EWMA Charts

NCSS Statistical Software NCSS.
com
Chapter 249
Exponentially Weighted
Moving Average (EWMA)
Charts
Introduction
This procedure generates exponentially weighted moving average (EWMA) control charts for variables. Charts
for the mean and for the variability can be produced. The format of the control charts is fully customizable. The
data for the subgroups can be in a single column or in multiple columns. This procedure permits the defining of
stages. The target value can be entered directly or estimated from the data, or a sub-set of the data. Sigma may be
estimated from the data or a standard sigma value may be entered. Means and ranges/standard deviations can be
stored to the spreadsheet.
Exponentially Weighted Moving Average Control Charts

Similarly to the CUSUM chart, the EWMA chart is useful in detecting small shifts in the process mean. These
charts are used to monitor the mean of a process based on samples taken from the process at given times (hours,
shifts, days, weeks, months, etc.). The measurements of the samples at a given time constitute a subgroup.
The EWMA chart relies on the specification of a target value and a known or reliable estimate of the standard
deviation. For this reason, the moving average chart is better used after process control has been established.
249-1
© NCSS, LLC. All Rights Reserved.
NCSS Statistical Software NCSS.com
Exponentially Weighted Moving Average (EWMA) Charts
Other Control Charts for the Mean and Variation of a Process

Common control charts for detecting large process changes, or for establishing initial control of the process are
the X-bar and R charts (or X-bar and s charts).
Once a reliable estimate of the mean and standard deviation is available, the EWMA and CUSUM charts are
useful in detecting smaller shifts in the process mean.
The EWMA chart may also be used when only a single response is available at each time point. Another option
for single responses are the individuals and moving range (I-MR) control charts.
Control Chart Formulas

Suppose we have k subgroups, each of size n. Let xij represent the measurement in the jth sample of the ith
subgroup.
Formulas for the Points on the Chart

The ith subgroup mean is calculated using
n
∑x
j =1
ij
xi =
n
The points of the chart are obtained from the xi ’s by the following exponential smoothing operation
zi = λ xi + (1 − λ )zi −1
The value of z0 is set to the target mean. The value of λ is specified by the user.
Values for the points of the EWMA variation chart are given by substituting xi with the range or standard
deviation.
Estimating the EWMA Chart Center Line (Grand Mean)

In the EWMA Charts procedure, the target mean may be input directly, or it may be estimated from a series of
subgroups. If it is estimated from the subgroups the formula for the grand average 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
249-2
Estimating Sigma – Sample Ranges

Either the range or the standard deviation of the subgroups may be used to estimate sigma, or a known (standard)
sigma value may be entered directly. If the standard deviation (sigma) is to be estimated from the ranges, 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.
When n is one, we cannot calculate Ri since it requires at least two measurements. The procedure in this case is to
use the ranges of successive pairs of observations. Hence, the range of the first and second observation is
computed, the range of the second and third is computed, and so on. The average of these approximate ranges is
used to estimate σ.
Estimating Sigma – 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 
249-3
Estimating Sigma – Weighted 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 = s =  i =1 k 
 ni − k 
 ∑ i =1 
EWMA Chart Limits

The lower and upper control limits for the EWMA chart are calculated using the formulas
 σˆ  λ
LCLi = µ 0 − m  [
1 − (1 − λ )
2i
]
 n  2−λ
UCLi = µ 0 + m
 σˆ  λ
 [
1 − (1 − λ )
2i
]
 n  2−λ
where m is a multiplier (often set to 2.7 or 3), and λ is the exponential smoothing constant.
Note that the values of limits change with each successive subgroup, but tend to level off at around the tenth
subgroup.
Values for the limits of the EWMA variation chart are given by substituting µ0 with the range or standard
deviation center line value (such as R-bar or s-bar).
249-4
Data Structure
In this procedure, the data may be in either of two 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
. . . . .
. . . . .
. . . . .
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.
249-5
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.
• 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
. . .
. . .
. . .
Variables – Response Column

Response Variable
Specify the column with the data values. The data values are separated into subgroups below using the Subgroup
Specification options.
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,
6, 4; the second would be 3, 7, 6; and so on.
249-6
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.
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
. . .
. . .
. . .
249-7
If only one variable is specified, NCSS automatically generates an individuals chart with a moving-range of size
2.
Stages
Number of Stages
Specify whether the analyses and charts are to be produced based on a single set of subgroups, or a series of sets
of subgroups. Typically a single stage is used unless you wish to have separate estimation for various segments of
the subgroups. When using multiple stages, the software requires that there be at least one subgroup in every stage
specified.
Stage Specification
Specify whether the various stages will be defined using a variable (column) with a unique value for each stage,
or by entering a range of rows for each stage.
• Stage Variable
X1 X2 X3 Stage
5 6 4 1
3 7 6 1
4 5 7 1
6 5 4 1
3 7 6 1
5 8 3 2
2 6 4 2
3 6 5 2
4 5 8 2
. . . .
. . . .
. . . .
• Enter a range for each stage

1-50, 51-100, 101-150
This would produce three stages. The first stage would be made up of rows 1 to 50, the second stage would be
rows 51 to 100, and the third stage would be rows 101 to 150.
Stage Variable
Specify the variable (column) that contains the identifiers for each stage.
X1 X2 X3 Stage Variable
5 6 4 1
3 7 6 1
4 5 7 1
6 5 4 1
3 7 6 1
5 8 3 2
2 6 4 2
3 6 5 2
4 5 8 2
. . . .
. . . .
. . . .
A new stage is created for each change in the Stage Variable, going down.
249-8
Stage Ranges
Enter a range for each stage using a dash. Separate each stage with a comma.
Example: 1-50, 51-100, 101-150
This would produce three stages. The first stage would be made up of rows 1 to 50, the second stage would be
rows 51 to 100, and the third stage would be rows 101 to 150.
Specify Rows in Calculations

Specification Method
Select which method will be used to specify the rows of the data to be used to form subgroups.
• 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 for use in calculations.
• First N Rows (Enter N)

The data beginning at Row 1 and ending at Row N will be used in calculations.
• Last N Rows (Enter N)

Subgroups will be formed from the last N rows of the dataset.
• Keep Rows Variable

Specify a variable and a value in that variable column that will be used to determine which rows are used to
form the subgroups.
• 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 to form the subgroups.
First Row
Specify the beginning row to be used for the first subgroup.
Last Row
Specify the last row to be used for the last subgroup.
N
Enter the number of rows to be used in forming subgroups.
Keep Rows Variable
This variable (column) is used to specify which rows of the data will be used to form the subgroups for the
calculations.
Keep Rows Value
This value determines which rows of the Keep Rows Variable will be used in the calculation portion of the
analysis.
249-9
Remove Rows Variable

This variable (column) is used to specify which rows of the data will not be used to form the subgroups for the
calculations.
Remove Rows Value
This value determines which rows of the Remove Rows Variable will not be used in the calculation portion of the
analysis.
Specify Rows in Charts

Specification Method
Select which method will be used to specify the rows of the data to be used to form subgroups for the charts.
• 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 for use in the plots.
• First N Rows (Enter N)

The data beginning at Row 1 and ending at Row N will be used in the plots.
• Last N Rows (Enter N)

Subgroups will be formed from the last N rows of the dataset.
• Keep Rows Variable

Specify a variable and a value in that variable column that will be used to determine which rows are used to
form the subgroups.
• 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 to form the subgroups.
First Row
Specify the beginning row to be used for the first subgroup.
Last Row
Specify the last row to be used for the last subgroup.
N
Enter the number of rows to be used in forming subgroups.
Keep Rows Variable
This variable (column) is used to specify which rows of the data will be used to form the subgroups for the plots.
Keep Rows Value
This value determines which rows of the Keep Rows Variable will be used in the plots.
Remove Rows Variable
This variable (column) is used to specify which rows of the data will not be used to form the subgroups for the
plots.
249-10
Remove Rows Value

This value determines which rows of the Remove Rows Variable will not be used in the plots.
Labels (Optional)
Subgroup Label Variable
Specify a variable (column) that contains the desired x axis (subgroup) labels for plots. If left blank, the plot will
use the subgroup number. The format of the labels is controlled on the x axis tab of the plot format window.
Point Label Variable
Specify a variable (column) that contains the desired individual point labels for plots. If left blank, no point labels
are shown. The format of the labels is controlled on the main chart tab of the plot format window.
Target & Sigma Tab

The options on the Target & Sigma tab are used to specify the target value and sigma.
Target Value Options

Target Value Specification
Specify whether the target value is estimated from the data, or whether it will be specified directly.
• From Rows in Calculations Data

Estimate the target value from the subgroups specified for calculations.
• Enter Target Value(s)

Specify the target value directly. If multiple stages are used, separate the target value for each stage by spaces.
• Use a Variable with Target Value(s)

Specify a column containing the target value in row 1. If multiple stages are used, a target value should be
entered in a separate cell for each stage, beginning with row 1 for the first stage.
Target Value(s)
Enter the target value to be used. If multiple stages are used, separate the target values for each stage by spaces.
Target Value(s) Variable
Specify a column containing the target value in row 1. If multiple stages are used, a target value should be entered
in a separate cell for each stage, beginning with row 1 for the first stage.
Limit Multipliers
Primary Limit Multiplier
This option specifies the multiplier of sigma for the primary control limits. For the well-known 3-sigma limits, the
multiplier is set to 3.
Specification Limits
Lower Limit
Enter an optional lower specification limit for display on the chart. These limits are not the control limits.
249-11
Upper Limit
Enter an optional upper specification limit for display on the chart. These limits are not the control limits.
Spec Value
Enter an optional specification value for display on the chart.
Sigma Estimation Options

Range or SD Estimation
Specify whether the estimation of sigma will be based on the average range or the average standard deviation.
• Range
The average range will be used to estimate sigma.
• SD
The average standard deviation will be used to estimate sigma.
Sigma Specification
Specify the method by which Sigma will be estimated for use in the charts.
• 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). Only the subgroups specified for use in calculations will be used.
• From Data – Weighted Approach Estimate

This method estimates s-bar using a special formula that is recommended when the subgroup size varies
across subgroups. This option should only be used when the Range or SD Estimation is set to SD. Only the
subgroups specified for use in calculations will be used.
• Enter Standard Sigma Value(s)

In this case the sigma value is entered directly. The corresponding R-bar or s-bar value is back-calculated
from this value. If multiple stages are used, separate the sigma values for each stage by spaces.
• Use a Variable with Standard Sigma Value(s)

Specify a column containing the standard sigma value in row 1. The corresponding R-bar or s-bar value is
back-calculated from this value. If multiple stages are used, a sigma value should be entered in a separate cell
for each stage, beginning with row 1 for the first stage.
Sigma Value(s)
Enter the value to be used for the standard sigma. If multiple stages are used, separate the sigma values for each
stage by spaces.
Sigma Variable
Specify a column containing the standard sigma value in row 1. The corresponding R-bar or s-bar value for
variation plots is back-calculated from this value. If multiple stages are used, a sigma value should be entered in a
separate cell for each stage, beginning with row 1 for the first stage.
249-12
Reports Tab
The following options control the format of the reports.
Specify Reports
Target Value and Sigma Summary Section
This report gives the numeric values of the target value and sigma, as well as the sigma estimation.
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.
EWMA Charts Tab

This panel sets the options used to define the appearance of the two EWMA charts.
Select Plots
EWMA Chart, EWMA Variability Chart
These charts are 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 Quality Control Chart Format Options
below for more formatting details).
3. A second checkbox used to indicate whether the chart can be edited during the run.
EWMA Parameter
This specifies the value of the smoothing parameter (λ) in the EWMA charts. The range for this parameter is 0 to
1 (not including 0). Typically, a value between 0.05 and 0.25 works well in practice. Popular choices include
0.05, 0.10, and 0.20.
249-13
Storage Tab
The options on this panel control the automatic storage of the means on the current dataset.
Storage Columns
Store Means or Ranges/SDs in Column
You can automatically store the means, ranges, or standard deviations of each subgroup into the column specified
here. Whether the range or standard deviation is stored depends on the choice of Range or SD Estimation on the
Target & Sigma tab.
Warning: Any data already in this column is replaced. Be careful not to specify columns that contain important
data.
Quality Control Chart Format Window Options

This section describes a few of the specific options available on the first tab of the control chart format window,
which is displayed when a quality control chart format button is pressed. Common options, such as axes, labels,
legends, and titles are documented in the Graphics Components chapter.
249-14
[Xbar] / [Range] Chart Tab
Symbols Section
You can modify the attributes of the symbols using the options in this section.
A wide variety of sizes, shapes, and colors are available for the symbols. The symbols for in-control and out-of-
control points are specified independently. There are additional options to label out-of-control points. The label
for points outside the primary control limits is the subgroup number. The label for points that are out-of-control
based on the runs test is the number of the first runs test that is signaled by this point.
The user may also specify a column of point labels on the procedure variables tab, to be used to label all or some
of the points of the chart. The raw data may also be shown, based on customizable raw data symbols.
Lines Section
You can specify the format of the various lines using the options in this section. Note that when shading is
desired, the fill will be to the bottom for single lines (such as the mean line), and between the lines for pairs of
lines (such as primary limits).
Lines for the zones, secondary limits, and specification limits are also specified here.
249-15
Titles, Legend, Numeric Axis, Group Axis, Grid Lines, and Background
Tabs
Details on setting the options in these tabs are given in the Graphics Components chapter. The legend does not
show by default, but can easily be included by going t1o the Legend tab and clicking the Show Legend checkbox.
Example 1 – EWMA Chart

This section presents an example of how to produce an EWMA chart. The data represent 50 subgroups of size 5.
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 EWMA Charts 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 EWMA Charts window.

• On the menus, select Graphics, then Quality Control Charts, then EWMA Charts. The EWMA Charts
procedure will be displayed.
• On the menus, select File, then New Template. This will fill the procedure with the default template.
3 Specify the variables.

• On the EWMA Charts 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 Specify the target value and sigma.

• On the Moving Average Charts window, select the Target & Sigma tab.
• Set Target Value Specification to Enter Target Value(s). Set the Target Value to 67.
• Set Sigma Specification to Enter Sigma Value(s). Set the Sigma Value to 8.
5 Run the procedure.

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

Target Value Section for Subgroups 1 to 50
Number of Subgroups 50
Target Value Specification Value

User-Specified Target Value 67
This section displays the target value that is used in the EWMA chart.
Target Value Specification
This is the method by which the target value is obtained.
249-16
Target Value
This is value of the target value used to create the moving average chart.
Sigma Specification Section

Sigma Specification Section for Subgroups 1 to 50
Sigma User-Specified
Specification Sigma Value
User-Specified 8
This section shows the specification of the standard deviation used in the EWMA chart.
Sigma Specification
This is the method by which the sigma value is obtained.
User-Specified Sigma Value
This is value of the sigma entered by the user, and is the value used to create the moving average chart.
EWMA Chart
This EWMA plot shows the progress of the weighted moving averages across the 50 subgroups. There does not
appear to be an indication of a change in the process mean.
249-17
Example 2 – EWMA Chart with Additional Formatting

This example uses the same setup as Example 1, except that a variety of improvements are made in the plot
format. These improvements are made by clicking the EWMA Chart format button on the EWMA Chart tab.
You can load the completed template Example 2 by clicking on Open Example Template from the File menu of
the EWMA Charts window.
EWMA Chart
As shown here, a variety of enhancements can be made to the formatting of the control charts to make the chart as
easy to read as possible.
249-18

EWMA Charts

Uploaded by

Copyright:

Available Formats

EWMA Charts

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EWMA Charts

Uploaded by

Copyright:

Available Formats

NCSS Statistical Software NCSS.

Exponentially Weighted Moving Average Control Charts

Other Control Charts for the Mean and Variation of a Process

Control Chart Formulas

Formulas for the Points on the Chart

Estimating the EWMA Chart Center Line (Grand Mean)

Estimating Sigma – Sample Ranges

Estimating Sigma – Sample Standard Deviations

Estimating Sigma – Weighted Approach

EWMA Chart Limits

Alternative example dataset

• Response Column and Subgroup Column or Subgroup Size

• Multiple Columns with One Subgroup Per Row

Variables – Response Column

Variables – Multiple Columns

• Enter a range for each stage

Example: 1-50, 51-100, 101-150

Specify Rows in Calculations

• Enter First Row and Last Row

• First N Rows (Enter N)

• Last N Rows (Enter N)

• Keep Rows Variable

• Remove Rows Variable

Remove Rows Variable

Specify Rows in Charts

• Enter First Row and Last Row

• First N Rows (Enter N)

• Last N Rows (Enter N)

• Keep Rows Variable

• Remove Rows Variable

Remove Rows Value

Target & Sigma Tab

Target Value Options

• From Rows in Calculations Data

• Enter Target Value(s)

• Use a Variable with Target Value(s)

Sigma Estimation Options

• From Data – R-bar or s-bar Estimate

• From Data – Weighted Approach Estimate

• Enter Standard Sigma Value(s)

• Use a Variable with Standard Sigma Value(s)

EWMA Charts Tab

Quality Control Chart Format Window Options

[Xbar] / [Range] Chart Tab

Example 1 – EWMA Chart

1 Open the QC dataset.

2 Open the EWMA Charts window.

3 Specify the variables.

4 Specify the target value and sigma.

5 Run the procedure.

Target Value Section

Target Value Specification Value

Sigma Specification Section

Example 2 – EWMA Chart with Additional Formatting

You might also like