Professional Documents
Culture Documents
Statistical Software
Agenda
Minitab History & Introduction
Minitab & Six sigma
Minitab Project Manager
Data, Calc menu items & Quality planning tools
Hypothesis testing
Capability Analysis
Correlation and Regression
Measurement System Analysis
Major Enhancements in MINITAB 15
Supports Provided by Minitab
Other Six-Sigma software by Minitab
Minitab History &
Introduction
History
Minitab Statistical Software was initially developed in
1972 by three members of the statistics faculty at the
Pennsylvania University.
Goal : To make statistics more interesting and
meaningful to students
Minitab soon became the world's leading statistical
software.
Currently Minitab is used by:
- 4000 colleges and universities worldwide
- Over half the companies in the Fortune 500
Introduction
The leading package for Sixsigma and Quality
improvement
It contains statistical method you need
Generates graphs that are easy to interpret, simple
to learn and use
Data can be imported and exported from .XLS, dat/txt
and from database
Spreadsheet based and compatible with Microsoft
products
User-friendly software
Minitab & Six
Sigma
Minitab & Six Sigma
Minitab Project
Manager
Project Manager
Project Manager organizes all the project elements
into a familiar folder structure for quick and easy
access to any item in the project.
The Project Manager includes:
• Worksheets folder
• Session folder
• Graphs folder
• History folder
• Report Pad folder
• Related Documents folder
Worksheets Folder
The Worksheet folders contains an automatically
updated summary of the current worksheet.
Helps to keep track of variables if you have a
large worksheet.
Enables to check the Count, Missing values and
datatype for a worksheet.
Each worksheet can contain 4000 columns and
10,000,000 rows.
Worksheets Folder
Session folder/window
A Minitab window that displays the text output
of your analysis, such as statistical test results
and related notes or error messages
Session window enables user to Edit, Save and
Print the contents for reference.
Session window text and Minitab graphs can be
combined in ReportPad or a word processor to
create reports.
Contents in the session folder are editable
Session folder/window
Graphs folder
To manage all of the graphs in your project.
By highlighting the graphs from the list in the
Graphs folder we can:
- Save, Copy, or Print one or more graphs
- Tile or Layout multiple graphs across the
Minitab screen for easy viewing and
Comparison
- Rename individual graphs
- Append graphs to the ReportPad folder
Graphs folder
History folder
Provides a convenient overview of what you have
done in your session along with the changes
done in the datawindow(worksheet)
To view the submitted commands through the
user interface during your Minitab session
Complex commands can be re-executed by
copying them from the History folder and pasting
them into the Command line Editor.
To automate a repetitive task by creating
MACROS.
Contents in the History folder are non-editable.
History folder
Report Pad folder
Enables to create reports on data and output very
quickly.
Minitab graphs and Session window output can
be appended to the ReportPad folder.
Once the report is added it can be enhanced by
using the built-in word processor to add text,
notes, captions, or headings.
Reports can be saved as RTF or HTML file
Contents in the Report Pad folder are editable
Report Pad folder
Related Documents
To access project related non-minitab files very
quickly.
Links can be added non-Minitab files or web
page internet addresses that are related to your
Minitab project for easy reference.
Description can be given for the related
documents for Identification.
Related Documents
Data, Calc menu items
&
Quality planning tools
Data Menu
Some of the important items in Data menu are
Split Worksheets
Merge Worksheets
Stack, Unstack & Transpose columns
Concatenate
Calc Menu
Some of the important items in Calc menu are
Calculator
Column Statistics
Row Statistics
Random data
Probability distributions
Quality Planning tools
Some of the quality planning tools covered in
Minitab are:
Scatter plot
Histogram
Dot plot
Box plot
Bar & Pie charts
Time series plot
Cause and Effects diagram (Fishbone)
Pareto chart
Run Chart
Hypothesis testing
Hypothesis Testing Concept
What Actually Is
Innocent Guilty
What was Your Decision
Innocent
ERROR
Correct Decision
TYPE II
Guilty
ERROR
Correct Decision
TYPE I
Understanding Risk
What Actually Is
Innocent Guilty
What was Your Decision
Innocent
Inference:
The P-value > 0.05.
We accept the Null Hypothesis and Reject the Alternate hypothesis
We are 95% confident about this decision
Conclusion: The sample data can be considered for Analysis
1-Sample t test
You have 20 sample data and you want to compare it with
the historical mean of 5 and standard deviation of 1. If the
sample data matches with the historical mean you can
consider it for your analysis
Inference:
The P-value > 0.05.
We accept the Null Hypothesis and Reject the Alternate hypothesis
We are 95% confident about this decision
Conclusion: The sample data can be considered for Analysis
2-Sample t test
You have 2 sets of samples and you want to compare
whether there is a difference between the 2 sets. If there is
difference you can consider the best one based on your
requirements
Inference:
The P-value > 0.05.
We accept the Null Hypothesis and Reject the Alternate hypothesis
We are 95% confident about this decision
Conclusion: There is no difference between the 2 sets of samples
Paired t test
You have received some sets of samples and you are not
ok with the output. In order to improve the output you
thought of providing a training. After the training you
collected the same sets of samples. Now you need to
check the effectiveness of the training
Inference:
The P-value < 0.05.
We accept the Alternate hypothesis and fail to Accept the Null Hypothesis
We are 95% confident about this decision
Conclusion: There is effect out of the training provided
1-Proportion test
A county district attorney would like to run for the office of
state district attorney. She has decided that she will give
up her county office and run for state office if more than
65% of her party constituents support her.
As her campaign manager, you collected data on 950
randomly selected party members and find that 560 party
members support the candidate.
Inference:
The P-value > 0.05.
We accept the Null Hypothesis and Reject the Alternate hypothesis
We are 95% confident about this decision
Conclusion: The proportion of party members that support the candidate is not
greater than the required proportion of 0.65. As her campaign manager, you
would advise her not to run for the office of state district attorney.
2-Proportion test
As your corporation's purchasing manager, you need to authorize the
purchase of twenty new photocopy machines. After comparing many
brands in terms of price, copy quality, warranty, and features, you
have narrowed the choice to two: Brand X and Brand Y. You decide
that the determining factor will be the reliability of the brands as
defined by the proportion requiring service within one year of
purchase.
Because your corporation already uses both of these brands, you
were able to obtain information on the service history of 50 randomly
selected machines of each brand. Records indicate that six Brand X
machines and eight Brand Y machines needed service. Use this
information to guide your choice of brand for purchase.
Inference:
The P-value > 0.05.
We accept the Null Hypothesis and Reject the Alternate hypothesis
We are 95% confident about this decision
Conclusion: The proportion of photocopy machines that needed service in the first
year did not differ depending on brand. As the purchasing manager, you need to
find a different criterion to guide your decision on which brand to purchase.
1-Variance test
You are a quality control inspector at a factory that builds
high precision parts for aircraft engines, including a metal
pin that must measure 15 inches in length. Safety laws
dictate that the variance of the pins' length must not
exceed 0.001in2.
You collect a sample of 100 pins and measure their length
in order to conduct the hypothesis test
Inference:
The P-value < 0.05.
We accept the Alternate hypothesis and fail to Accept the Null Hypothesis
We are 95% confident about this decision
Conclusion: You should conclude that the variance of pin length is small enough
to meet specifications and ensure passenger safety.
2-Variance test
A study was performed in order to evaluate the
effectiveness of two devices for improving the efficiency
of gas home-heating systems. Energy consumption in
houses was measured after one of the two devices was
installed. The two devices were an electric vent damper
(Damper = 1) and a thermally activated vent damper
(Damper = 2). You are interested in comparing the
variances of the two populations
1 P-Value 0.558
Lev ene's Test
Test Statistic 0.00
P-Value 0.996
2
5 10 15 20
BTU.In
2-Variance test
Inference:
The P-value > 0.05.
We accept the Null Hypothesis and Reject the Alternate hypothesis
We are 95% confident about this decision
Conclusion: These data do not provide enough evidence to claim that the two
populations have unequal variances
1-Way ANOVA
Twenty four golf balls with different dimple patterns are
selected and checked for the distance traveled. Now
Analysis needs to be performed whether there is a
difference in the distance traveled of the golf balls
Dimple 1 Dimple 2 Dimple 3 Dimple 4
277 281 304 250
268 299 295 277
281 317 317 268
263 286 299 272
290 304 281
295 304 286
281
263
Inference:
The P-value < 0.05.
We accept the Alternate hypothesis and Fail to Accept the Null Hypothesis We
are 95% confident about this decision
Conclusion: At least one dimple pattern is different
1-Way ANOVA
Individual Value Plot of Distance vs Dimples Boxplot of Distance
320 320
310 310
300 300
290 290
Distance
Distance
280 280
270 270
260 260
250 250
Inference:
The P-value < 0.05 for Dimples and >0.05 for Players
We accept the Alternate hypothesis for Dimples and Null hypothesis for Players
We are 95% confident about this decision
Conclusion: There is a difference in the dimple pattern but No difference in the Players
Capability Analysis
Capability 6-Sixpack
A manufacturer of cable wire wants to assess if the
diameter of the cable meets specifications. A cable wire
must be 0.55 + 0.05 cm in diameter to meet engineering
specifications.
Analysts evaluate the capability of the process to ensure
it is meeting the customer's requirement of a Ppk of 1.33.
Every hour, analysts take a subgroup of 5 consecutive
cable wires from the production line and record the
diameter.
Capability 6-Sixpack
Indicates
Capability 6-Sixpack Stable process
0.56
_ LS L 0.5
_
X=0.54646 U S L 0.6
0.54
0.05 _
R=0.0431
0.00 LCL=0
1 3 5 7 9 11 13 15 17 19 0.50 0.55 0.60
35
Mileage (km/Lit)
30
25
20
15
25 35 45 55 65 75
Speed (km/h)
Scatter diagram
A scatter diagram depicts the relationship as a pattern
that can be directly read.
If Y increases with X, then X and Y are positively
correlated.
If Y decreases as X increases, then the two types of
data are negatively correlated.
If no significant relationship is apparent between X
and Y, then the two data types are not correlated.
Different scatter diagram patterns
Correlation
Analysis is done to check whether there is correlation
between the Marks scored in entrance examination vs
the Marks scored in Graduation. Data samples of 20
persons are taken for Analysis
Inference:
The P-value < 0.05
We accept the Alternate hypothesis and Fail to accept the Null hypothesis
We are 95% confident about this decision
Conclusion: There is correlation between the marks scored in Entrance
examination vs the Graduation degree marks
Scatterplot of Graduate Marks vs Entrance Marks
100
95
Graduate Marks
90
85
Strong
80 Positive
correlation
75
70
70 75 80 85 90 95 100
Entrance Marks
Regression
Regression is the prediction of dependent variable
from knowledge of one or more other independent
variables.
Regression Analysis is a statistical technique for
estimating the parameters of an equation relating a
particular value of dependent variable to a set of
independent variables. The resulting equation is
called Regression Equation.
Linear regression is the regression in which the
relationship is linear.
Curvilinear regression is the regression in which the
best fitting line is a curve.
Simple linear regression
Only a single predictor variable or independent
variable „X‟ (e.g.: cutting speed) and a response
variable or dependent variable „Y‟ (e.g: tool life).
Inference:
The P-value < 0.05
We accept the Alternate hypothesis and Fail to accept the Null hypothesis
We are 95% confident about this decision
Conclusion: There is correlation between the TV watched HRS and Marks Scored
90
Reg-Marks scored
80
70
Strong
60
Negative
correlation
50
1 2 3 4 5 6 7
Reg-TV watched
Regression
Since there is a correlation we can proceed to
Regression Analysis
Regression Regression equation calculated
for the data given, by Minitab
Inference:
The P-value < 0.05
We accept the Alternate hypothesis and Fail to accept the Null hypothesis
We are 95% confident about this decision
Conclusion: Hours spent in watching TV significantly affects the Marks scored
Prediction
Based on the regression data we
concluded Hours spent in watching
TV significantly affects the Marks
scored. So we can use the below
regression equation for prediction
Inference:
Predicted value is 51.64. If I spend 8hrs watching TV the marks I will score is 51.6%
Measurement System
Analysis
Possible Sources of Process Variation
Observed Process Variation
Performance is Poor.
Need training
Minitab – Attribute GRR
Attribute GRR – Examples
Call Center Call Quality Agent – Call agent Rating -
Score of 1 to 5
Canteen Lunch – Good / Bad
Performance appraisals – Discrete rating scale
Painted parts – Accept or reject
Continuous GRR – Case Study
In a supermarket the customers started complaining
about wrong weights of the apples sold. The
Supermarket management already familiar with the
concept of MSA, was sure that the measurement
system could be the possibility. So they decided to
conduct the Continuous Gage R & R studies.
% Contribution of Total Gage R&R indicate the resolution of the measuring equipment.
Less than 1% - the measurement system is acceptable.
Between 1% and 9% - the measurement system is acceptable depending on the
application, the cost of the measuring device, cost of repair, or other factors.
Greater than 9% - the measurement system is unacceptable and should be improved.
Minitab – Continuous GRR
The Range chart provides the range of the weight measured by each
operator for each apple. Ideally, all the values should be on the bottom Red
line. But as long as they are within the 2 red lines (Control Limits), things
are OK
Minitab – Continuous GRR
The Xbar chart gives an idea about the ability of the measuring instrument
to distinguish different types of parts. It plots the average of all the
Operators, apple-wise for the various repetitions. If all of them are within
the 2 red lines (Control Limits), that means that all the means seem to be
similar or in other words the Measuring system cannot distinguish between
the different parts (apples). As a thumb rule, 50% or more points outside the
2 Red lines is a good indication
Minitab – Continuous GRR
The Weight By Part plots the weight of each apple measured different times.
Ideally, all the points of a certain apple should be a single dot. The variation
for each of the apple measurements shows lack of repeatability or
reproducibility
Minitab – Continuous GRR
Weight by Operator plots the average weight of all the parts (Apples) by
each operator. Ideally, this should be a straight line. The operator with the
deviation indicates some kind of problem with the operator while making
the measurement.
In this example, Sandeep seems to have a problem
Minitab – Continuous GRR
The Operator * Part Interaction plots the values of the various apples against
the operators. Ideally, a certain apple measured by any operator should be the
same. If any difference is shown, that in conjunction with the “By Operator”
and the “By Part” graphs gives an idea whether the problem is with the
operator, part or between the operator to part interaction.
Major Enhancements
in
MINITAB 15
Formulas in Worksheet
Column calculates by formula, updates
with new data
Displayed on:
Worksheets
- Graphs
Format Dialog
- Selected statistical output
$ 10.00
Revenue
$ 8.00
$ 6.00
$ 4.00
$ 2.00
$ 0.00
Year 2002 2003 2004 2005
New column formats and calculations
Elapsed time (new Elapsed function in Calculator)
Can calculate work days
Date formats: like January 1, 2006 or no
separators (20060101)
Thousands of a second
Multiple Undo/Redo in Data Window
Undo mistakes, redo changes
Release 14: one-step change for Edit menu functions
- Release 15: multiple changes and supports more functions, including:
Editor > Replace, Editor > Format, Formulas