home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Explore the World of Soft…e: Engineering & Science
/
Explore_the_World_of_Software_Engineering_and_Science_HRS_Software_1998.iso
/
programs
/
statistc
/
spcex21s.exe
/
INFO.HYP
< prev
next >
Wrap
Text File
|
1993-06-10
|
43KB
|
1,253 lines
@
MORE_INFO_MENU
ON LINE DOCUMENTATION
#m>HELP<;hypertext help#m - Click on the >HELP< or Press Enter to learn
how to operate the hypertext help system.
#m(X.Y.Z);MENU_NOTATION;screens.hyp#m SPC EXpert menu navigation notation.
#mA. SPC EXpert Licensing Information;system0#m
#mB. SPC EXpert Custom Enhancement Services;custom#m
#mC. Technical/Customer Support;support#m
#mD. Addressing email feedback;email#m
#mE. Planned Enhancements to SPC EXpert;enhance#m
#mF. SPC Abbreviations;SPC Abbreviations;define.hyp#m
#mG. SPC Definitions;SPC Definitions;define.hyp#m
#mH. SPC EXpert system overview;system1#m
#mI. Data File Formats;format1#m
#mJ. Graphs/Diagrams Overview;graph0#m
#mK. Control Charts Overview;charts0#m
#mL. On line SPC EXpert User's Manual;users manual;manual.hyp#m
@
email
Sending Electronic Mail
You are strongly encouraged to send your comments or questions via
electronic mail since telephone support will only be available from
6:00pm-9:00pm Eastern time.
EMAIL : marks@meaddata.com
If this does not work, you may need to try the alternative email
addressing format as follows:
...!uunet!meaddata!marks
That is, once you can get your message to UUNET, it can get to us.
Now, if your email system "knows" UUNET, then the following may work:
uunet!meaddata!marks
However, if this does not work, you may need to add another system name
to this email address. For example, if your full email address is
myname@abc.def.GOV, then you might try using the following email address:
gov!uunet!meaddata!marks
Other possibilities include:
mil!uunet!meaddata!marks
edu!uunet!meaddata!marks
net!uunet!meaddata!marks
@
system0
SPC EXpert - Copyright Quality Software Designs, Inc. 1993
SPC EXpert is a SHAREWARE product. You may freely reproduce and
re-distribute SPC EXpert for evaluation, educational, and
governmental purposes. If you make commercial use of this software,
you are expected to pay a licensing fee as outlined below. Upon
receipt of your licensing fee you will receive a printed manual, a
license agreement, and notices of future upgrades to the software.
*Academic and federal,state & local government agencies may use and
re-distribute SPC EXpert without charge (FREE) but should register.
Individual User($40);PC LAN($100);Site License($200);*Academic(FREE)
Mutli-Site License($500); *Federal, state, & local Government(FREE)
** Send Check or money order to Quality Software Designs Inc
933 East Stroop Road
phone: (513) 299-3023 Kettering OH 45429 USA
@
custom
SPC EXpert Custom Enhancement Services
Quality Software Designs, Inc., can provide you with a custom enhanced
version of SPC EXpert tailored to your specific needs. In many cases,
a tailor-made version of SPC EXpert can be produced at a cost comparable
to the single-copy off-the-shelf cost of commercial SPC software products.
To inquire about this service, contact:
Quality Software Designs, Inc.
933 East Stroop Road
Kettering, Ohio 45429
(513) 299-3023
email: marks@meaddata.com or
...!uunet!meaddata!marks
Possible areas for customization include,
1. Placing domain-specific advice in the expert consultation text.
e.g. In place of
"A shift-up in the R chart means that your
process has become less consistent..."
we display
"A shift-up in the R chart has historically been
caused (90%) by a loose ball bearing in the
preprocessing machine."
2. Placing company-specific terminology and text in the help screens
and expert consultation text.
3. Translating SPC EXpert into another non-English language.
4. Adding a unique type of statistical/graphical tool.
Remember, Quality Software Designs, Inc., is in the business of
providing Custom Software Solutions. When you consider which
SPC software package to buy, call up the commercial vendors and
ask then if their $200-$1000 per copy price-tag includes such
customer-driven tailoring.
@
support
SPC EXpert Technical/Customer Support
1. Phone support is available 6:00pm-9:00pm eastern time (513) 299-3023
2. Messages can be left on an answering machine which is regularly
checked all day, every day. (513) 299-3023
3. Questions and detailed problem reports can be sent over the INTERNET
(email) to marks@meaddata.com or ...!uunet!meaddata!marks
4. Send written comments and questions to
Quality Software Designs, Inc.
933 East Stroop Road
Kettering, Ohio 45429
Suggestions are strongly encouraged. Without your help and feedback,
we cannot improve SPC EXpert to better meet your (changing) needs.
@
enhance
Planned Enhancements To SPC EXpert
SHORT TERM
S.1 Include Short-Run Control Charts
a. Nominal Xbar-R
b. Target Xbar-R
c. Short Run X-Bar-R
d. Short Run XMR
e. Short Run Moving Average Moving Range
f. Short Run C
g. Short Run U
h. Short Run NP
i. Short Run P
j. Standardized S Chart
S.2 Include Variable Sub-group Size Xbar-R & Xbar-S Charts
S.3 "Multiple" Versions of existing tools
a. multi bar graph
b. stacked bar graph
c. stacked pareto diagram
d. multi line (run chart) graph
MID TERM
M.1 MicroSoft Windows Based Version
M.2 Import Data from Properly Structured DBase-Type FIles
- The database structure for each type of
control chart or graphic will be specifically
given in the documentation.
LONG TERM
L.1 "Real-Time" Data Analysis and Monitoring
L.2 Macintosh Version
L.3 MicroSoft Windows Based Graphical TQM Tools
a. Flow-Charting
b. Fish-bone Diagrams
c. Cause & Effect Diagrams
d. Organizational Charts
@
format1
SPC EXpert
DATA FILE FORMATS
#mGeneral;format2#m
#mRun Charts;format3#m
#mMR Charts;format3#m
#mXMR Charts;format3#m
#mXBAR Charts;format6#m
#mR Charts;format6#m
#mS Charts;format6#m
#mXBAR-R Charts;format6#m
#mXBAR-S Charts;format6#m
#mPN Charts;format9#m
#mP Charts;format11#m
#mC Charts;format13#m
#mU Charts;format15#m
#mBar Graphs;format17#m
#mPareto Diagrams;format17#m
#mPie Charts;format17#m
#mTables;format17#m
#mHistograms;format3#m
#mFrequency Polygons;format3#m
#mOgives;format3#m
#mScatter Diagrams;format26#m
#mMultiple Charts;format28#m
@
format2
DATA FILE FORMATS
1. SPCEX will generate 11 different types of control charts
- RUN,XBAR,R,S,MR,XBAR_R,XBAR_S,XMR,PN,P,C,U.
2. SPCEX will generate 8 different types of diagrams/graphs
- BAR, PARETO, PIE, data TABLE, HISTOGRAM Frequency
POLYGON, Ogive (CDF) and SCATTER diagram.
3. Each type of chart/diagram requires a slightly different
format for the ASCII file which contains the process data.
4. The #mfirst 4 lines#m of every data file contain titles & axis
labels
5. For all types of charts, entries in the data file must be
separated by RETURNS.
* DO NOT SEPARATE ENTRIES WITH COMMAS, ETC. !!!
@
first 4 lines
DATA FILE FORMATS
FIRST 4 LINES
The first 4 lines of every data file regardless of type
will always contain the same type of information:
LINE 1 : A TITLE for the graph/diagram
LINE 2 : A SUBTITLE for the graph/diagram
LINE 3 : The X-axis (horizontal axis) label
LINE 4 : The Y-axis (vertical axis) label
@
format3
RUN, MR, XMR, HISTOGRAM, POLYGON, CDF
Data File Format
1. The #mfirst 4 lines#m contain titles & axis labels
2. All entries consist of a numeric value separated by spaces
followed by an optional string of characters and a carriage
RETURN. Only one numerical entry per line is allowed.
3. The length of a RUN, MR, or XMR control chart will be
the number of entries.
4. SPC EXpert limits the number of entries to 16000 or less.
5. Sample data files are located on your SPCEX System Disk in
the following files:
"#trun.dmo#t"
#u<RUN graphic>;RUN run.dmo default.cfg#u
#u<MR graphic>;MR run.dmo default.cfg#u
#u<XMR graphic>;XMR run.dmo default.cfg#u
#u<HISTOGRAM graphic>;HISTOGRAM run.dmo default.cfg#u
#u<POLYGON graphic>;POLYGON run.dmo default.cfg#u
#u<CDF graphic>;CDF run.dmo default.cfg#u
@
format6
XBAR, R, S, XBAR_R, XBAR_S
Data File Formats
1. The #mfirst 4 lines#m contain titles & axis labels
2. The fifth line is an INTEGER between 2 and 200 inclusive.
This number represents the logical group size or sample size.
3. The remaining entries (REAL or INTEGER) represent the actual
process measurements.
4. The remaining lines may be coded one of two ways.
(a) If the logical group size is small enough, all entries
for each group may be placed on a single line separated by spaces.
For example, the data file below depicts measurements of each of
five items for each sample number.
Pitch Diameter
Units of .0001 inch in excess of .4000 inch
Sample Number
Measurement
5
36 35 34 33 32
31 31 34 32 30
30 30 32 30 32
32 33 33 32 35
(b) The second way to create the data file is to place each measurement
on a single line. The data file below will produce the exact same
results as the one above. It is particularly useful when the sample
size is too large to it all sample measurements on a single line.
Pitch Diameter
Units of .0001 inch in excess of .4000 inch
Sample Number
Measurement
5
36
35
34
33
32
31
31
34
32
30
30
30
32
30
32
32
33
33
32
35
4. The length of the chart(s) will be the number of process
measurements divided by the logical group size. For example,
if we have a group size of 3 and 69 process measurements, then
the length of the XBAR and R Charts will be 23.
5. Sample XBAR-R data files are located on your SPCEX System Disk
in files:
"#tr.dmo#t"
#u<XBAR graphic>;XBAR r.dmo default.cfg#u
#u<R graphic>;R r.dmo default.cfg#u
#u<S graphic>;S r.dmo default.cfg#u
#u<XBAR_R graphic>;XBAR_R r.dmo default.cfg#u
#u<XBAR_S graphic>;XBAR_S r.dmo default.cfg#u
"#ts.dmo#t"
#u<XBAR graphic>;XBAR s.dmo default.cfg#u
#u<R graphic>;R s.dmo default.cfg#u
#u<S graphic>;S s.dmo default.cfg#u
#u<XBAR_R graphic>;XBAR_R s.dmo default.cfg#u
#u<XBAR_S graphic>;XBAR_S s.dmo default.cfg#u
@
format9
PN CHART
DATA FILE FORMAT
1. The #mfirst 4 lines#m contain titles & axis labels
2. All numeric entries in a PN Chart data file must be INTEGERS.
3. The fifth line contains an integer which represents the
fixed sample size.
4. The remaining numeric entries represent the number of defectives
(rejects/failures) out of a fixed number of samples. The length
of the PN chart will be one less than the number of entries in
the data file. A sample PN data file is shown below:
PN Control Chart
Lot Size = 200
Production Order Number
Number of Rejects
200
23
15
17
15
41
0
25
31
29
0
8
16
Only one entry per line is allowed.
5. A sample PN data file is located in the file
"#tpn.dmo#t"
#u<PN graphic>;PN pn.dmo default.cfg#u.
@
format11
P CHART
DATA FILE FORMAT
1. The #mfirst 4 lines#m contain titles & axis labels
2. All remaining entries in a P Chart must be PAIRS of INTEGERS!
Only one pair per line is allowed. Each pair may be followed
by an optional string of characters.
3. The first integer in each pair represents the sample size.
4. The second integer represents the number of defectives or
rejects.
5. The length of the P Chart will be the number of pairs of
entries.
A sample P data file is shown below:
P Chart
Paragraph 5.6 Data
Day
Percent Rejected
28 4
27 9
28 3
16 2
26 1
24 1
25 5
20 5
25 14
14 1
6. A sample P data file is located in the file
"#tp.dmo#t"
#u<P graphic>;P p.dmo default.cfg#u.
@
format13
C CHART
DATA FILE FORMAT
1. The #mfirst 4 lines#m contain titles & axis labels
2. All entries in a C Chart data file must be INTEGERS.
3. Each entry represents the number of defects (errors) in a
sample of fixed size/weight/length/etc. Only one entry per
line followed by an optional string of characters is allowed
per line.
4. The length of the C chart will be the number of entries
in the data file.
A sample C data file is shown below:
C Chart
Paragraph 5.7 Data
Sample number
Number of Alignment Errors
7
6
6
7
4
7
8
12
9
9
5. A sample C data file is located in the file
"#tc.dmo#t"
#u<C graphic>;C c.dmo default.cfg#u.
@
format15
U Chart
DATA FILE FORMAT
1. The #mfirst 4 lines#m contain titles & axis labels
2. All entries in a U Chart must be PAIRS of entries. Only one
pair per line followed by an optional string of characters
is allowed.
3. The first entry in each pair can be either INTEGER or REAL
and represents the sample size/length/weight/etc.
4. The second entry must be an INTEGER and represents the number
of defects/errors/etc. in the sample.
5. The length of the U Chart will be the number of pairs of
entries.
A sample U data file is shown below:
U Chart
# Typos in Daily Calendar
Day
Typos per Word
150 2
100 1
120 0
140 3
200 6
120 1
6. A sample U data file is located in the file
"#tu.dmo#t"
#u<U graphic>;U u.dmo default.cfg#u.
@
format17
BAR, PARETO, PIE, TABLE
DATA FILE FORMAT
1. The #mfirst 4 lines#m contain titles & axis labels
2. The numeric data for a BAR, PARETO, ... Graphs MUST BEGIN
ON THE FIFTH LINE OF THE DATA FILE.
3. The numeric data consists of PAIRS (maximum 100) of entries.
4. The first entry can be any NON-NEGATIVE number.
5. The second entry can be anything - this will be the category.
#mSample File;format18#m
6. A sample BAR graph data file is located on your SPCEX System
Disk in the file "#tbar.dmo#t". This will produce the
following graphics:
#u<BAR>;BAR bar.dmo default.cfg#u
#u<PARETO>;PARETO bar.dmo default.cfg#u
#u<PIE>;PIE bar.dmo default.cfg#u
#u<TABLE>;TABLE bar.dmo default.cfg#u
@
format18
SAMPLE BAR GRAPH
DATA FILE
Any Title You Want
Any Subtitle you want
X AXIS LABEL
Y AXIS LABEL
25 Type A
15 Type B
2 Type C
55 Type D
@
format26
SCATTER DIAGRAM
DATA FILE FORMAT
1. The #mfirst 4 lines#m contain titles & axis labels
2. The numeric data for a SCATTER Diagram MUST BEGIN ON THE FIFTH
LINE OF THE DATA FILE.
3. The numeric data consists of PAIRS of entries. Both entries in
each pair can be any number, INTEGER or REAL. Only one pair
followed by an optional string of characters is allowed per line.
4. The first number in each pair represents the x-coordinate for that
pair, and the second represents the y-coordinate.
#mSample File;format27#m
5. A sample Scatter diagram data files are located in files
"#tscat.dmo#t"
#u<graphic>;SCATTER scat.dmo default.cfg#u
@
format27
SAMPLE SCATTER DIAGRAM
DATA FILE
Height vs Weight
A Hypothetical Sample
Weight (lbs)
Height (inches)
160 70
180 61
220 75
105 61
155 69
@
format28
MULTIPLE Data File Format.
1. SPC EXpert software allows the user to design custom charts by putting
up to 6 separate graphs in a single multiple graph. The graphs are
automatically sized and placed unless otherwise specified.
2. The first two lines for these files contain the multiple chart title
and subtitle. The length of each line may be anywhere from 0 to 70
characters followed by a carriage return.
3. Multiple data files do not have X-axis and Y-axis labels.
4. Beginning on the third line, the multiple file consists of filename
and chart-type pairs. A multiple file may contain up to 6 pairs.
A sample is shown below:
Multiple Graph
Paragraph 5.10 Sample Output
CAR.DAT BAR
CAR.DAT PARETO
CAR.DAT PIE
CAR.DAT TABLE
5. Sample MULTIPLE chart files are located on your SPCEX System Disk:
"#tmulti3.dmo#t"
#u<MULTIPLE graphic WITH titles>;MULTIPLE multi3.dmo default.cfg#u
#u<MULTIPLE graphic WITHOUT titles>;MULTIPLE multi3.dmo notitles.cfg#u
"#tmulti2.dmo#t"
#u<MULTIPLE graphic WITH titles>;MULTIPLE multi2.dmo default.cfg#u
#u<MULTIPLE graphic WITHOUT titles>;MULTIPLE multi2.dmo notitles.cfg#u
"#tmulti1.dmo#t"
#u<MULTIPLE graphic WITH titles>;MULTIPLE multi1.dmo default.cfg#u
#u<MULTIPLE graphic WITHOUT titles>;MULTIPLE multi1.dmo notitles.cfg#u
@
system1
SPC EXpert
STATISTICAL PROCESS CONTROL SOFTWARE
SPC EXpert is a software tool which automates the statistical and
graphical techniques used in Statistical Process Control (SPC),
Total Quality Management (TQM), and many other techniques designed
for continuous process improvement. SPC EXpert,
1. CONSTRUCTS, ANALYZES, and INTERPRETS 8 different types
of control charts - both attribute and variables.
2. CONSTRUCTs BAR graphs, PIE charts, PARETO diagrams,
HISTOGRAMS, Frequency POLYGONS, and Ogives (CDFs).
3. SPC EXpert CONSTRUCTS and ANALYZES scatter diagrams.
SPC EXpert supports multiple printers; exports to graphics files
which can be imported into commercial graphics and desktop publishing
software such as Harvard Graphics, WordPerfect, MicroSoft PowerPoint,
and MicroSoft Word; produces regression and process capability
statistics; automatically analyzes control charts for out-of-control
conditions and provides expert advice on their meaning; provides
and integrated word processor; imports data from ASCII files;
Comparable commercial SPC software will cost from $200-$1000 per
single-user copy and from $5000-$20000 for a single-site license.
SPC EXpert was developed by
Quality Software Designs, Inc.
933 East Stroop Road
Kettering, Ohio 45429
SPC EXpert Detailed Description
1. PROCESS DATA is collected by the user and placed
into an ASCII file. This file can be created by any
word processing tool. SPC EXpert includes a
built-in text editor for this purpose. The editor can
be used by selecting #mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option J.
2. The user enters the name of the ASCII file which contains
the process data. This is done using #mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m
option A.
3. The user then selects the appropriate control chart or
diagram type. This is done using #mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option B.
4. While the user may directly select any chart/diagram type,
SPCEX provides an automated CONTROL CHART selection service.
The user will be asked 1 to 3 questions concerning the nature
of the process data. From these responses, the appropriate
control chart type will be selected. This is also done using
#mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option B.
5. The user may at this point view the control chart or
diagram using #mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option C.
6. The user may now search a control chart for unusual
patterns. This is done using #mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option D.
AVAILABLE FOR CONTROL CHARTS ONLY.
7. After the pattern search, the user may then view the unusual
patterns highlighted on the control chart. Each pattern may
be view separately or all together. This is done using
#mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option E.
AVAILABLE FOR CONTROL CHARTS ONLY.
8. After the pattern search, the user may also obtain expert
advice concerning the unusual patterns highlighted on the
control chart. Information about each pattern may
be viewed separately or all together. This is done using
#mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option F.
AVAILABLE FOR CONTROL CHARTS ONLY.
9. The user may print out a paper copy of any chart or diagram
by selection #mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option H. Several different
printers,print densities, and paper orientations are available.
10. At any time, the user may edit the current data file (or
any other ASCII text file) by using the SPC integrated
editor. The SPCEX Integrated Text Editor is available by
selecting #mSPCEX MAIN MENU;TOP_MENU_RESPONSES;screens.hyp#m option J.
@
in_control1
IN CONTROL!
Your chart is in control! This means that there are
no unusual patterns within the chart which suggest
the presence of #massignable causes of variation;in_control3#m.
Only #mnatural variation;in_control2#m is present.
@
in_control2
NATURAL VARIATION
The random fluctuation of points within the limits results from
variation built into the process. Such random variation is
natural, results from common causes within the system (e.g. design,
choice of machine, preventative maintenance, etc.), and can only
be affected by changing the system itself.
@
in_control3
UNNATURAL VARIATION
However, points which fall outside of the control limits or which
form "unnatural" patterns indicate that some of the variation
within the process may be due to assignable causes. Assignable
causes of variation (e.g.measurement errors, unplanned events,
freak occurrences, etc.) can be identified and result from
occurrences that are not part of the process.
@
maybe_in_control1
IN CONTROL?
Your chart MAY BE in control. This means that the
SHORT or ABORTED SEARCH found no unusual patterns
within the chart which suggest the presence of
#massignable causes of variation;in_control3#m.
Only #mnatural variation;in_control2#m is present.
SUGGEST LONG SEARCH
The SHORT or ABORTED SEARCH could find no unusual patterns
within the control chart. To ensure that there are absolutely
no unnatural patterns, use a LONG SEARCH of the control chart.
@
charts0
CONTROL CHARTS OVERVIEW
TABLE OF CONTENTS
#mThe Need for Control Charts;charts1#m
#mInterpretation of Control Charts;charts3#m
#mControl Limits;charts5#m
#mTypes of Control Charts;charts6#m
#mChart Type Selection Summary Table;charts11#m
#mRUN Charts;charts12#m
#mMoving Range (XMR) Charts;charts14#m
#mXBAR-R Charts;charts13#m
#mXBAR-S Charts;charts16#m
#mP & PN Charts;charts18#m
#mC & U Charts;charts21#m
@
charts1
THE NEED FOR CONTROL CHARTS
Histograms and check sheets consolidate process data to show the
overall picture, while Pareto diagrams are used to indicate problem
areas. These methods group the data for a specified period and
express them in static form. However, in our processes we also
want to know more about the nature of the changes that take place
over a specified period of time, that is, the dynamic form.
This means that we not only have to see what changes in data occur
over time; we must also study the impact of the various factors in
the process that change over time. Thus, if the materials, the
workers, or the working methods or equipment were to change during
this time, we would have to note the effect of such changes on our
process.
One way of following these changes is by using control charts.
@
charts3
INTERPRETATION OF CONTROL CHARTS
Now the problem is to find out whether the points on the graph are
abnormal or not. Such a determination cannot be made unless
standards of evaluation are set. Without such standards, one is
liable to make an arbitrary judgment or one favorable to oneself
and the graph will not be meaningful. When irrational evaluations
are made, necessary action may be missed or unsuitable action may
be taken in haste, thus causing confusion. This will result in
inappropriate conclusions being drawn, thus lowering quality and
efficiency.
STANDARDS FOR EVALUATION
Limit lines can be drawn on graphs to indicate standards for
evaluation. These lines will indicate the dispersion of the data
on a statistical basis and indicate if an abnormal situation occurs
in your process.
@
charts5
CONTROL LIMITS
A graph or a chart with limit lines is called a CONTROL CHART, and
the lines are called control lines. There are three kinds of
control lines :
(1) Upper Control Limit (UCL)
(2) Central Line (or Average)
(3) Lower Control Limit (LCL)
@
charts6
TYPES OF CHARTS
A control chart's form varies widely according to the kind of data
it contains. Certain data are based on measurements, such as the
measurement of length of a product (in mm) or the measurement of
the time required to sort a batch of mail (in minutes). These are
known as indiscrete values or continuous data. Other data are
based upon counting, such as the number of defective articles or
the number of defects in a product. They are known as discrete
values or enumerated data. Control charts based upon these two
categories of data will differ. Click #mX;charts11#m for a summary
table.
Three major factors affect the choice of chart type :
1. #mIndiscrete vs Discrete Data;charts8#m
2. #mConstant vs Variable Sample Size;charts9#m
3. #mCounting Defectives vs Counting Defects;charts10#m
@
charts8
Indiscrete vs Discrete Data
1. Discrete (Attribute Data)
Examples include : number of errors
number of rejects
number of reworks
2. Indiscrete (Measurement) Data
Examples include : length
weight
time
@
charts9
Constant vs Variable Sample Size
1. Constant - Examples include :
(a) # pin holes in a fixed area of sheet metal
(b) # rejects out of a fixed batch size
2. Variable - Examples include :
(a) # pin holes in pieces of sheet metal differing
in area
(b) # rejects out of batches differing in size
(c) # typographic errors in letters (of differing
length)
@
charts10
Counting Defectives vs Counting Defects
1. Defectives - This is the number of parts/units/etc that
are rejected or un-usable. Examples include :
(a) # letters returned for rework.
(b) # sheets of metal rejected.
2. Defects - This is the number of flaws in a part/unit/etc.
Example include :
(a) # typographic errors in a letter.
(b) # pin holes in a sheet of metal.
@
charts11
CHART TYPE SELECTION SUMMARY TABLE
Chart-type Measurement-type Sample-size Defects/Defectives
---------- ---------------- ----------- ----------------
Run Continuous 1 N/A
XMR Continuous 1 N/A
XBAR Continuous Constant > 1 N/A
R Continuous Constant > 1 N/A
XBAR-R Continuous Constant > 1 N/A
S Continuous Constant > 10 N/A
XBAR-S Continuous Constant > 10 N/A
PN Discrete Constant Defectives
P Discrete Variable Defectives
C Discrete Constant Defects
U Discrete Variable Defects
@
charts12
RUN CHART
Run charts are employed to visually represent data. They are
used to monitor a process to see whether or not the long range
average is changing.
Run charts are the simplest tool to construct and use. Points
are plotted on the graph in the order in which they become
available. It is common to graph the results of a process
such as machine downtime, yield, scrap, typographical errors or
productivity as they vary over time.
-> Select for an #mExample Run Chart Process Measurement;charts17#m
@
charts13
XBAR-R Chart
The XBAR-R control chart is one that shows both the mean value,
XBAR, and the range, R of a sample. This is the most common type
of control chart using indiscrete or continuous values. The XBAR
portion of the chart mainly shows any changes in the mean value of
the process, while the R portion shows any changes in the
dispersion of the process. This chart is particularly useful
because it shows changes in mean value and dispersion of the
process at the same time, making it a very effective method for
checking abnormalities in the process.
-> Select for an #mExample XBAR-R Chart Process Measurement;charts17#m
@
charts14
MOVING RANGE (XMR) Chart
Sometimes collecting enough data to produce a XBAR-R chart is
impossible or at least impractical. Sometimes the natural subgroup
size should be one (1) like when a measurement represents a lot
or batch. In this case we need to be able to look at just a single
measurement as a subgroup, hence an individuals chart. But what
about the fact that the range is based on the variation between
subgroup members? In this case, we use a range chart made of the
range of the last two individuals or a moving range chart.
The XMR chart does not detect changes in the process as fast as
an XBAR-R chart. So we should only use them when it is not
practical to use the XBAR-R chart because of limits in data
availability or if the rational subgroup is one (1).
-> Select for an #mExample XMR Chart Process Measurement;charts17#m
@
charts16
XBAR-S Chart
The XBAR-S chart is identical to the XBAR-R chart except that the
R (range) chart is replaced by an S (standard deviation) chart.
XBAR-S charts should be used when the logical group size is
larger than ten (10).
-> Select for an #mExample XBAR-S Chart Process Measurement;charts17#m
@
charts17
SAMPLE PROCESS MEASUREMENTS
RUN, XMR, XBAR-R, & XBAR-S Chart
(a) The length of time a document sits in an "in-box"
(b) The number of hours required to sort a bundle of mail
(c) The diameter of a bore hole in a machining process
@
charts18
P and PN CHARTS
A P chart is one that shows the fraction defective (p), whereas a
PN chart shows the number of defectives (pn). Basically, they are
the same except that a PN chart is used when the size of the
subgroup (sample size) (n) is constant and a P chart is used when
it is not constant. Obviously, when the size of the subgroup (n)
varies, the defective measurement can only be meaningful in
fractional or proportional terms. The P and PN charts are not used
together as are the XBAR and R charts. This is because P and PN
charts show the characteristics of both mean and dispersion of the
process.
-> Select for an #mExample PN Chart Process Measurement;charts19#m
-> Select for an #mExample P Chart Process Measurement;charts20#m
@
charts19
SAMPLE PROCESS MEASUREMENTS
PN Chart
(a) The number of bytes of a 1 Megabyte file incorrectly
transferred via modem.
(b) The number of broken eggs in a carton of eggs.
@
charts20
SAMPLE PROCESS MEASUREMENTS
P Chart
(a) The number of letters returned for rework each day
(This assumes that the number of letters worked each
day may be different.)
(b) The number of customers who leave each day before getting
service due to long delays.
(c) The number of documents which spent more than a total of 4
hours that day sitting in in-baskets.
(d) The number of defective components in batches of varying size.
@
charts21
C and U CHARTS
A U chart is used in dealing with the number of defectives when the
material being inspected is not constant in area and length such
as the unevenness of woven materials or pin holes in enamel wire.
A C control chart is used in dealing with the number of defects
which appear in fixed unit samples, such as the number of
imperfectly soldered connections in radios, etc.
-> Select for an #mExample C Chart Process Measurement;charts22#m
-> Select for an #mExample U Chart Process Measurement;charts23#m
@
charts22
SAMPLE PROCESS MEASUREMENTS
C Chart
(a) The number of incorrectly soldered connections on a particular
circuit board.
(b) The number of surface flaws in identical pieces of sheet metal.
@
charts23
SAMPLE PROCESS MEASUREMENTS
U Chart
(a) The number of misspellings in documents of varying length.
(b) The number of surface flaws in pieces of sheet metal of varying
area.
@
graph0
GRAPHS/DIAGRAMS OVERVIEW
Select/Click-On An Item of Interest
#mBar Graphs;graph1#m
#mPareto Diagrams;graph3#m
#mPie Charts;graph5#m
#mScatter Diagrams;graph6#m
#mHistograms;graph9#m
#mFrequency Polygons;graph10#m
#mOgives (CDFs);graph11#m
@
graph1
Bar Graphs
Bar Graphs are a very simple way of illustrating the nature of your
process data. A bar graph simply illustrates the relative
frequencies or magnitude of data which can be broken down into
distinct categories.
#m * ;graph2#m Sample Bar Graph Data
@
graph2
SAMPLE BAR GRAPH DATA
Frequency Category
----------- --------------
20 Type A
15 Type B
55 Type C
22 Type D
5 Type E
7 Type F
@
graph3
The Need For Pareto Charts
There are many aspects of any process that could be improved :
defectives, time allocation, cost savings, rework, etc. In fact,
each problem consists of so many smaller problems that it is
difficult to know just where to begin solving them. In order to
be efficient, a definite basis is needed for any action.
A Pareto chart is the tool to use when you need to display the
relative importance of all of the problems or conditions in order
to : choose the starting point for problem solving, monitor
success, or identify the basic cause of a problem.
#mWHAT IS A PARETO CHART?;graph4#m
@
graph4
WHAT IS A PARETO CHART?
A Pareto Chart is a special form of a vertical bar graph which
helps us determine which problems to solve in what order. Doing
a Pareto Chart based upon either Check Sheets or other forms of
data collection helps us direct our attention and efforts to the
truly important problems. We will generally gain more by working
on the tallest bar than tackling the smaller bars. More information
concerning the use of Pareto Charts is available through SPC's
MORE INFORMATION MENU.
@
graph5
Pie Charts
Pie Charts are simply graphs in which the entire circle represents
100% (not 360 degrees) of the data to be displayed. The circle
(pie) is divided into percentage slices that clearly show the
largest shares of data. This is useful in the same way as a Pareto
Chart. The Pie Chart is sometimes even more useful since it is
widely used to display data on T.V. or in the newspapers. More
information concerning the use of Pareto Charts is available
through SPC's MORE INFORMATION MENU.
@
graph6
Scatter Diagrams
A Scatter Diagram is the tool to use when you need to display what
happens to one variable when another variable changes in order to
test a theory that the two variables are related.
A Scatter Diagram is used to study the possible relationship
between one variable and another. The Scatter Diagram is used to
test for possible cause and effect relationships. It cannot prove
that one variable causes the other, but it does make it clear
whether a relationship exists and the strength of that
relationship. Select for #mSAMPLE SCATTER DIAGRAM DATA;graph8#m.
A Scatter Diagram is set up whereby the horizontal axis (x-axis)
represents the measurement values of one variable, and the vertical
axis (y-axis) represents the measurement values of the second
variable. More information concerning the use of Scatter Diagrams
is available through SPC's MORE INFORMATION MENU.
@
graph8
SAMPLE SCATTER DIAGRAM DATA
Person Weight Height
1 160 70
2 180 61
3 220 75
. . .
. . .
50 105 61
@
graph9
Histograms
As we have already seen with the Pareto Chart, it is very helpful
to display in bar graph form the frequency with which certain
events occur (frequency distribution). The Pareto Chart, however,
only deals with characteristics of a product or service, e.g., type
of defect, problem, safety hazards, etc. (attribute data). A
Histogram takes measurement data, e.g., temperature, dimensions,
etc., and displays it's distribution. This is critical since we
know that all repeated events will produce results that vary over
time. A Histogram reveals the amount of variation that any process
has within it.
@
graph10
Frequency Polygons
Another kind of graphical display of a frequency distribution
(Histogram) is the Frequency Polygon. Here the cell (data range)
frequencies are plotted at the midpoint of each cell and the
midpoints are joined by straight lines.
@
graph11
Ogives (Cumulative Distribution Function)
If the same technique in the Frequency Polygon is applied to the
cumulative distribution function, we obtain what is called an
Ogive (rhymes with alive) or CDF. The only difference is that in
constructing the Ogive (CDF), the cell boundary is used as the plot
point rather than the middle of the cell.
@
hypertext help
HYPERTEXT INFORMATION SYSTEM
1. You are currently using the hypertext information system.
2. Text which is highlighted in light green or light blue are
referred to as hot links.
3. This is a #mSample Text Hot Link#m Click on this with the
left mouse button or by pressing <ENTER>.
4. "#tbar.dmo#t" is a sample Text File Hot Link. Click
on it with the left mouse button or move the highlight
over it using TAB or DOWN-ARROW and then press
<ENTER>.
5. HYPERTEXT window help is available by pressing <F1>
6. The entire contents of any hypertext window may be saved to
an ASCII (DOS TEXT) file by pressing <F3>. This is the best
way to print the contents of a hypertext window by importing
the resulting ASCII file into a standard word processor.
@
Sample Text Hot Link
After you click on a Text Hot Link, additional text will come
up in a window. #mOther hot links#m may also be in this window.
To go back to the previous window, press ESC or click the
right mouse button.
@
Other hot links
Again, a text box appears. This
may repeat for several levels.
To go back to the previous window,
press ESC or click the right mouse
button.
@