In Touch SPC
In Touch SPC
Revision A
September 2002
Contents
Contents
Welcome to SPCPro...........................................7
Installing SPCPro ................................................................................... 7
Upgrading SPCPro from a Previous Version...................................... 8
System Requirements............................................................................. 8
The Basis of SPC.................................................................................... 9
SPC Program Design Considerations....................................................11
SPCPro Features....................................................................................11
SPC Chart Wizards ............................................................................11
Variable Sized Display Windows ..................................................... 12
SPC Analysis .................................................................................... 12
Integrated Alarming.......................................................................... 13
Distributed SPC ................................................................................ 13
Multiple Data Entry Schemes........................................................... 13
About this Manual................................................................................ 15
Technical Support................................................................................. 16
Your FactorySuite License ................................................................... 17
Contents
Contents
Contents
Welcome to SPCPro
Welcome to SPCPro
Contents
Installing SPCPro
System Requirements
The Basis of SPC
SPC Program Design Considerations
SPCPro Features
About this Manual
Technical Support
Your FactorySuite License
Installing SPCPro
The Wonderware FactorySuite installation program is used to install InTouch
and its add-ons. InTouch runs on Microsoft Windows 2000 or Windows NT
operating systems. The installation program creates directories as needed and
copies files from the compact disk to your hard drive.
System Requirements
To run SPCPro, we recommend the following hardware and software:
Welcome to SPCPro
10
In SPC terms, the natural variations of a process are referred to as the control
limits. They are determined from many measurements of data taken over a
long period of time. Within the scope of SPC, a measurement refers to a single
instance of data, for example, a temperature is 78 degrees. A measurement is
also called a reading or observation. A sample is a collection of one or more
measurements used to calculate a point on an SPC chart. A group is any
complete set of samples that is used to create an SPC chart. When multiple
measurements are used as a sample, they are sometimes referred to as a subgroup.
Although the mathematical calculations involved in determining system
control limits aren't that difficult, the theory behind them is obscure to most.
The end result is that the control limits calculated for a given process represent
a range into which 99.7% of the in control samples will fall. The calculation of
control limits is an iterative process. In the first pass, all samples are used and
initial limits are defined. Then, samples falling outside of the limits are
eliminated and the limits are recalculated. This process continues until all
remaining samples fall within the control limits, which represent true common
cause control limits. Future samples can now be graphed against these limits.
When a process is in control, a completely random pattern will emerge in the
charts. The user can use these plots to monitor how the process is doing. Any
discernible trends, unusual cycles or out-of-limit samples will be cause for
alarm.
Typical SPC charts deal with two types of data: Variable and Attribute.
Variable data consists of measurable parameters such as diameters, weights,
temperatures, etc. Attribute data is a quantitative measurement, such as the
number of rejected parts, number of defects per assembly, etc.
There are two distinct phases involved in setting up a quality program for a
manufacturing process. The first phase is analytical. Quality Control engineers
must evaluate the process to determine what data should be monitored and how
often samples should be collected. Data must then be collected to obtain a
history of the process. A process description can be derived from this history
that will indicate the center of the process and natural variation.
Phase two of the program involves using the process description as a control
element for alarm monitoring and optimization. Samples are collected on a
regular basis and plotted on control charts whose centerline and range were
determined by the historical data. The user can monitor these charts on their
user interface screens, to see if the process is stable and within limits as it runs
on the factory floor. If there are Special Cause variations in the operation,
appropriate actions can be taken based on these observations to correct out-oflimit operation. In addition, the effects of any process setpoint changes or
optimization strategies can also be observed and evaluated objectively. These
continual improvement and optimization strategies are at the heart of practical
implementation of Total Quality Management (TQM).
Welcome to SPCPro
11
SPCPro Features
The SPCPro add-on program provides you with the following features:
12
For more information on using the SPC Chart Wizards, see Chapter 3, "Using
the SPC Chart Wizards."
SPC Analysis
The following briefly describes each supported SPC Analysis type:
X Individual
This type of chart is used for a single value of variable data to determine the
stability of the center of the process. The control limits for these charts are
determined by calculating the standard deviation of the sample population.
X bar, R chart
These types of charts are used for multiple values of variable data to determine
the stability of the center and range of the process. The width is determined by
subtracting the minimum value from the maximum value.
X bar, s charts
These types of charts are used for multiple values of variable data to determine
the stability of the center and standard deviation of the process. The width is
determined by calculating the standard deviation of the measurements.
C charts
C charts (number of nonconformity's) are used for monitoring the number of
defects in a fixed sample size. The scale is the count.
P charts
P charts (fraction nonconforming measurements) are used for monitoring the
proportion of defects for any given sample size. The scale is the fractional
defects.
Welcome to SPCPro
13
NP charts
NP charts (number of nonconforming items) are used in lieu of a P chart when
the sample size is constant.
U charts
U charts (nonconformity's per unit) are used in lieu of a C chart when the
sample size is not constant.
EWMA charts
EWMA charts (exponentially-weighted moving averages) are created to place
more emphasis on recent samples and less on the values some distance back in
the collected data.
CUSUM charts
CUSUM control charts (cumulative sum) can detect small process shifts faster
than standard control charts. But they are not a direct substitute for standard
control charts, they should be used in combination with standard control
charts.
Integrated Alarming
SPC alarms are associated with InTouch tagnames and are displayed and
acknowledged the same as any other InTouch alarm. Alarms can also be
acknowledged by right-clicking the sample to display the SPC menu and then
clicking Ack Sample. Alarm monitoring can be selected for out-of-limit
conditions, four custom run rules and seven different Western Electric run
rules.
Distributed SPC
SPC allows you to build Distributed ODBC Based SPC Applications.
Distributed SPC allows remote modification of datasets, automatic update of
displayed data with values from remote datasets, and the displaying of
products in charts that are not being collected.
14
Welcome to SPCPro
15
Flagged Samples
Samples can be flagged to alert the user.
16
Assumptions
This manual assumes you are:
Technical Support
Wonderware Technical Support offers a variety of support options to answer
any questions on Wonderware products and their implementation.
Prior to contacting technical support, please refer to the relevant chapter(s) in
your InTouch User's Guide for a possible solution to any problem you may
have with your InTouch system. If you find it necessary to contact technical
support for assistance, please have the following information available:
1.
2.
3.
The type and version of the operating system you are using. For example,
Microsoft Windows NT Version 4.0 SP4 workstation.
4.
5.
Any relevant output listing from the Wonderware Logger, the Microsoft
Diagnostic utility (MSD) or, any other diagnostic applications.
6.
Details of the attempts you made to solve the problem(s) and your results.
Welcome to SPCPro
7.
8.
17
2.
Click View License. The License Utility - LicView dialog box appears.
For more information on the licensing viewing utility, see your online
FactorySuite System Administrator's Guide.
18
C H A P T E R
19
Configuring an ODBC
Database
You will need to configure SPCPro before you can correctly use it. You must
have Microsoft ODBC drivers installed to use SPCPro. The two ODBC drivers
that SPCPro currently support are: Microsoft Access driver version
4.00.3711.08 and Microsoft SQLServer version 3.70.06.23.
It is extremely important to setup your new SPCPro database and to import any
datasets previously created with SPC (version 6.0 or earlier) prior to running
your InTouch application. This chapter explains how to configure a Microsoft
Access database or, a Microsoft SQLServer database, and how to setup user
ID's.
Contents
Configuring a SPC Database
20
Chapter 1
Start WindowMaker.
2.
On the Special menu, point to SPC and then, click Database or, in the
Application Explorer under SPC, double-click Database. The Configure
SPC Database dialog box appears.
3.
Click the Database Type arrow and select Microsoft Access (Local).
21
4.
Click the ODBC Data Source arrow and select <NEW> .The ODBC
Data Source Administrator dialog box appears.
5.
Click the User DSN tab and then, select your ODBC data source in the list
or, click Add. The Create New Data Source dialog box appears.
22
Chapter 1
6.
Select your ODBC driver in the list and then, click Finish. The ODBC
Microsoft Access 97 Setup dialog box appears.
Note This dialog box also appears when you click the Modify button in
the Configure SPC Database dialog box to edit an existing database
configuration.
7.
In the Data Source Name box, type a unique name for your data source.
For example, "SPCdata."
8.
9.
23
In the Database Name box, type a new name for your database. For
example "SPCdata.mdb."
10. Select a directory to store your new database file and then, click OK. A
message box appears confirming the successful creation of the database.
11. Click OK. The ODBC Microsoft Access Setup dialog box reappears.
12. Click OK. The ODBC Data Source Administrator dialog box reappears.
Select your newly created User Data Source. For example, SPCdata.
13. Click OK. The Configure SPC Database dialog reappears.
14. Click Save. A message box appears warning you that the new database is
not initialized.
15. Click Yes to initialize the database. A message box appears informing you
that the database was successfully initialized.
16. Click OK.
17. Click Verify. The connection status for the ODBC database will be
verified. If your ODBC connection is valid, a green light is displayed in
the dialog box.
18. Click OK.
Start WindowMaker.
2.
On the Special menu, point to SPC and then, click Database or, in the
Application Explorer under SPC, double-click Database. The Configure
SPC Database dialog box appears.
Note To configure an SQL database for use with SPCPro you must create
a database on a Microsoft SQLServer 7.0 using Microsoft Enterprise
Manager (Start/Programs/Microsoft SQLServer 7.0). Also, you must have
Administrator privileges (SA).
24
Chapter 1
Tip If you are using Microsoft SQLServer to store data for SPCPro, use
these instructions to connect to your database.
3.
Click the Database Type arrow and select Microsoft SQLServer in the
list.
25
4.
Click the ODBC Data Source arrow and select your ODBC Data Source
in the list. The ODBC Data Source Administrator dialog box appears.
5.
Click the User DSN tab and then, select your ODBC data or, click Add.
The Create New Data Source dialog box appears.
6.
Select the ODBC driver that you want to use. For example, SQLServer.
26
Chapter 1
7.
Click Finish.
Note The ODBC SQLServer setup dialog boxes appear. Follow the on
screen instructions. For more information on these dialog boxes consult
your database administrator or your Microsoft SQLServer documentation.
8.
9.
In the Admin User ID box, type the user name for your logon account to
the SQLServer.
27
Tip Data is retained until the specified number of days is exceeded. Once
the number of days is exceeded, the data is automatically purged. For
example, if 2 is entered, by the 4th day, the first day's data files is purged.
There will actually be 3 days of data; the previous two day's plus the
current day's.
Application performance will be impacted if you don't periodically purge
old data. By default, this field is set to zero (0) (represents infinity) and the
data is never deleted. If you use the default, it is recommended that you
periodically either purge the older data or, archive it to avoid running out
of disk space.
13. Click Save, then click OK to close the dialog box.
Start WindowMaker.
2.
On the Special menu, point to SPC and then, click Users or in the
Application Explorer under SPC, double-click Users. The SPC Users
dialog box appears.
3.
4.
5.
28
Chapter 1
In the Full Name box, type in your full name. (This is an optional
descriptive field.)
7.
In the User Type box, click the arrow and select Agent.
8.
Click OK or click Save to configure another user and then, click OK.
Note SPC users are for SPC only and not SQLServer. These users are not
added to the SQLServer's user list.
C H A P T E R
29
In order to use SPC Pro, you must create SPCPro Datasets, Indirect Datasets,
and Products for each Dataset. This chapter will cover how to create datasets
and how to import and convert Datasets created in older versions of SPC to use
with SPC Pro.
Contents
Configuring SPC Datasets
Configuring Indirect Datasets
Importing SPC Datasets
30
Chapter 2
Start WindowMaker.
2.
On the Special menu, point to SPC, and then click Datasets or, in the
Application Explorer under SPC, double-click Datasets. The SPC
Datasets Configuration dialog box appears.
3.
4.
In the Dataset Name box, type a new unique dataset name or, click Select
(to choose an existing Dataset). The Select a Dataset dialog box appears.
Select the Dataset that you want to use. The dialog box will close, and the
selected Dataset name will automatically be inserted into the Dataset
Name box.
31
5.
6.
Select the tagname that you previously defined to be used for data
collection. The dialog box will close and the selected tagname will
automatically be inserted into the Collection Tagname box.
7.
Select the Scooter Tagname option if you plan to use a scooter with your
SPC chart, and then type a Analog (Real or Integer) memory type tagname
in the box or, double-click the blank box and select the tagname as
described above.
Tip You must define the tagname that you plan to use for the scooter prior
to configuring the Dataset.
Note WindowViewer will not include any tagname used in SPCPro as a
Collection Tag or Scooter Tag in the tagname usage count.
32
Chapter 2
8.
Click the Analysis button. The SPC Analysis Selection dialog box
appears.
9.
In the Analysis Type group, select the analysis type that you want to use
for this dataset, and then click OK.
Note After clicking Save, the SPC Analysis Selection cannot be
changed.
For more information on SPC Analysis Types see the "Welcome to
SPCPro" section.
10. In the Samples Per Chart group boxes, type the valid number of samples
to be displayed for each SPC Chart type or, use the default of 20.
For more information on the SPC Chart types, see Chapter 3, "Using the
SPC Chart Wizards."
33
11. The Sample Info boxes become active based upon the type of Analysis
selected as follows.
Sample Size
Time Based
34
Chapter 2
Event Based
13. The EWMA Parameters group becomes active when the Analysis Type
EWMA is selected.
Tighter Control(2.58 sigma)
Smoothing Factor
14. Click Save. The Products, Alarms, and Causes buttons will become
active.
35
Note Once you have configured a new Dataset, you must configure at
least one Product for it before you can close the SPC Database
Configuration dialog box.
For more information on configuring a Product, see "Configuring Dataset
Products."
Start WindowMaker.
2.
On the Special menu, point to SPC, and then click Datasets or, in the
Application Explorer under SPC, double-click Datasets. The SPC
Datasets Configuration dialog box appears.
36
Chapter 2
3.
4.
In the Name box, type the name that you want to use for the product.
5.
The Center Chart group's options are used to set chart values for the
control limits, specification limits, centerline, and target.
6.
The Width Chart group's options are used to set the mean and the control
limits of Range or Standard Deviation charts. Enter a reasonable value in
each box. (The values can be changed at any time.) The new values will be
reflected in the next entered sample. These options can also be changed
through DDE at Runtime. The Dataset keeps a separate copy of chart
values for each configured product. This option is available for X bar, R,
X bar, s and Moving X, Moving R charts.
7.
In the Display Titles group type the titles that you want to use for each
display for every product. A separate title can be displayed for each type
of chart.
Tip If multiple products are defined for this Dataset, click Save after each
defined product, and then click New to define the next product.
8.
Start WindowMaker.
2.
On the Special menu, point to SPC, and then click Datasets or, in the
Application Explorer under SPC, double-click Datasets. The SPC
Datasets Configuration dialog box appears.
3.
4.
5.
Select the product that you want to modify. The Products dialog box
reappears displaying the selected product's configuration settings.
6.
37
Start WindowMaker.
2.
On the Special menu, point to SPC, and then click Datasets or, in the
Application Explorer under SPC, double-click Datasets. The SPC
Datasets Configuration dialog box appears.
3.
In the SPC Dataset Configuration dialog box, click Alarms. The SPC
Alarms Selection dialog box appears.
38
Chapter 2
4.
Select the desired SPC Alarm options that you want to be monitored by
the SPC program. There are three types of alarm areas, Limit Alarms,
Standard Deviation Alarms, and Consecutive Alarms. Here are a few
tips:
Tip In the ! of Last ! Samples Outside of ! Standard Deviations,
type three numbers for the alarm. This alarm can include points on both
sides of the center line. The second alarm option is the same, except all
points are either above or below the center line. Also, the first two fields
are integers and the last field is a real.
If the Consecutive Samples Outside of 1 Standard Deviation option is
set to 8, this will allow eight conditions outside of the standard deviation
range to be reached before the alarm condition would be reported to the
dataset.
5.
All of the Alarm options have a Priority level. A valid priority level is
between 1 and 999. This value represents the severity of the alarm with 1
being the most severe. By creating alarm ranges using priorities and
assigning alarms to each, you can easily filter out critical alarms from noncritical ones.
For more information on alarm priorities, refer to your InTouch User's
Guide.
6.
Click OK.
2.
3.
\Spcpro!$system
\\NodeName\spcpro!$system
39
Tip Both can be used as the Alarm Query provider. If there are two
providers make sure you separate the provider names with a space. For
example: \InTouch!$system (space) \spcpro!$system
40
Chapter 2
4.
5.
Configure your Message property sheet.In the dialog below, the following
configuration was selected.
Option
Description
Date
Time
Comment
Alarm Name
Alarm
Provider
shows the node and the application that provided the alarm.
41
Note The preceding options were configured for the example below
Start WindowMaker.
2.
On the Special menu, point to SPC, and then click Datasets or, in the
Application Explorer under SPC, double-click Datasets. The SPC
Datasets Configuration dialog box appears.
42
Chapter 2
3.
In the SPC Dataset Configuration dialog box, click Causes. The Special
Cause Configuration dialog box appears.
4.
In the New Special Cause box, type a cause description with a code
number. Then click the >> button or the ENTER key, to add the new
special cause to the Special Causes list box. This number is used to
identify the columns in the Pareto display chart. For example, type 1Startup
5.
Add as many special cause descriptions as you need. Each special cause
configured will be displayed in the list box.
Note Special Causes can be attached to any sample during runtime
through the Sample Information dialog box or through DDE.
6.
Click OK.
Tip The Special Cause Configuration dialog box can be displayed in
runtime by right-clicking a sample, and then selecting Add/Delete
Causes.
Start WindowMaker.
43
2.
On the Special menu, point to SPC, and then click Indirect Datasets or,
in the Application Explorer under SPC, double-click Indirect Datasets.
The Indirect SPC Datasets Configuration dialog box appears.
3.
In the Indirect Dataset Name box, type a unique name for the indirect
dataset (maximum of 31 characters).
4.
In the Default Dataset box, type the name of the Dataset to which you
want to link the Indirect Dataset (maximum of 31 characters) or, click
Select. The Select a Dataset dialog box appears.
5.
44
Chapter 2
Start the spcpro.exe utility. The SPCPro Server utility program appears.
Note You can use the SPCPro Server utility to perform all SPCPro
functions. For example, configure a Database, configure users, and create
Datasets and Indirect Datasets. By using the SPCPro Server utility, you
do not need to configure SPC from within WindowMaker.
2.
On the Database menu, click Import. The Open dialog box appears.
45
3.
4.
In the Import Options group, select the Select All Datasets option to
import all Datasets from the selected application or, Select All Products
to import all Products from the selected application or, pick and choose the
datasets and products from the listing on the left.
5.
Select what samples you want to import in the second Import Options
group as follows:
Import No Samples
Import Range of Samples Import only the samples for the specified From
Date and To Date.
6.
Click Import. The datasets will be imported and converted to the new
SPCPro format and are now ready to view in SPC Pro.
Note If there is a lot of data to import, this process could take some time.
46
Chapter 2
C H A P T E R
47
SPC Chart wizards are used to display the contents of Datasets. There are three
types of SPC chart objects available: Control Chart, Histogram and Pareto
charts. SPC chart objects are "wizards" that are simply pasted into your
window and then configured and linked to a Dataset.
You can configure the SPC Control Chart wizard to display X Individual, X bar
- R, X bar - s, Moving-X Moving-R, CUSUM, EWMA, C, P, U, and NP
Charts.
For more information on configuring Datasets, see Chapter 2, "Creating SPC
Datasets."
Contents
Control Charts
Histograms
Pareto Charts
Installing the SPC Chart Wizards
SPC Limits Wizard
Configuring the SPC Control Chart Wizard
Configuring the SPC Histogram Wizard
Configuring the SPC Pareto Chart Wizard
Configuring the SPC Limit Wizard
Control Charts
SPC data is displayed using X-Y charts, calculating a point for each sample or
sub-group. The points are connected by lines to form a control chart, which
provides graphical feedback to the user about the process being monitored. A
centerline is used to show the average of all points in a group. An upper control
limit line is displayed at three standard deviations above the centerline. The
lower control limit line is the centerline minus three standard deviations. Upper
and lower specification limit lines show the arbitrary top and bottom limits of
acceptable output. A target line shows the desired average value of the process
(which should be the same as the centerline) and zone lines are reference lines
that are plus and minus one and two standard deviations away from the
centerline.
InTouch SPCPro Users Guide
48
Chapter 3
Note Working in an SPC Control Chart does not affect the inactivity timer in
Runtime.
If a sample falls outside of the control limits (or breaks one of the run rules), an
alarm will be generated and the user may note the Special Cause that produced
the out of control sample. The following is an example of the SPC Control
Chart wizard:
Note Control charts will not refresh or scroll if another object such as a
pushbutton overlaps the chart.
49
Histograms
Histograms are created from the raw measurement data used for the control
charts and are used to display the distribution and frequency of the collected
data. A normal process will have a bell-shaped distribution of data values. Any
other distribution shape is cause for investigation. The following is an example
of the SPC Histogram wizard:
50
Chapter 3
Pareto Charts
Pareto charts are used to graphically present the number of occurrences of
Special Causes. Since the user usually enters Special Cause notations when
acknowledging alarms, the Pareto chart would use these entries over some
specified number of samples and present them in the form of a descending bar
graph. While there are many possible causes for out of control samples, it is
usually only one or two Special Causes that produce the bulk of bad samples.
Pareto charts help zero in on the most offensive causes. The following is an
example of the SPC Pareto Chart wizard:
Note Pareto charts recalculate control limits for every new sample entered.
Start WindowMaker.
2.
51
3.
52
Chapter 3
Note If SPC Chart wizards do not appear in the Wizard Selection dialog
box, you will need to initialize them. On the Special menu, point to
Configure, and then click Wizard/ActiveX Installation. You must click
the OK button to install SPC Chart wizards.
2.
3.
In the display area, select the SPC chart wizard that you want to use, and
then click OK or double-click the wizard. The dialog box will close and
your window reappears.
Tip To add the wizard to the Wizards/ActiveX Toolbar, select the
wizard, and then click Add to toolbar. Once you add a wizard to the
Wizards/ActiveX Toolbar, you can select it and paste it into your open
window at any time.
4.
5.
53
Note Pareto charts recalculate control limits for every new sample entered.
54
Chapter 3
2.
3.
Select the SPC Limits Wizard in the display area, and then click OK or
double-click the wizard. The dialog box will close and your window
reappears.
Tip To add the wizard to the Wizards/ActiveX Toolbar, select the
wizard, and then click Add to toolbar. Once you add a wizard to the
Wizards/ActiveX Toolbar, you can select it and paste it into your open
window at any time.
4.
5.
55
2.
Click Remove from toolbar. The Remove Wizard from Toolbar dialog
box appears.
3.
Select the wizard(s) that you want to remove from the toolbar.
4.
Click OK.
56
Chapter 3
Paste the SPC Control Chart wizard into your window, and then doubleclick it. The SPC Chart Display Configuration dialog box appears.
2.
Type the Dataset name in the input box or, click Select SPC Dataset. The
Select a Dataset dialog box appears.
Tip A previously defined dataset name must be entered to configure the
SPC Control Chart.
3.
4.
The Chart Setup group's options are used to set up and help define your
chart as follows:
Show Title
57
Disable AutoScaling
5.
Click Title Fonts, to open the Font dialog box. Select the font, font style
and size for the chart's Title. Click the color box to select the color in the
color palette that you want to use for the Title.
6.
Click Value Fonts, to open the Font dialog box. Select the font, font style
and size for the values displayed on the chart. Click the color box to select
the color in the color palette that you want to use for the values.
7.
The Process Center Chart group's options are used to define and display
a Center Chart. You must select the Show Chart option to configure the
lines you want displayed on your Center Chart. Click each color box to
select the color in the color palette that you want to use for the Spec.
Lines, Control Lines, Zone Lines, Centerline and Target Line.
The Normal, Alarm and ACK sample points can also be set to different
colors. This helps to distinguish the Normal Sample points from the
Alarm and ACK samples.
58
Chapter 3
8.
The Enable Custom Zones, Zones Required 3,4,5 or 6, and Zone Name
options all define the zones used within your Center Chart as follows:
Note In Runtime, right-click the chart, and then select the Zone Center
option. All the chart samples plotted will now appear in the center of your
custom zones.
9.
The Process Width Chart group's options are used to define and display a
Width Chart. You must select the Show Chart option to configure the
lines you want displayed on your Width Chart.
Control Lines and Centerline. Click the color box to select another color
in the color palette.
Note The Normal, Alarm, and ACK sample points can also be set to
different colors. This helps to distinguish the Normal Sample points from
the Alarm and ACK samples. Click the color box to select another color
in the color palette.
10. In the Sample Display Style group's options you can change the sample
style and the point character size used in both your Center Chart and your
Width Chart.
11. Select Disable Manual Input to disable entering manual sample inputs
from the chart in runtime.
Tip Manual samples can also be entered by using InTouch scripting.
12. Select Enable Right Click Menu to enable the right-click sample menu.
This menu includes Acknowledging Alarms, Deleting samples, Modifying
samples, Zone Centering option and Adding and Deleting Special Causes.
13. Select Show Zone Names to display the zone names on your Center
Chart.
14. Select Allow Sample Del/Modify to allow the user to delete and modify
sample values through the right-click menu.
Tip All sample modifications and deletions are logged to a file named
spcxact.log that is stored in your application directory.
15. Click OK to save.
59
Paste the SPC Histogram wizard into your window, and then double-click
it. The SPC Histogram Display Configuration dialog box appears.
2.
Type the Dataset name in the input box or, click Select SPC Dataset. The
Select a Dataset dialog box appears.
Note A previously defined dataset name must be entered to configure the
SPC Histogram.
3.
4.
The Number of Zones group's options are used to configure the number
of displayed zones on the Histogram chart. This allows you to choose a
fixed number of samples. For example, if your sample size was 300, and
your chart was based on sample size, your graph would end up unreadable.
Fixed Number
5.
The Chart Setup group's options are used to set up and help define your
chart as follows:
Show Title
60
Chapter 3
Specification Lines, Control Click the respective color box to choose the
Lines Centerline
color in the color palette that you want to use
in the chart for each item.
Show Normal Curve
6.
Click Title Fonts, to open the Font dialog box. Select the font, font style
and size for the chart's Title. Click the color box to select the color in the
color palette that you want to use for the Title.
7.
Click Value Fonts, to open the Font dialog box. Select the font, font style
and size for the values displayed on the chart. Click the color box to select
the color in the color palette that you want to use for the values.
8.
Click OK to save.
Paste the SPC Pareto wizard into your window, and then double-click it.
The SPC Pareto Display Configuration dialog box appears.
2.
Type the Dataset name in the input box or, click Select SPC Dataset. The
Select a Dataset dialog box appears.
Note A previously defined dataset name must be entered to configure the
SPC Pareto chart.
3.
61
Tip The dialog box will close and the selected Dataset name will
automatically be inserted into the input box.
4.
The Number of Zones group's options are used to configure the number
of displayed zones on the Pareto chart.
Fixed Number
5.
The Chart Setup group's options are used to set up and help define your
chart as follows:
Background Color,
Zone Color
Border Color,
Percentage Line
Show Title
Show Legend
6.
Click Title Fonts, to open the Font dialog box. Select the font, font style
and size for the chart's Title. Click the color box to select the color in the
color palette that you want to use for the Title.
7.
Click Value Fonts, to open the Font dialog box. Select the font, font style
and size for the values displayed on the chart. Click the color box to select
the color in the color palette that you want to use for the values.
8.
Click OK to save.
62
Chapter 3
Paste the SPC Limit wizard into your window, and then double-click it.
The SPC Limit Display Configuration dialog box appears.
2.
Type the Dataset name in the input box or, click Select SPC Dataset. The
Select a Dataset dialog box appears.
Tip A previously defined dataset name must be entered to configure the
SPC Limit wizard.
3.
4.
In the Tags group boxes, type the tagnames that you have defined in your
Tagname Dictionary for the various items or, click Suggest for the wizard
to automatically suggest tagnames for each item.
Note Clicking suggest or having more than one SPC Limits Wizard will
create multiple Access Names.
63
Tip If you double-click on the blank Tags input box, the Tag Browser
appears displaying all tagnames defined for the selected tag source. For
example.
Double-click the tagname that you want to use or, select it, and then click
OK. The tag browser will close and the selected tagname will
automatically be inserted into the selected field.
For more information on using the Tag Browser, see your online InTouch
User's Guide.
5.
Click OK to save.
64
Chapter 3
C H A P T E R
65
This chapter describes the SPC application techniques which allow the
operator to dynamically change datasets, manipulate Control Charts, obtain
detailed sample information, and perform corrective action for a sample.
Contents
Changing Datasets
Manipulating Control Charts
Detailed Sample Information
Changing Datasets
This section describes how to change Indirect Datasets, change products within
a Dataset, and how to create new products in runtime.
2.
Associate this tagname to an InTouch Access Name using SPC for the
application name and a valid configured Indirect Dataset Name for the
topic name.
Note The Dataset Names are Case Sensitive. If the DSN is called
"Indirect", and you enter "indirect" for the Topic in the Access Name,
none of the I/O items for that Access Name will work.
3.
Type the SPC DDE Item Name DatasetName in the tagname's Item input
box.
4.
66
Chapter 4
Indirect_DatasetName = "SPC1";
Or, create the following to allow the user to select the Dataset name. For
example:
Indirect_DatasetName = SPCSelectDataset();
Once the QuickScript executes, the SPC Chart will display the
configuration settings for the specified Dataset name. For example, the
control limits, and the last samples collected. (The last collected sample
will be the last sample displayed in the chart.)
2.
Associate this tagname to an InTouch Access Name using SPC for the
application name and a valid configured Dataset Name for the topic name.
3.
Type the SPC DDE Item Name ProductCollected in the tagname's Item
input box.
4.
Or, create the following to allow the user to select the Product name. For
example:
ProductCollected = SPCSelectProduct("Dataset");
5.
6.
Once this QuickScript executes, the specified Product name (in this case,
Product1) is written to the I/O Message tagname, ProductCollected.
67
2.
Associate this tagname to an InTouch Access Name using SPC for the
application name and a valid configured Dataset Name for the topic name.
3.
Type the SPC DDE Item Name ProductDisplayed in the tagname's Item
input box.
4.
Or create the following to allow the user to select the Product name. For
example:
ProductDisplayed = SPCSelectProduct("Dataset");
5.
6.
Once this QuickScript executes, the specified Product name (in this case,
Product1) is written to the I/O Message tagname, ProductDisplayed. The
SPC Chart will display the current ProductDisplayed's configuration
settings. For example, the control limits, and the last samples collected.
(The last collected sample will be the last sample displayed in the chart.)
2.
3.
You can set the new products, Control Chart, Pareto Chart and Histogram
Title by using these DDE items:
NewProductCtrlTitle
NewProductParetoTitle
NewProductHistTitle
68
Chapter 4
4.
If these are set before you call NewProduct the newly created product
will use the new title names. For example:
NewProductCtrlTitle = "Product2";
NewProductParetoTitle= "Product2";
NewProductHistTitle= "Product2";
NewProduct= "Product2";
5.
When this script executes, the specified Product name (in this case,
Product2) is written to the I/O Message tagname NewProduct. SPC will
also automatically set the value of the SPC DDE Item ProductCollected
to the value of NewProduct.
Scrolling a Chart
You can create a 3-D button and attach a Touch Pushbutton - Action script to
scroll forward and/or backwards through the currently displayed Dataset's
historical data. This is accomplished by changing the value of the SPC DDE
Item LastSampleDisplayed in the scripts linked to the objects. Another SPC
DDE Item, SamplesPerControlChart can also be used to control how much
data is displayed in a chart.
Tip A Limit Wizard can be used to scroll forward and/or backwards through
the currently displayed Dataset's historical data.
To scroll backwards through the history
1.
Create a graphic object such as a 3-D button, and attach the following
Touch Pushbutton - Action script to it:
LastSampleDisplayed = LastSampleDisplayed SamplesPerControlChart;
2.
In runtime, when the operator clicks the 3-D button, the script will execute
causing SPC to subtract the value of SamplesPerControlChart from the
current LastSampleDisplayed number and automatically scroll the chart
backwards to display the resulting sample number as the last sample in the
chart.
For example, if the current LastSampleDisplayed is number 860 and
SamplesPerControlChart item's value is 20, the chart will scroll
backwards twenty samples resulting in sample number 821 being the first
sample displayed in the chart and sample number 840 being the last
sample displayed.
69
Create a graphic object such as a 3-D button, and attach the following
Touch Pushbutton - Action script to it:
LastSampleDisplayed = LastSampleDisplayed +
SamplesPerControlChart;
2.
In runtime, when the operator clicks the 3-D button, the script will execute
causing SPC to add the value of SamplesPerControlChart to the current
LastSampleDisplayed number and automatically scroll the chart forward
to display the resulting sample number as the last sample in the chart.
3.
Create a graphic object such as a 3-D button, and attach the following
Touch Pushbutton - Action script to it:
LastSampleDisplayed = CurrentSampleNumber;
In runtime, when the operator clicks the 3-D button, the script will execute
causing SPC to set the LastSampleDisplayed item equal to the
CurrentSampleNumber and fill the chart with previous samples
according to the value of the SamplesPerControlChart item's value,
including the CurrentSampleNumber.
3.
70
Chapter 4
Comment
C-Chart
MI_M1
P-Chart
NP-Chart
MI_M1
U-Chart
71
In runtime, click the sample currently displayed (unless your mouse access
is disabled) in the SPC Control Chart. The Sample Information dialog
box appears.
The Sample No. box displays the number for the sample that was clicked
in the SPC Control Chart.
3.
The Date Time box displays the date and time the sample was taken.
4.
5.
The Alarms window displays all alarm conditions for the displayed
sample.
6.
The Measurements list displays the actual values of all the measurements
used in the calculation of the sample.
7.
8.
9.
Click the Special Causes arrow to select a Special Cause for the sample.
72
Chapter 4
10. Select Flag Sample if you want to flag the sample on the chart.
11. Select Ignore Value to force the SPC Control Chart to be redrawn,
ignoring the selected sample from the auto-scaling calculation.
Tip The sample will still be plotted but appears to be off the chart display.
The sample value will also be ignored in the Histogram plot.
Note Selecting this option will not exclude the point during control limit
calculation, just in the SPC Control Chart display.
12. Click New to access the Manual Input dialog box to manually enter
measurements for the sample.
For more information on this dialog box, see "Manual Input Information."
13. Click Corrective Action to access the Corrective Action dialog box to
take corrective action on the sample.
For more information on this dialog box, see "Performing Corrective
Action on a Sample."
14. Click OK.
73
2.
3.
In the Measurements box, type the value of each measurement taken for
the sample, and then press the ENTER key. (The value is entered into the
box below the input box.)
Tip The number of required measurements is specified above the input
box. You cannot click OK to save the input until all the expected number
of measurements are entered. The dialog box only accepts the specified
number of measurements for the sample.
Note The ManualInputDialog will write the current sample number to
LastSampleDisplayed.
4.
By default, the Date and Time boxes, will display the current date and
time that will be associated with this new sample. To change the date or
time, type the new information into the respective input boxes.
5.
If required, click the Special Cause arrow to select a Special Cause to the
sample. Otherwise, the default of None will be used.
6.
7.
In the Sample Comment box, type any comments relating to the sample
(maximum of 50 characters).
8.
Click OK to add the measurements to the Dataset and close the dialog
box. The Sample Information dialog box reappears.
9.
Click OK.
74
Chapter 4
Note This value is in milliseconds so 1500 = 1.5 seconds and 1000 = 1.0
seconds. This should be done before starting WindowViewer.
75
Click the alarmed sample in the SPC Control Chart. The Sample
Information dialog box appears.
76
Chapter 4
2.
3.
In the Corrective Action Notes window, type the notes regarding the
corrective action taken for the sample.
Tip The Other Corrective Action window will list all corrective actions
taken for samples in the SPC Dataset linked to the SPC Control Chart. If
you select a listed correction action, its respective notes will be displayed
in the lower window.
You can scroll through the text. You can also select the text, press the
Ctrl+C keys to copy it, and then press the Ctrl+V keys to paste the text in
the Correction Action Notes window. You can then modify the text as
required for your corrective action.
4.
Click OK. A message box appears asking if you are sure you want to
perform the corrective action for the sample. Click Yes to insert the
corrective action into the SPC Database, or click No to cancel the action.
The Corrective Action dialog box reappears.
5.
6.
Click OK.
Tip The sample will be marked with the symbol [c] in the SPC Control Chart
to indicate that corrective action has been taken for the sample and the SPC
alarm counters will reset, according to the switch setting in the SPC.INI file.
Taking a Corrective Action on the last sample (CurrentSampleNumber) will
also reset the SPCPro Run Rule Counts.
77
Note When a corrective action is taken the SPC Alarm counters will
automatically be reset to zero. This reset is controlled through a switch setting
in the SPC.INI file. The default setting is to reset only those alarms that exist for
the sample for which the correction action was performed. However, you can
change this to resetting all SPC alarm counters by including the following line
in your SPC.INI file:
[General]
ResetAllAlarmCounters=1
To Modify a Sample, right click on the point that you want to modify.
78
Chapter 4
2.
3.
The Sample Information dialog box appears with the Modify button
available.
4.
5.
6.
To delete a sample
1.
To Delete a Sample, right click on the point that you want to delete.
2.
Select the Delete Sample option. The following message box will appear.
3.
79
80
Chapter 4
C H A P T E R
81
Contents
Using SPC DDE Items
SPC Functions
82
Chapter 5
With the addition of displayed and collected products, many SPC DDE Items
apply only to the collected product. These items are flagged in the following
lists by an asterisk (*) preceding the SPC DDE Item name.
Item Name
DDE Type
Access
Description
AutoCollection
Discrete
R/W
*CalculateControlLimits
Discrete
R/W
DatasetName
Message(32)
R/W
HistogramLCL
Real
RO
HistogramUCL
Real
RO
Kurtosis
Real
RO
LastSampleDisplayed
Integer
R/W
*ManualInputDialog
Discrete
R/W
MeasurementsPerSample Integer
RO
NewProduct
Message
(32)
R/W
NewProductCtrlTitle
Message
(32)
R/W
NewProductParetoTitle
Message
(32)
R/W
NewProductHistTitle
Message
(32)
R/W
*ProductCollected
Message
(32)
R/W
ProductDisplayed
Message
(32)
R/W
SampleSize
Integer
RO
SamplesControlChart
Integer
R/W
SamplesPerHistogram
Integer
R/W
SamplesPerLimitCalc
Integer
R/W
SamplesPerPareto
Integer
R/W
83
Item Name
DDE Type
Access
Description
SelSPCOutSpecMsg
Message
RO
Skewness
Real
RO
SPCAllowSampDelMod
Discrete
R/W
SPCConnection
Discrete
RO
SPCConnectType
Message
RO
SPCLowDBSpace
Discrete
RO
SPCResetRunRules
Discrete
R/W
StartCollection
Discrete
R/W
84
Chapter 5
For distributed SPC, initially all the values are set to zero. SPC connects to the
database and checks for new data every 5 seconds. The item values are updated
whenever new information is found. Modifications to the current sample
values are buffered locally until the CurrentUpdate item is set to 1. Then the
values are placed in a current sample packet and sent to the remote dataset
node for analysis and storage. Current sample modifications that indicate a
different collected product and a current sample number that is not the last
recorded sample will be rejected by the Server.
With the addition of displayed and collected products, all of the "Current" SPC
DDE Items apply only to the collected product.
Item Name
DDE Type
Access
Description
CurrentCauseCode
Integer
R/W
CurrentCauseString
Message
(128)
RO
CurrentComment
Message
(50)
R/W
CurrentCp
Real
RO
CurrentCpk
Real
RO
CurrentDate
Message
(10)
R/W
CurrentFlag
Discrete
R/W
CurrentIgnoreValue
Discrete
R/W
CurrentMx
Real
R/W
CurrentR
Real
RO
CurrentRBar
Real
R/W
CurrentRLCL
Real
R/W
CurrentRUCL
Real
R/W
85
Item Name
DDE Type
Access
Description
CurrentSample
Real
RO
CurrentSampleBar
Real
R/W
CurrentSampleNumber
Integer
RO
CurrentTarget
Real
R/W
CurrentTime
Message (8)
R/W
CurrentUpdate
Discrete
R/W
CurrentXLCL
Real
R/W
CurrentXLSL
Real
R/W
CurrentXUCL
Real
R/W
CurrentXUSL
Real
R/W
SPC2L3Out2SD
Integer
RO
SPC2L3Out2SDMsg
Message
RO
SPC4L5Out1SD
Integer
RO
SPC4L5Out1SDMsg
Message
RO
SPCConSampAltUpDn
Integer
RO
SPCConSampAltUpDnMsg Message
RO
SPCConSampIn1SD
RO
Integer
86
Chapter 5
Item Name
DDE Type
Access
Description
SPCConSampIn1SDMsg
Message
RO
SPCConSampIncDec
Integer
RO
SPCConSampIncDecMsg
Message
RO
SPCConSampOneSideCL
Integer
RO
SPCConSampOneSideCL
Msg
Message
RO
SPCConSampOut1SD
Integer
RO
SPCConSampOut1SDMsg
Message
RO
SPCNLNOutNSD
Integer
RO
SPCNLNOutNSDMsg
Message
RO
SPCNLNOutNSDSS
Integer
RO
SPCNLNOutNSDSSMsg
Message
RO
SPCOutRCtrl
Integer
RO
SPCOutRCtrlMsg
Message
RO
SPCOutXCtrl
Integer
RO
SPCOutXCtrlMSG
Message
RO
87
Item Name
DDE Type
Access
Description
SPCOutSpec
Integer
RO
SPCOutSpecMsg
Message
RO
SPCRecalculateCp
Discrete
R/W
SPCResetAlarmCounters
Discrete
R/W
SPCResetRunRules
Discrete
R/W
88
Chapter 5
With the addition of displayed and collected products, all of the "Manual" SPC
DDE Items apply only to the collected product.
Item Name
DDE Type
Access
Description
MI_CauseCode
Integer
WO
MI_CauseString
Message
(127)
RO
MI_Comment
Message
(50)
WO
MI_Date
Message
(10)
WO
MI_Flag
Discrete
WO
MI_IgnoreValue
Discrete
WO
MI_Mx
Real
WO
MI_Save
Discrete
WO
MI_Time
Message (8)
WO
89
For distributed SPC, selected sample DDE items are private to each node.
They are the sample values recorded by the remote node for a specified sample
number of the collected product. When the Selection DDE item is set to a
sample number, the sample information is retrieved from the remote node's
sample file. Old data cannot be changed, but Special Cause Codes, Flags, and
Comments can be added by changing the appropriate DDE item and setting the
SelectionUpdate item to 1. When SelectionUpdate is set to 1, the Special
Cause Code, Comment, Flag, and Ignore Value items are sent to the remote
node in a packet for storage.
Note With the addition of displayed and collected products, all of the
"Selection" SPC DDE Items apply only to the collected product.
Item Name
DDE Type
Access
Description
Selection
Integer
R/W
SelectionCauseCode
Integer
R/W
SelectionCauseString
Message
(128)
RO
SelectionComment
Message
(50)
R/W
SelectionCp
Real
RO
SelectionCpk
Real
RO
SelectionDate
Message
(10)
RO
SelectionFlag
Discrete
R/W
SelectionIgnoreValue
Discrete
R/W
SelectionMx
Real
RO
SelectionProduct
Message
(32)
RO
SelectionRUCL
Real
RO
SelectionRLCL
Real
RO
SelectionR
Real
RO
SelectionRBAR
Real
RO
90
Chapter 5
Item Name
DDE Type
Access
Description
SelectionSample
Real
RO
SelectionSampleBar
Real
RO
SelectionTarget
Real
RO
SelectionTime
Message (8) RO
SelectionUpdate
Discrete
R/W
SelectionXUSL
Real
RO
SelectionXLSL
Real
RO
SelectionXUCL
Real
RO
SelectionXLCL
Real
RO
SelSPC2L3Out2SDMsg
Message
RO
SelSPC4L5Out1SDMsg
Message
RO
SelSPCConSampAltUpDnMsg
Integer
RO
SelSPCConSampIn1SDMsg
Message
RO
SelSPCConSampIncDecMsg
Message
RO
SelSPCConSampOneSideCLMsg
Message
RO
SelSPCConSampOut1SDMsg
Message
RO
SelSPCNLNOutNSDMsg
Message
RO
SelSPCNLNOurNSDSSMsg
Message
RO
SelSPCOutRCtrlMsg
Message
RO
SelSPCOutXCtrlMsg
Message
RO
SelSPCOutSpecMsg
Message
RO
91
Multiple Indirect Datasets can be setup and linked to the same real Dataset.
Then the Selection value of each Indirect Dataset can be set to a different
sample number. This allows you to view detail information of multiple samples
within a Dataset.
SPC Functions
You can control the SPC program from WindowViewer by using the SPC
Functions described below in InTouch scripts.
92
Chapter 5
For detailed information on all SPC Functions, see your online InTouch
Reference Guide.
Function
Description
SPCConnect
SPCDatasetDlg
SPCDisconnect
SPCDisplayData
SPCLocateScooter
SPCMoveScooter
93
Function
Description
SPCSaveSample
SPCSelectDataset
SPCSelectProduct
SPCSetControlLimits
SPCSetMeasurement
SPCSetProductCollected
94
Chapter 5
Function
Description
SPCSetProductDisplayed
SPCSetRangeLimits
SPCSetSpecLimits
C H A P T E R
95
This chapter describes how to use the Wonderware supplied SPCPro utility.
(spcutil.exe). This utility can be found in your InTouch install directory. It used
to upgrade your SPCPro database to a newer version when upgrading to a
newer release of SPCPro. It also provides a utility to manage your Microsoft
Access database. Microsoft Access databases grow in size rapidly and
should be compacted regularly. You will see a significant reduction in the
Microsoft Access database *.mdb file and your hard disk usage after
compacting it.
Contents
Upgrading Your SPCPro Database
Managing Your Microsoft Access Database
2.
96
Chapter 6
3.
On the File menu, select Convert Schema. The Select Data Source
dialog appears.
4.
In DSN Name box, type the name that points to the database you want to
convert or, click the Machine Data Source tab and select it from the
defined data sources on your computer.
5.
After selecting your data source name, click OK. Your database will be
converted to the latest version.
Tip A backup file will be saved in the database directory before
converting. After you have verified that the new database is working
correctly, you can delete the database backup file.
2.
3.
4.
5.
Select the Database location and file name, then click Open.
6.
97
Tip A backup copy of the old database is saved before the compaction
starts. The backup file can be found in the database directory. After you
verify that the compacted database is OK, you can delete the backup
database file to restore hard disk space.
98
Chapter 6
Technical References
C H A P T E R
99
Technical References
This chapter describes the formulas that each type of SPC chart uses to
perform its calculations. The iterative process for calculating control limits, a
Bibliography and Glossary of Terms are also included.
Contents
SPC Calculations
Iterative Calculation Method
Bibliography
Glossary of Terms
SPC Calculations
The SPC calculations are performed in the following manner.
X Individual
Xbar (Centerline) = X / N
where: X = Individual Sample Values
N = Number of samples used in the control limit calculation
(configured in the Dataset definition)
Control Limits
Xbar +/- 3s
where: s = Standard Deviation = SQRT( ( X2 - N*Xbar2) / (N -1) )
Xbar - s
Xbar = X / n
where: X = Individual Measurement Values
n = Number of measurements used for the sample
(configured in the Dataset definition)
100
Chapter 7
Xbarbar(Centerline) = Xbar / N
where: Xbar = Individual Sample Values
N = Number of samples used in the control limit calculation
(configured in the Dataset definition)
s = Standard Deviation = SQRT(( X2 - n*Xbar2 ) / (n -1) )
sbar = s / N
Control Limits (XUCL, XLCL)
Xbarbar +/- A3 * sbar
where: A3 = statistical constant based on subgroup size
Control Limits (sUCL, sLCL)
sUCL = B4* sbar
where: B4 = statistical constant based on subgroup size
sLCL = B3* sbar
where: B3 = statistical constant based on subgroup size
Technical References
101
Note On Xmoving - Rmoving charts any samples that do not contain the
correct amount of measurements are eliminated from the calculation. For
example - if an Xmoving - Rmoving dataset is configured to use 2 samples
then the first collected sample in the dataset will not be used in any
calculations.
C chart
Cbar(Centerline) = C / N
where: C = Individual counts
N = Number of samples used in the control limit calculation
(configured in the Dataset definition)
Control Limits
CUCL = Cbar + 3 * SQRT(Cbar)
CLCL = Cbar - 3 * SQRT(Cbar)
P chart
Pbar(Centerline) = NP/ Nn
where: P = Individual % Defective samples
N = Number of samples used in the control limit calculation
(configured in the Dataset definition)
NP = Total number of rejects (not defects)
n = Sample size
nBAR = n/N
Control Limits
PUCL = Pbar + 3 * SQRT(Pbar * ( 1-Pbar) / nBAR)
PLCL = Pbar - 3 * SQRT(Pbar * ( 1-Pbar) / nBAR)
NP chart
NPbar = NP / N
Pbar = NPbar / n
The NP chart is similar to the C chart.
where: n = sample size used in the calculation
(configured in the Dataset definition)
NP = total number of rejects (not defects)
N = the number of samples used in Control Limit calculation
Control Limits
UCL = NPbar + 3 * SQRT(NPbar * ( 1-Pbar) )
102
Chapter 7
U chart
Ubar = (Ci) / (ni)
Nbar = Average of ni
The U chart is similar to the P chart.
where: ni = sample size
Ci = total number of defects (not rejects)
Control Limits
UCL = Ubar + 3 * SQRT(Ubar / Nbar)
LCL = Ubar - 3 * SQRT(Ubar / Nbar)
EWMA chart
The inputs for configuring the dataset are the Smoothing Factor and the Tighter
Controls.
The input for the chart is Xi values. For each Xi value entered, the EWMA
Points are calculated and plotted on the screen.
EWMA ( i ) = ( Smoothing Factor ) Xi + ( 1 - Smoothing Factor ) EWMA (
i-1)
Control Limits
UCL = Xbar + 3 s * SQRT(Smoothing Factor / (2 - Smoothing Factor))
LCL = Xbar - 3 s * SQRT(Smoothing Factor / (2 - Smoothing Factor))
If Tighter Control is checked in the dataset configuration, the UCL and LCL
values are calculated with 2.58 instead of 3.0
where: Smoothing Factor / Weighting Factor = alpha symbol
Xbar = Average of Xi values
s = Standard Deviation
Also, you can input the samples by MI_M1 and MI_Save.
Your input Xi value is saved as an Input Sample in the database.
CuSum chart
There is no specific input required for configuring the datasets. There is
Control limit calculations for this chart.
The input for the chart is Xi value. The calculated CuSum Score is plotted on
the screen.
where: CuSum(i) = CuSum(I-1) + (Xi-Target) and CuSum (0) = 0
X = individual sample value.
Technical References
103
Note In the sample input, you can enter Multiple Xi's, not the measurements.
In CuSum chart, Measurement / Sample = 1.
Also, you can input the samples by MI_M1 and MI_Save.
Your input Xi value is saved as an Input Sample in the database.
Histogram
Mean = X / n*N
where: X = Individual Measurement Values
n = Number of measurements/sample
N = Number of samples used in the display
Control Limits
UCL = Mean + 3s
LCL = Mean - 3s
where: s = Standard Deviation = SQRT(( nt* X2 - ( X )2) / nt*(nt -1) )
where: nt = n*N
Skewness = m3 / (m2) 3/2
where: m3 and m2 are the third and second moments about the mean of the
sample.
Kurtosis = m4 / (m2) 2
where: m4 and m2 are the fourth and second moments about the mean of the
sample.
Capability
Cp = ( USL - LSL ) / 6s
where: s = Standard Deviation = SQRT( ( n* X2 - ( X )2 ) / n*(n -1) )
Cpk = (minimum of ( USL - Mean) OR ( Mean - LSL ) ) / 3s
After the 24th sample the SPC program will query the database for the
records of the last 24 samples.
104
Chapter 7
2.
3.
4.
5.
If any samples fall outside the control limits the SPC program will
eliminate the worst offending sample and repeat steps 1 through 4. This
process will continue until all samples fall within the control limits.
6.
7.
8.
Individual samples are compared against the control limits. If any samples
fall outside the control limits the SPC program will eliminate the worst
offending sample and repeat steps 1 through 6. This process will continue
until all samples fall within the control limits.
Bibliography
DataMyte Corporation, DataMyte Handbook, (DataMyte Corporation,
Minnetonka, Minnesota 1989)
Duncan, Acheson J. Quality Control and Industrial Statistics (Irwin,
Homewood, Illinois 1986)
Evans, James R. A Statistical Process Control for Quality Improvement, "A
Training Guide to Learning SPC," (Prentice Hall, Englewood Cliffs, New
Jersey 1991)
Holmes, Donald Introduction to SPC, (Copley Publishing Group, Littleton,
Massachusetts 1988)
Ishikawa, Kaoru Guide to Quality Control, (Quality Resources, White Plains,
New York 1990)
Montgomery, Douglas C. Introduction to Statistical Quality Control (John
Wiley & Sons 1985)
Ott, Ellis R. and Schilling, Edward G. Process Quality Control
"Troubleshooting and Interpretation of Data," (McGraw-Hill Publishing
Company 1990)
Wheeler, Donald J. and Chambers, David S. Understanding Statistical Process
Control - Second Edition, (SPC Press, Knoxville, Tennessee 1992)
Technical References
105
Glossary of Terms
Assignable Causes
Bell-shaped Curve
c Chart
Centerline
Common Causes
Control Chart
Control limits
106
Chapter 7
Cp
Cpk
CuSum Chart
Cycles
Defective Unit
EWMA Chart
Frequency distribution
Histogram
Technical References
107
Kurtosis
LCL
See LCL.
See LSL.
LSL
Mean
Moving X, Moving R
108
Chapter 7
Moving Range
Normal Distribution
Normal Curve
NP Chart
Out of Control
Pareto Analysis
Pareto Chart
Percent Defective
Technical References
109
Process Capability
Range
RChart
Sample
Sample Size
Skewness
Standard Deviation
Statistical Control(of a
process)
Subgroups
Tolerance
Trends
110
Chapter 7
U Chart
UCL
See UCL.
See USL.
USL
Variables
X and R Charts
X Individual Charts
Xbar Charts
Zone Centering
Zones
111
Index
A
Alarms 37
Integrated Alarming 13
Monitoring the Alarm State of an SPC
Tagname 38
Analysis Types 12
C charts 12
CUSUM charts 13
EWMA charts 13
Moving X, Moving R charts 12
NP charts 13, 108
P charts 12
U charts 13
X bar, R chart 12
X bar, s charts 12
X Individual 12
Application Techniques
Changing Collected Products within a Dataset 66
Creating New Product During Runtime 67
C
Calculations
C chart 101
Capability 103
CuSum chart 102
EWMA chart 102
Histogram 103
Iterative Calculation Method 103
NP chart 101
P chart 101
U chart 102
X Individual 99
Xbar - R, Xmoving - Rmoving 100
Xbar - s 99
Causes 41
Changing
Collected Products within a Dataset 66
Datasets 65
Default Corrective Action Name 77
Displayed Products within a Dataset 66
Indirect Datasets 65
Compacting your database 96
Configuring
Indirect Datasets 42
ODBC Database 19
SPC Control Chart Wizard 55
SPC Database Users 27
SPC Datasets 29
SPC Histogram Wizard 59
SPC Limit Wizard 61
SPC Pareto Chart Wizard 60
Configuring Datasets
Alarms 37
Causes 41
Products 35
Control Chart Manipulation 68
Filling a Chart with Current Sample Data 69
Scrolling a Chart 68
Corrective Action 72
D
Data Entry Schemes 13
Automatic Control Limit Calculation 14
Automatic Time Based Collection 14
Detailed Sample Information 15
Event Driven Automatic Collection 14
Flagged Samples 15
Historical Data Review 15
Manual Data Collection 14
Multiple Product or Batches per Dataset 14
Special Cause/Comment Entry 15
Text and Symbols associated with Sample
Number 15
Database Configuration 19
Distributed 23
Single Node 19
Deleting Sample Information 78
Disabling Manual Input on Control Charts 58
Distributed SPC 13
E
Enabling Right Click Menu 58
Entering Corrective Action information 74
F
Functions 91
SPCConnect 92
SPCDisconnect 92
SPCDisplayData 92
SPCLocateScooter 92
SPCMoveScooter 92
SPCSaveSample 93
SPCSelectDataset 93
SPCSelectProduct 93
SPCSetControlLimits 93
SPCSetMeasurement 93
SPCSetProductCollected 93
SPCSetProductDisplayed 94
SPCSetRangeLimits 94
SPCSetSpecLimits 94
H
Histogram Configuration 59
Chart Setup 59
Number of Zones 59
Select SPC Dataset 59
I
Importing SPC Datasets 43
Indirect Datasets 42
Installation 7
Installing
112
SPC Wizards 50
Item Names
SPC Control and Display DDE Items 81
SPC Current Sample DDE Items 83
SPC Manual Input DDE Items 87
SPC Selection DDE Items 88
L
Limit Wizard 61
Select SPC Dataset 62
Tags 62
M
Managing your Microsoft Access database 96
Manipulating Control Charts 68
Manual Input
Measurements 73
Manual Input Information 72
Sample Information 73
Microsoft Access database 19
Microsoft Access database configuration 19
Microsoft SQL Server database 23
Microsoft SQL Server database configuration 23
Modifying and Deleting Sample Information 77
Modifying Sample Information 78
P
Pareto Chart Configuration 60
Chart Setup 61
Number of Zones 61
Select SPC Dataset 60
Performing Corrective Action on a Sample 74
Products 35
Remote datasets 13
Remove Wizard from Toolbar Dialog Box 55
Tag Browser 63
Tagname Dictionary 31
Technical References 99
The Basic's of SPC 9
Total Quality Management 10
Sample Information 70
Corrective Action 72
Manual Input 72
Special Causes 71
SPC Application Techniques 65
SPC Calculations 99
SPC Chart Display Configuration 55
Allow Sample Del/Modify 58
Centerline 58
Chart Setup 56
Control Lines 58
Disable AutoScaling 57
Disable Manual Input 58
Enable Custom Zones 58
Enable Right Click Menu 58
Process Width Chart 58
Sample Display Style 58
V
Verifying the database connection 23
W
WindowMaker Application Explorer 8
Wizard Tool
Removing Wizards from the Toolbox 54
Wizard/ActiveX Installation 50
113
114
Index