NCSS Statistical Software NCSS.

com

Chapter 255

Capability Analysis
Introduction
This procedure summarizes the performance of a process based on user-specified specification limits. The
observed performance as well as the performance relative to the Normal distribution are output. Process capability
ratios such as Cp and Cpk are produced. Cpm and Cpkm may also be generated if a specification target is entered. A
capability histogram with specification limit lines may also be produced in this procedure. Normality Tests are
also given in this procedure. Subgroup data or individual values may be used.

Capability Analysis
Capability analysis, or process capability analysis, is the comparison of the distribution of sample values to the
specification limits, and possibly also the specification target. One basic measure of the capability of the process
is the proportion of values falling inside (or outside) the specification limits. Another measure of capability is the
proportion of values that would fall inside (or outside) the specification limits if the data are assumed to follow
the normal distribution. Several capability ratios, or capability indices, have been developed to summarize how
well the process yields measurements within the specification limits. Those produced in this procedure are Cp,
Cpk, Cpm, and Cpkm. Cpm and Cpkm additionally take into account the nearness of the process to the specification
target.
Process data are typically gathered as samples or individual measurements taken from the process at given times
(hours, shifts, days, weeks, months, etc.). If more than one value is taken at a time, the measurements of the
samples at a given time constitute a subgroup.

255-1
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Typically, an initial series of subgroups or individual values is used to estimate the mean and standard deviation
of a process. The mean and standard deviation can then be used to estimate the capability of the process.
Because the assumption of normality is integral to the usefulness of the summaries, an important part of capability
analysis is determining whether the data follow a Normal distribution, at least approximately. Normality tests and
the capability histogram can be useful for investigating this assumption.

Other Procedures for Process Capability

Some of the other procedures in NCSS that may be useful for analyzing process capability are X-bar and R (or s)
charts, IM-R Charts, Descriptive Statistics, Stem-and-Leaf Plots (for smaller samples), Normality Tests, Outlier
Tests, Distribution Fitting, Box-Cox Transformation, and the Data Simulation Tool.

Process Capability Formulas

The formulas for estimating the mean and sigma depend on whether the data is subgroup data or individual value
data.

Estimating the Mean – Subgroup Data

Suppose we have k subgroups, each of size n. Let xij represent the measurement in the jth sample of the ith
subgroup.
The ith subgroup mean is calculated using
n

∑x
j =1
ij

xi = ,
n
The formula for the overall mean is
k ni

∑∑ x
i =1 j =1
ij

x= k
.
∑n
i =1
i

If the subgroups are of equal size, the above equation for the grand mean reduces to
k

∑x i
x1 + x 2 +  + x k
x= i =1
= .
k k

Estimating the Mean – Individual Values Data

Suppose we have k individual values. The estimate of the overall mean is given by
k

∑x i
x= i =1
.
k

255-2
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Estimating Sigma – Subgroup Data

In this procedure, sigma can be entered directly, or there are three options for estimating sigma from subgroup
data: sample ranges, sample standard deviations, and the weighted approach. Suppose we have k subgroups, each
of size n. Let xij represent the measurement in the jth sample of the ith subgroup.

Estimating Sigma – Subgroup Data – Sample Ranges

If the standard deviation (sigma) is to be estimated from the ranges, Ri, it is estimated as
R
σˆ =
d2
where
k

∑R
i =1
i
R =
k
E (R ) µR
d2 = =
σ σ
The calculation of E(R) requires the knowledge of the underlying distribution of the xij’s. Making the assumption
that the xij’s follow the normal distribution with constant mean and variance, the values for d2 are derived through
the use of numerical integration. It is important to note that the normality assumption is used and that the accuracy
of this estimate requires that this assumption be valid.
In the procedure, this calculation is performed if Sigma Estimation is set to From Data – R-bar or s-bar Estimate,
and Range or SD Estimation is set to Range.

Estimating Sigma – Subgroup Data – Sample Standard Deviations

If the standard deviation (sigma) is to be estimated from the standard deviations, it is estimated as
s
σˆ =
c4
where
k

∑s
i =1
i
s =
k
E (s ) µs
c4 = =
σ σ
The calculation of E(s) requires the knowledge of the underlying distribution of the xij’s. Making the assumption
that the xij’s follow the normal distribution with constant mean and variance, the values for c4 are obtained from
n
Γ 
2  2
c4 =
n − 1  n − 1
Γ 
 2 
In the procedure, this calculation is performed if Sigma Estimation is set to From Data – R-bar or s-bar Estimate,
and Range or SD Estimation is set to SD.

255-3
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Estimating Sigma – Subgroup Data – Weighted (SD) Approach

When the sample size is variable across subgroups, a weighted approach is recommended for estimating sigma
(Montgomery, 2013):
1/ 2
 k 2
 ∑ ( ni − 1) si 
σˆ =  i =1 k 
 ni − k 
 ∑ i =1

In the procedure, this calculation is performed if Sigma Estimation is set to From Data – SD Approach.

Estimating Sigma – Individual Values Data

Suppose we have k individual values. There are two methods in this procedure for estimating sigma: moving
ranges and overall sample standard deviation.

Estimating Sigma – Individual Values Data – Moving Ranges

If there is only one observation per time point, a moving range may be calculated by finding the range of each
value with its previous value:
Ri = xi − xi −1
Then the standard deviation (sigma) is estimated from the ranges, Ri, in the same manner as for subgroup data,
namely,
R
σˆ =
d2
where
k

∑R
i =1
i
R =
k
E (R ) µR
d2 = =
σ σ
In the procedure, this calculation is performed if the data are individual values data, and Sigma Estimation is set
to From Data – R-bar or s-bar Estimate, and Range or SD Estimation is set to Range.

Estimating Sigma – Individual Values Data – Overall Standard Deviation

If there is only one observation per time point, and the process is assumed to be in control, sigma may be
estimated using the sample standard deviation
k

∑(x i − x )2
σˆ = i =1
k −1
In the procedure, this calculation is performed if the data are individual values data, and if Sigma Estimation is set
to From Data – SD Approach.

255-4
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Process Capability Ratios

Several capability ratio formulas are presented below. Further details may be found in Montgomery (2013) and
Ryan (2011).

Cp
The process capability ratio Cp is given by
USL − LSL
Cp =
6σ
where USL and LSL are the upper and lower specification limits, respectively. An estimate of Cp is produced by
substituting a suitable estimate of σ , namely σ̂ .
Confidence intervals for Cp are given as

χ n2−1,α / 2
C p lower = C p
n −1
χ n2−1,1−α / 2
C p upper = C p
n −1

Cpl and Cpu

The one-sided capability ratios Cpl and Cpu are defined as
µ − LSL
C pl =
3σ
and
ULS − µ
C pu =
3σ
These are estimated by substituting mean and standard deviation estimates.

Cpk
Cpk is the lesser of Cpl and Cpu, or
C pk = min(C pl , C pu )
The lower and upper confidence limits for Cpk reported in NCSS are given by

n −1  C pk 2 
C pk lower = C pk − z1−α / 2 + 1 + 6 
9n( n − 3)  2n − 6  n − 1

n −1  C pk 2 
C pk upper = C pk + z1−α / 2 + 1 + 6 
9n( n − 3)  2n − 6 
 n − 1

255-5
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Cpm
A capability ratio which incorporates the nearness to the specification target is defined as
USL − LSL
C pm =
6 σ 2 + (µ − T )
2

where T refers to the specification target. A suitable estimate of Cpm is made by substituting estimates of the mean
and standard deviation.

Cpmk
Similarly to Cpm, Cpmk also accounts for nearness to the specification target:
C pk
C pmk =
µ −T 
2

1+  
 σ 
where T refers to the specification target. A suitable estimate of Cpmk is made by substituting estimates of the
mean and standard deviation.

Data Structure
In this procedure, the data may be in any of three formats. The first data structure option is to have the data in
several columns, with one subgroup per row.

Example dataset
S1 S2 S3 S4 S5
2 6 3 8 5
8 8 7 7 9
6 2 2 4 3
5 6 7 6 10
48 2 6 5 0
. . . . .
. . . . .
. . . . .

255-6
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

The second data structure option uses one column for the response data, and either a subgroup size or a second
column defining the subgroups.

Alternative example dataset

Response Subgroup
2 1
6 1
3 1
8 1
5 1
8 2
8 2
7 2
7 2
9 2
6 3
2 3
. .
. .
. .

In the alternative example dataset, the Subgroup column is not needed if every subgroup is of size 5 and the user
specifies 5 as the subgroup size. If there are missing values, the Subgroup column should be used, or the structure
of the first example dataset.
If there are no subgroups (individual values only), the only input needed is a single column of values.

Response
2
6
3
8
5
8
8
7
7
9
6
2
.
.
.

255-7
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Procedure Options
This section describes the options available in this procedure. To find out more about using a procedure, go to the
Procedures chapter.

Variables Tab
This panel specifies the variables that will be used in the analysis.
Input Type
Specify whether the data is in a single response column or in multiple columns with one subgroup per row.
Specify whether the data is in a single response column with an associated subgroup column, multiple columns
with one subgroup per row, or a column with individual values (no subgroups).

• Response Column and Subgroup Column or Subgroup Size

Response Subgroup
5 1
6 1
4 1
3 2
7 2
6 2
4 3
5 3
8 3
. .
. .
. .

• Multiple Columns with One Subgroup Per Row

X1 X2 X3
5 6 4
3 7 6
4 5 8
. . .
. . .
. . .

255-8
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

• Response Column with Individual Values (no subgroups)

Response
3
8
4
5
4
6
7
5
6
.
.
.

Variables – Response Column

Response Variable
Specify the column with the data values.
Subgroup Specification
Specify whether subgroups are defined by a Subgroup ID variable, or by a subgroup size. If the subgroup size is
3, then subgroups are formed by going down the response column in groups of 3. The first subgroup would be 5.
Subgroup ID Variable
Specify the column containing the subgroup identifiers.

Response ID Variable
5 1
6 1
4 1
3 2
7 2
6 2
4 3
5 3
8 3
. .
. .
. .

A new subgroup is created for each change in the Subgroup ID Variable, going down.

255-9
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Subgroup Size
Specify the number of individuals in each subgroup.

Response
5
6
4
3
7
6
4
5
8
.
.
.

If the subgroup size is 3, then subgroups are formed by going down the response column in groups of 3. The first
subgroup would be 5, 6, 4; the second would be 3, 7, 6; and so on.

Variables – Multiple Columns

Data Variables
Specify the columns containing the sample responses. Each row represents a subgroup.

X1 X2 X3
5 6 4
3 7 6
4 5 8
. . .
. . .
. . .

Specify Rows
Specification Method
Select which method will be used to specify the rows of the data.

• All Rows
All rows in the response column(s) will be used.

• Enter First Row and Last Row

Specify the first row and the last row of the data.

• First N Rows (Enter N)

The data beginning at Row 1 and ending at Row N will be used.

• Last N Rows (Enter N)

The last N rows of the dataset will be used.

255-10
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

• Keep Rows Variable

Specify a variable and a value in that variable column that will be used to determine which rows are used.

• Remove Rows Variable

Specify a variable and a value in that variable column that will be used to determine which rows will not be
used.
First Row
Specify the beginning row to be used.
Last Row
Specify the last row to be used.
N
Enter the number of rows to be used.
Keep Rows Variable
This variable (column) is used to specify which rows of the data will be used.
Keep Rows Value
This value determines which rows of the Keep Rows Variable will be used.
Remove Rows Variable
This variable (column) is used to specify which rows of the data will not be used.
Remove Rows Value
This value determines which rows of the Remove Rows Variable will not be used.

Limits & Estimation Tab

The options on the Limits & Estimation tab are enter the specification limits, and to specify the method by which
sigma is estimated.

Specification Limits
Lower Limit
Enter the lower specification limit. This lower specification limit is required for calculations of Cp, Cpl, Cpm, and
may be used in calculations of Cpk and Cpkm. This lower limit can also be displayed on the capability histogram.
Upper Limit
Enter the upper specification limit. This upper specification limit is required for calculations of Cp, Cpl, Cpm, and
may be used in calculations of Cpk and Cpkm. This lower limit can also be displayed on the capability histogram.
Target Specification Value
Enter the target value. The target value is required for calculation of Cpm and Cpkm. This target value can be
displayed on the capability histogram.

Mean Value Options

Mean Value Estimation
Specify whether the mean will be estimated from the data, or whether it will be specified directly.

255-11
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

• From Data
Estimate sigma based on the mean range. Only the subgroups specified for use in calculations will be used.

• Enter Mean Value

Specify the mean value directly.

• Use a Variable with Mean Value

Specify a column containing the mean value in row 1.
Mean Value
Enter the value to be used for the mean in all calculations.
Mean Value Variable
Specify a column containing the mean value in row 1. This value will be used for the mean in all calculations.

Sigma Estimation Options

Sigma Estimation
Specify the method by which Sigma will be estimated for the capability analysis calculations.

• From Data – R-bar or s-bar Estimate

Estimate sigma based on the average of the ranges or standard deviations (whichever is specified under Range
or SD Estimation). When there are no subgroups (individual values only), sigma will be estimated with the
ranges of each value with the previous value, as s-bar cannot be estimated with only one value at each time
point.

• From Data – SD Approach

When there are subgroups, this method estimates sigma using a weighted approach estimate formula that is
recommended when the subgroup size varies across subgroups. When there are no subgroups (individual
values only), sigma is estimated using the common sample standard deviation formula.

• Enter Sigma Value

In this case the sigma value is entered directly. This sigma value is used in all calculations involving sigma.

• Use a Variable with Sigma Value

Specify a column containing the sigma value in row 1. This sigma value is used in all calculations involving
sigma.
Sigma Value
Enter the value to be used for sigma. This sigma value is used in all calculations involving sigma
Sigma Variable
Specify a column containing the sigma value in row 1. This sigma value is used in all calculations involving
sigma.

255-12
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Reports Tab
The following options control the format of the reports.

Specify Reports
Estimation Summary Section
This report gives the estimated mean and sigma to be used in the remaining calculations.
Capability Analysis Section
This report gives a performance summary as well as several process capability ratios.
Normality Tests Section
This report gives three normality tests.

Report Options
Precision
Specify the precision of numbers in the report. A single-precision number will show seven-place accuracy, while
a double-precision number will show thirteen-place accuracy. Note that the reports are formatted for single
precision. If you select double precision, some numbers may run into others. Also note that all calculations are
performed in double precision regardless of which option you select here. This is for reporting purposes only.
Variable Names
This option lets you select whether to display variable names, variable labels, or both.
Page Title
This option specifies a title to appear at the top of each page.
Plot Subtitle
This option specifies a subtitle to appear at the top of each plot.

Histogram Tab
This panel sets the options used to define the appearance of the histogram.

Select Plots
Capability Histogram
Each chart is controlled by three form objects:
1. A checkbox to indicate whether the chart is displayed.
2. A format button used to call up the plot format window (see Histogram Format Options below for
more formatting details).
3. A second checkbox used to indicate whether the chart can be edited during the run.
Specification Limits on Histogram
Check this box to include the specification limit lines on the histogram.

255-13
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Histogram Format Options

To learn details regarding the format of the histogram, go to the Histograms chapter of the documentation.
Common options, such as axes, labels, legends, and titles are documented in the Graphics Components chapter.

Storage Tab
The options on this panel control the automatic storage of the means and ranges on the current dataset.

Storage Columns
Store Means in Column
You can automatically store the means of each subgroup into the column specified here.
Warning: Any data already in this column is replaced. Be careful not to specify columns that contain important
data.
Store Ranges/SDs in Column
You can automatically store the range or standard deviation of each subgroup into the column specified here.
Warning: Any data already in this column is replaced. Be careful not to specify columns that contain important
data.

255-14
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Example 1 – Capability Analysis for Subgroup Data

This section presents an example of how to run a capability analysis for subgroups. The data represent 50
subgroups of size 5 that are assumed to be in control. The specification limits for the process are 50 and 80, with a
specification target of 65. The data used are in the QC dataset. We will analyze the variables D1 through D5 of
this dataset.
You may follow along here by making the appropriate entries or load the completed template Example 1 by
clicking on Open Example Template from the File menu of the Capability Analysis window.

1 Open the QC dataset.

• From the File menu of the NCSS Data window, select Open Example Data.
• Click on the file QC.NCSS.
• Click Open.

2 Open the Capability Analysis window.

• Using the Analysis menu or the Procedure Navigator, find and select the Capability Analysis procedure.
• On the menus, select File, then New Template. This will fill the procedure with the default template.

3 Specify the variables.

• On the Capability Analysis window, select the Variables tab.
• Double-click in the Data Variables text box. This will bring up the variable selection window.
• Select D1 through D5 from the list of variables and then click Ok. “D1-D5” will appear in the Data
Variables box.

4 Set the Specification Limits.

• On the Capability Analysis window, select the Limits & Estimation tab.
• Enter 50 for Lower Limit. Enter 80 for Upper Limit.
• Enter 65 for Target Specification Value.

5 Run the procedure.

• From the Run menu, select Run Procedure. Alternatively, just click the green Run button.

Mean Estimation Section

Mean Estimation Section for Subgroups 1 to 50
Number of Subgroups 50

Estimation Type Estimate

Estimated Grand Average 67.12

This section displays the estimated mean to be used in all calculations.

Estimated Grand Average
This value is the average of all the observations. If all the subgroups are of the same size, it is also the average of
all the X-bars.

255-15
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Sigma Estimation Section

Sigma Estimation Section for Subgroups 1 to 50

Estimation Estimated Estimated

Type Value Sigma
Ranges (R-bar)* 18.14 7.798796
Standard Deviations (s-bar) 7.365443 7.835698
Weighted Approach (s-bar) 7.902911 7.902911

* Indicates the estimation type used in this report.

This report gives the estimation of the population standard deviation (sigma) based on three estimation
techniques. The estimation technique used for the calculations in this procedure is based on the ranges.
Estimation Type
Each of the formulas for estimating sigma is shown earlier in this chapter in the Process Capability Formulas
section.
Estimated Value
This column gives the R-bar and s-bar estimates based on the corresponding formulas.
Estimated Sigma
This column gives estimates of the population standard deviation (sigma) based on the corresponding estimation
type.

Capability Analysis Section

Data Summary
Number of Values 250
Sigma (Estimated) 7.798796
Mean (Estimated) 67.12

Number of Values
This is the number of observations in the capability analysis. While subgroups were used in the estimation of
sigma, they are no longer distinguished in the remainder of the capability analysis.
Sigma
This is the sigma that will be used for the capability analysis calculations.
Mean
This is the mean to be used for the capability analysis calculations.

Specification Summary
Corresponding
Specification Value Z-Value
Lower Limit 50 -2.195211
Upper Limit 80 1.651537
Target Value 65 -0.271837

This report lists the user-specified specification values, as well as the corresponding Z-value.
Specification
This column identifies the specification value type.

255-16
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Value
This is the user-input specification value. The target value is only required for the Cpm and Cpkm capability ratios.
Corresponding Z-Value
These are the z-values of the specification limits and target value, calculated using the formula
spec − µˆ
z spec =
σˆ

Performance Summary
LL ------- Observed Performance ------- - Normal Distribution Performance -
Specification Value # < LL % < LL PPM < LL % < LL PPM < LL
Lower Limit (LL) 50 4/250 1.6000% 16000.00 1.4074% 14074.25

UL ------- Observed Performance ------- - Normal Distribution Performance -

Specification Value # > UL % > UL PPM > UL % > UL PPM > UL
Upper Limit (UL) 80 11/250 4.4000% 44000.00 4.9314% 49314.49

----------- Observed Performance ----------- - Normal Distribution Performance -

Specification # Outside % Outside PPM Outside % Outside PPM Outside
Outside Both Limits 15/250 6.0000% 60000.00 6.3389% 63388.74

Range ------------ Observed Performance ------------ - Normal Distribution Performance -

Specification (UL - LL) LL ≤ # ≤ UL LL ≤ % ≤ UL LL ≤ PPM ≤ UL LL ≤ % ≤ UL LL ≤ PPM ≤ UL
Between Limits 30 235/250 94.0000% 940000.00 93.6611% 936611.26

This report gives the percentage of values inside or outside the specification limits. In this example, the observed
performance is similar to the Normal distribution (expected) performance.
Specification
This identifies the region to be examined for performance.
LL, UL, and Range
The LL and the UL values are the user-specified lower and upper specification limits. The range is the lower limit
subtracted from the upper limit.
Observed Performance – #
This gives the actual number of observed values in the corresponding region.
Observed Performance – %
This gives the percent of observed values in the corresponding region.
Observed Performance – PPM
This gives the parts per million number of observed values in the corresponding region.
Normal Distribution Performance – %
If the values are assumed to follow a normal distribution with mean µ̂ and standard deviation σ̂ , this is the
percent of values that would fall in the corresponding region. This is sometimes called the expected performance.
Normal Distribution Performance – PPM
If the values are assumed to follow a normal distribution with mean µ̂ and standard deviation σ̂ , this is the parts
per million number of values that would fall in the corresponding region. This is sometimes called the expected
performance.

255-17
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Process Capability Ratios

Process Capability Ratios
Confidence Level: 95.00%

Capability ---- Confidence Interval ----

Ratio Value Lower Limit Upper Limit
Cp (with C.I.) 0.641125 0.584820 0.697364
Cpk (with C.I.) 0.550512 0.486211 0.614813
Cpl 0.731737
Cpu 0.550512
Cpm 0.618673
Cpkm 0.531234

This report gives the values of the various capability ratios. Confidence intervals are given for Cp and Cpk. We
refer the reader to Montgomery (2013) or Ryan (2011) for interpretation details of each ratio.
Capability Ratio
This identifies the capability ratio of each line.
Value and Confidence Interval Limits
The formulas for each of these values are given earlier in this chapter in the Process Capability Ratios section
under Process Capability Formulas.

3- to 6-Sigma Limits
Limit
Type Mean Lower Limit Upper Limit
3-Sigma Limits 67.12 43.72361 90.51639
4-Sigma Limits 67.12 35.92482 98.31519
5-Sigma Limits 67.12 28.12602 106.114
6-Sigma Limits 67.12 20.32722 113.9128

Limits
The formulas for the limits are
LL = µˆ − mσˆ
UL = µˆ + mσˆ
where m is the multiplier 3, 4, 5, or 6.

Normality Tests Section

Normality Tests Section
User-Specified Alpha Level: 0.05

Normality Test Prob

Test Statistic Level Conclusion
Shapiro-Wilk 0.995 0.668324 Do Not Reject Normality Assumption
Anderson-Darling 0.455 0.268189 Do Not Reject Normality Assumption
Chi-Square 1.637 0.441089 Do Not Reject Normality Assumption

The details of the Shapiro-Wilk and Anderson-Darling (and other) Normality tests are discussed in the Normality
Tests procedure. The Chi-Square goodness of fit test for normality is obtained by dividing the data into bins, and
then comparing the observed counts to the expected counts for each bin using

255-18
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

χ =∑
2
k
(Oi − Ei )2
i =1 Ei
The individual observed and expected counts are detailed in the Chi-Square Test Frequency Distribution Details
section.

Chi-Square Test Frequency Distribution Details

------ Bin Boundaries ------
Lower Upper Actual Normal Diff. Actual Normal Diff. Chi-Sqr
Boundary Boundary Count Count Count Percent Percent Percent Amount
47.62301 1.0 1.6 -0.6 0.4 0.6 -0.2 0.00
47.62301 55.42181 19.0 15.1 3.9 7.6 6.1 1.5 0.65
55.42181 63.2206 59.0 60.4 -1.4 23.6 24.2 -0.6 0.03
63.2206 71.0194 101.0 95.7 5.3 40.4 38.3 2.1 0.29
71.0194 78.81819 55.0 60.4 -5.4 22.0 24.2 -2.2 0.49
78.81819 86.61699 14.0 15.1 -1.1 5.6 6.1 -0.5 0.17
86.61699 1.0 1.6 -0.6 0.4 0.6 -0.2 0.00
Total 250.0 250.0 0.0 100.0 100.0 0.0 1.64

This section summarizes the contribution of each bin to the Chi-Square goodness of fit test statistic.

Histogram Section

This histogram also displays the (vertical line) specification limits as well as the specification target.

255-19
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Example 2 – Capability Analysis for Individual Value Data

The capability analysis of individual value data is nearly the same as the analysis of subgroup data. The only
difference is the way in which sigma is estimated. This section presents an example of how to run a capability
analysis for individual value data. The data represent 200 part widths of a process that is assumed to be in control.
The specification limits for the process are 300 and 400, with a specification target of 350. The data used are in
the Capability dataset. We will analyze the variable Width of this dataset.
You may follow along here by making the appropriate entries or load the completed template Example 2 by
clicking on Open Example Template from the File menu of the Capability Analysis window.

1 Open the Capability dataset.

• From the File menu of the NCSS Data window, select Open Example Data.
• Click on the file Capability.NCSS.
• Click Open.

2 Open the Capability Analysis window.

• Using the Analysis menu or the Procedure Navigator, find and select the Capability Analysis procedure.
• On the menus, select File, then New Template. This will fill the procedure with the default template.

3 Specify the variables.

• On the Capability Analysis window, select the Variables tab.
• Change the Input Type to Response Column with Individual Values (no subgroups).
• Double-click in the Response Variable text box. This will bring up the variable selection window.
• Select Width from the list of variables and then click Ok. “Width” will appear in the Response Variable
box.

4 Set the Specification Limits.

• On the Capability Analysis window, select the Limits & Estimation tab.
• Enter 300 for Lower Limit. Enter 400 for Upper Limit.
• Enter 350 for Target Specification Value.

5 Set the Sigma Estimation Method.

• On the Capability Analysis window, select the Limits & Estimation tab.
• Change the Sigma Estimation to From Data – SD Approach.

6 Run the procedure.

• From the Run menu, select Run Procedure. Alternatively, just click the green Run button.

255-20
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Output
Mean Estimation Section for All Individuals
Number of Individuals 200

Estimation Type Estimate

Estimated Grand Average 346.79

Sigma Estimation Section for All Inidividuals

Estimation Estimated Estimated

Type Value Sigma
Ranges (R-bar) 28.84422 25.57112
Standard Deviation* 25.37945 25.37945

* Indicates the estimation type used in this report.

Capability Analysis Section

Data Summary
Number of Values 200
Sigma (Estimated) 25.37945
Mean (Estimated) 346.79

Specification Summary
Corresponding
Specification Value Z-Value
Lower Limit 300 -1.843618
Upper Limit 400 2.096578
Target Value 350 0.126480

Performance Summary

LL ------- Observed Performance ------- - Normal Distribution Performance -

Specification Value # < LL % < LL PPM < LL % < LL PPM < LL
Lower Limit (LL) 300 4/200 2.0000% 20000.00 3.2619% 32619.45

UL ------- Observed Performance ------- - Normal Distribution Performance -

Specification Value # > UL % > UL PPM > UL % > UL PPM > UL
Upper Limit (UL) 400 6/200 3.0000% 30000.00 1.8015% 18015.47

----------- Observed Performance ----------- - Normal Distribution Performance -

Specification # Outside % Outside PPM Outside % Outside PPM Outside
Outside Both Limits 10/200 5.0000% 50000.00 5.0635% 50634.91

Range ------------ Observed Performance ------------ - Normal Distribution Performance -

Specification (UL - LL) LL ≤ # ≤ UL LL ≤ % ≤ UL LL ≤ PPM ≤ UL LL ≤ % ≤ UL LL ≤ PPM ≤ UL
Between Limits 100 190/200 95.0000% 950000.00 94.9365% 949365.09

Process Capability Ratios

Confidence Level: 95.00%

Capability ---- Confidence Interval ----

Ratio Value Lower Limit Upper Limit
Cp (with C.I.) 0.656699 0.592195 0.721123
Cpk (with C.I.) 0.614539 0.537410 0.691669
Cpl 0.614539
Cpu 0.698859
Cpm 0.651509
Cpkm 0.609682

3- to 6-Sigma Limits

Limit
Type Mean Lower Limit Upper Limit
3-Sigma Limits 346.79 270.6516 422.9283
4-Sigma Limits 346.79 245.2722 448.3078
5-Sigma Limits 346.79 219.8927 473.6873
6-Sigma Limits 346.79 194.5133 499.0667

255-21
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Capability Analysis

Normality Tests Section

User-Specified Alpha Level: 0.05

Normality Test Prob

Test Statistic Level Conclusion
Shapiro-Wilk 0.986 0.039277 Reject Normality Assumption
Anderson-Darling 0.594 0.120995 Do Not Reject Normality Assumption
Chi-Square 2.714 0.257411 Do Not Reject Normality Assumption

Chi-Square Test Frequency Distribution Details

------ Bin Boundaries ------

Lower Upper Actual Normal Diff. Actual Normal Diff. Chi-Sqr
Boundary Boundary Count Count Count Percent Percent Percent Amount
283.3414 0.0 1.2 -1.2 0.0 0.6 -0.6 0.00
283.3414 308.7208 12.0 12.1 -0.1 6.0 6.1 -0.1 0.14
308.7208 334.1003 52.0 48.3 3.7 26.0 24.2 1.8 0.28
334.1003 359.4797 77.0 76.6 0.4 38.5 38.3 0.2 0.00
359.4797 384.8592 51.0 48.3 2.7 25.5 24.2 1.3 0.15
384.8592 410.2386 6.0 12.1 -6.1 3.0 6.1 -3.1 2.15
410.2386 2.0 1.2 0.8 1.0 0.6 0.4 0.00
Total 200.0 200.0 0.0 100.0 100.0 0.0 2.71

Histogram Section for All Individuals

The output descriptions for each section of the output are presented in Example 1. The only difference in formulas
in Example 2 compared to Example 1 is the difference in the calculation of the sigma estimate. In Example 2, the
common sample standard deviation formula using all the individual values is used to calculate the sigma estimate.

255-22
© NCSS, LLC. All Rights Reserved.