbol for the
upper specification is USL–upper specification limit.
Variability : Variability refers to the differences among individual outputs of a process. In control chart
pairs, it refers to the differences between individual observations and is analyzed in range, sigma, and
moving range charts.
Variables : Variables data is data that is acquired through measurements, such as length, time, diameter,
strength, weight, temperature, density, thickness, pressure, and height. X-bar and range, X-bar and sigma,
and individuals and moving range charts are used to analyze variables data.
Variation : Variation is the inevitable differences that occur among individual outputs of a process.
Sources of variation may be grouped into two major categories: common causes and special causes.
X-bar : X-bar is the average or mean of values in a group of observations.
X-bar chart : The X-bar chart is a variables control chart that shows the subgroup averages. The
subgroup size for this chart must be larger than one and consistent.
Zlower : The symbol for the Z value for the lower specification limit. It represents the number of standard
deviations between the average and the lower specification limit.
Zmin : The minimum of the Z values, either Zupper or Zlower. It is used to calculate the Cpk index in capability
analysis
Zupper : The symbol for the Z value for the upper specification limit. It represents the number of standard
deviations between the average and the upper specification limit.
Z value : Used in capability analysis, it is the symbol for the number of standard deviations between the
average and a specification limit for a normal distribution.