Understanding Six Sigma

Understanding Six Sigma
Understanding Six Sigma Manual
6σ
History of Six Sigma Understanding Six Sigma
The U.S Defense system developed a system known as SQC to manage the
complex weapon system & to handle the distributed defense contractors.
SQC:- is a set of tools that originated in the Military Standards and the basis of
SQC process was 3 sigma limits which yields a rate of 2700 defects per million.
After World war 2 US companies returned to their Original strategy while the
defeated countries were rebuilding their Industries. General Mc Arthur who was the
Governor general of JAPAN at that time Imported some of the U.S. Pioneers of
SQC to help train their counterparts in JAPAN.
By 1970~ 1980 Japanese producers were renowned for their Quality & durability.
U.S Companies slowly realized that to attain the desired Quality level two things
are necessary
One should be able to measure the quality level I.e it should be Quantifiable
& Measurable.
Motorola pioneered the Use of Six Sigma , Bill Smith VP &Senior QA
Manager of Motorola is regarded as the Father of Six Sigma.
Quality Management & Six Sigma

Total Optimization
of R&D,
“ Six Sigma Management” ---
Production , Sales
Lead by American Companies Six Sigma
and Service is
necessary
• Small group of sections

Total Quality
• Toward total Solution Management
Improve Product
Quality Total Quality
Control
Quality
` “ Production line focussed Improvement “
Control
--- Lead by Japanese Companies
Most of service called products were

from reworked products at the factories
• Bill Smith Report Hidden Factory and Rolled throughput
Yield concept are Induced
• Actual Practice Strategy by Dr. Michael Harry
• The Malcom Baldridge Award of 1988 of Motorola
• After Motorola, Texas Instruments, ABB, Allied Signal , GE, LG

Electronics, Polaroid, Nokia, Lockheed Martin, …. Sony started
Six Sigma
Definitions Understanding Six Sigma
Population Sample Data

Total Study Group Small Group taken The facts derived
from Population from the sample
The Group of Sample

100
people who
Age and Number of
possesses a Fan
decision makers
purchasing a Fan
Definitions Understanding Six Sigma
Data Point : The single entity in the sample.

Data : The trend of data points in a sample I.e the facts derived from the sample.
Information : The data presented in a form which conveys some result, Conclusion
Sample : A Sample is a portion of the whole collection of Items (population)
Population : The population consists of the set of all measurements in which the
investigation is interested. It is also called Universe.
Statistic : A numerical value such as standard deviation or mean , that characterizes the
sample.
Statistics : An application theory & method to reach appropriate & wise decisions in
unknown circumstances.
Population The Characteristic of populaton : parameter
Sample
The Characteristic of Sample : Statistic
Types of Data
A.) Continuos Data : - The Data which can be measured and has unit
associated with it is called continuous data. It can be in fractions.
E.g Length of a Playground, Thickness of the paint coating, ect.
B) Attribute Data :- The data which can has only two options Yes/No,
True/False is called Attribute data.
E.g Quality of Food ( OK/NG),
C) Discreet Data : - The data which can be measured only in whole

number and has no units associated with it is called Discreet data. It is the
Count of the Number of Attributes E.g number of Heart Beats in one
minute, Number of Type A defects.
Characteristics of Normal Curve

Normal Distribution Curve is known as Density
Curve meaning the area under the curve is
equal to one.
A.) The Normal curve is a bell shaped curve and it
has single peak (Mode ) at the center.
B) The mean & median of the distribution are equal
and are located at the peak.
x LSL
C) The Normal distribution curve is symmetrical USL
about the mean.
D) The curve is asymptotic I.e the curve gets close
to X-Axis but it never touches it
Standard Normal curve is one having Mean=0,
and standard deviation =1.
Measures of Central Tendency of Data

A.) Mean : The mean of a set of Observations is their average. It is equal to the sum of all
Observations divided by the number of Observations in the set. E.g Consider the data set
given below
1 , 2 , 3 , 4 , 5 ; Mean = 1+ 2 + 3 + 4 + 5 = 15/5 = 3
X = ΣX =
n
Mean of Sample, i x1 +x2 +x3 +x4+……..xn
I=1
n
b.) Median : The Median term of the given data is given by
Median = n + 1 th term, where n is the number of
2
Observations in the given data(arranged in increasing order).
E.g 3, 5 , 1 , 8 , 2 , 7 , 1 , 4 . No. of Terms = 8, Arrange data in Increasing Order = 1 , 1 , 2 , 3 , 4 , 5 , 7, 8
Median Term = 4.5th Term = 4th Term + 0.5 ( 5th - 4th Term ) = 3 + 0.5 ( 4- 3) = 3.5
Measures of Central Tendency of Data
C) Mode : The value Occurring maximum number of times

E.g 3, 4 , 3, 6, 5, 3, 7, 4 , Mode = 3
E.g Calculate the Median of the Following data :

2, 4, 6, 8, 12, 15, 10
Sol. Arrange the data in Increasing Order
2, 4, 6 , 8 , 10 , 12 , 15
No.of Observations = (7 + 1 ) / 2 = 4th Term = 8
Measures of Spread/Dispersion of Data
a.) Range : The Difference between Maximum & Minimum value.

b) Standard Deviation ( σ ) : It gives tells us about the variation in data .
c ) Variance ( σ 2 ) : It is defined as the square of the standard deviation to
account for the total variation observed in the data.
E.g : The Process Specification = 10 ± 2

Sol. USL = 12 , LSL =8 Range = USL - LSL = 12 - 8 = 4 Mean = 3
Consider the data set 3, 2, 5, 1, 4 Data Variation Variation 2
1 -2 4
Mean = 3 2 -1 1
Standard Deviation ( σ ) = (10/4 )1/2 = 1. 5 3 0 0
4 1 1
Variation ( σ 2 ) = 2.25 5 2 4
0 10
Measures of Spread/Dispersion of Data

d) Quartile : It divides the total range into 4 equal parts (quarters ) and tells that in
which Quartile a particular data point is lying.
th th th
Q1 = n + 1 , Q2 = 2 * n + 1 , Q3 = 3 * n + 1
4 4 4
where Q1,Q2 and Q3 are 1st 2nd & 3rd Quartiles resp. Interquartile Range ( IQR ) = Q3 -
Q1, contains 50 % of the Total data points
Upper Limit UL = Q3 + 1.5 IQR
Lower Limit LL = Q1 - 1.5 IQR
Eg. Calculate the First , second & Third Quartile for the data given below. Also calculate the IQR &
Upper & Lower Limits. 10, 17, 14, 23, 18 , 11 , 9, 13
Sol . Arrange data in Increasing order = 9 , 10 , 11, 13, 14 , 17 , 18 , 23
Q1 = [ ( 8 +1) / 4 ] th = 2.25th Term = 2nd Term + 0.25 ( 3rd Term - 2nd Term ) = 10 + 0.25( 1) = 10.25
Q2= 2* 2.25th Term = 4.5th Term = 13 + 0.5 (14 - 13 ) = 13.5, Q3= 3* 2.25th Term = 6.75th Term = 17.75
IQR = Q3 -Q1 = 17.75 - 10.25 = 7.50 ; UL = 17.75 + 1.5( 7.50 ) = 29 ; UL = 10.25 - 1.5 (7.50) = -1
Six Sigma : Introduction Understanding Six Sigma
What is Six Sigma ….???????
1) Statistical Measurement :
We measure defect rates in all the Processes through an expanding

statistical concept, and we use ‘ σ ’in measuring process capability.
2) Business Strategy :
We gain a competitive edges in Quality, Cost, Customer Satisfaction.
3) Philosophy :
We should work smarter, not harder

Z Level Harvest
Sweet Fruit
Z PPM Design for Manufacturablity
5 σ Wall, Improve Designs
6 3.4
Bulk of Fruit
5 233 Process Characterization and Optimization
4 σ Wall, Improve Processes
4 6,210
Lower Hanging Fruit
Seven Basic Tools
3 66,807
3 σ Wall, Work with suppliers
2 308,537 Ground Fruit
Logic and Intuition
Process Defect
Capability Opportunity
The Percentage 68.3 %

1σ
Acceptable Area under 95.45 %
2σ
the Curve Increases as 99.73 %
the Z Value ( the Number 3σ
of Standard Deviations σ 99.9936 %
4σ
) Increases .
99.99 99 4 %
5σ
99.99 99 99 8 %
6σ
USL x LSL
The Area Under the Curve represents the Acceptance or Yield,

whereas the Area outside the Curve represents the Rejection
Six Sigma : D-M-A-I-C Understanding Six Sigma
Objective
Theme Selection
Theme
Selection
• Define problem
(Define) • Define range
Measurement
• Measuring capability
Y Capability OK ? Measurement
Redesign of CTQ
N • Clearfy measuring method
Analysis
Analysis • Clearfy factors
Redesign ? Y
N
• Find vital few
Improvement Improvement • Optimize process
N Capability OK ?
• Control vital few
Y Control
• Set up control system
Control
3σ Vs 6σ Company Understanding Six Sigma
The 3 Sigma Company The 6 Sigma Company
• Spends 15-25% of sales dollars on • Spends 5% of sales dollars on cost

cost of failure of failure
• Produces 66,807 defects per million • Produces 3.4 defects per million
opportunities
opportunities
• Relies on capable prcesses that
• Relies on inspection to find defects don’t produce defects
• Belives high quality is expensive • Knows that the high quality producer
is the low cost produer
• Benchmarks themselves against
their competition • Benchmarks themselves against
the best in the world
• Believes 99% is good enough • Believes 99% is Unacceptable
• Defines CTQs internally • Defines CTQs from customers
Define-Application of Six Sigma Understanding Six Sigma
Six Sigma is a tool that can be applied to all business systems, Design,
Manufacturing, Sales & Service
Guarantee for design completion

• Selecting CTQ to meet customer requirement
R&D
• Deciding reasonable Tolerance
• Guarantee CTQ’s through capability analysis
Six Sigma
Quality Assurance in Manufacturing Stages

• Improve serious Problems
Mfg
• Real Time Monitoring
• CTQ control system
Maximizing Sales & Service

SVC • Improve cycle time & accuracy
• Cost Improvement
Define Understanding Six Sigma
Define Phase
• Pareto Analysis
• Process Mapping
• Logic Tree
• 3/5 Why Analysis
• RTY
• QFD
• FMEA
• Brainstorming
Pareto Analysis : The Origin of the Tool lies with the Italian Economist Vilfredo
Pareto. Pareto Principle is also known as 80/20 .20% of items purchased by the
company accounts for 80% of the value 1st Item in the figure below indicates the
highest no of faults
Pareto Charts are a type of bar chart in which the horizontal axis represents categories
of interest, rather than a continuous scale. The categories are often "defects." By ordering
the bars from largest to smallest, a Pareto chart can help you determine which of the defects
comprise the "vital few" and which are the "trivial many." A cumulative percentage line helps
you judge the added contribution of each category. Pareto charts can help to focus
improvement efforts on areas where the largest gains can be made.
Pareto chart can draw one chart for all your data (the default), or separate charts for groups
within your data.
Example :The company you work for manufactures metal bookcases. During final
inspection, a certain number of bookcases are rejected due to scratches, chips, bends, or
dents. You want to make a Pareto chart to see which defect is causing most of your
problems. First you count the number of times each defect occurred, then you enter the
name of the defect each time it occurs into a worksheet column called Damage.
Damage Counts • Choose Stat > Quality Tools > Pareto Chart.
Scratch 274 • Choose Chart defects data in and enter
Scratch 59 Damage. Click OK.
Bend 19
Graph window output
Chip 43
Dent 4
Scratch 8
Chip 6
Scratch 10
Process Mapping
• Process mapping is used to document process to examine part and information flow.
• It is a key tool in identifying opportunities for improvement.
The Process Mapping Method
• Define the Process boundary. (General area or specific process you intend to
improve)
• Brainstorm and order process steps with your team.
• Code activities using symbols for easy analysis.
• Walk through the process to validate map.
• Add key process metrics
- yield, costs, rolled throughput yield, scrap, overtime $, capacity, %schedule, %OTD
• Analyze map for key business issues -could be in the areas of :
- Process loss or waste
- Cycle time improvements
- Quality improvements
- Flow improvements
Process Mapping [ex] Vehicle Purchasing
Meeting Look around Look around

Car Shop arrival Small Talk
Salesman first car second car
Look around Trial No Shop around

third car Driving? for another shop
Yes No
Operation
Receive Trial Whether you Yes Decide Review the
key and Tag Driving buy or not? to buy sales manager
Decision
Decide
Contract Value
Delay
Decide the Whether you No Visit another measurement

Price purchase another or not? car shop
Yes
Storage
No
Decide Review the Decide Yes Make out
Reasonable
to buy sales manager the price price? final contract
w
Transmission
Get a Credit Stand by Drive the
loan check for a loan new car
Process Mapping [ex] Refrigerator - R1 Line Rolled Throughput Yield
Door Ass’y 89.7%

D/Plate plate/paint
99.0%
D/Liner extrusion/mold Door forming Door assembly

99.7% 93.4% 97.3%
99.6% 81.0%
I/Case extrusion/mold Case forming Cycle assembly

83.8%
97.7%
Front - CTQ, L painting O/Case, B/Plate LQC & appearance

99.2% 91.7% 96.5%
Case Ass’y 73.4%
Output
Rolled Though put Yield = 73.4% ×
89.7% ×97.7% × 83.8%× 96.5% = 52.0%
▲ ▲ ▲ ▲ ▲
Case Door Cycle assembly LQC& appearance
Logic Tree (Structure Tree)

• Used to break down problem into manageable groups to identify root cause or area
of focus.
• Breakdown the problem on the base of MECE
- MECE - Mutually Exclusive Collectively Exhaustive
Mutually Exclusive : When a Problem is broken into further sub parts there should not
be anything common among the factors.
Collectively Exhaustive : Also there should be nothing left to represent the Main factor
Why
Electromagnetic Losses
Lamination Inductance
Why Mechanical OD
Rotor
Endrings Area A Core length
RPM Stator
Area B
Assembly
6σ is a kind of type which can improve the problem (RPM) by practicing improvement
activity for the lower level displayed in the long run
5 Why Analysis : Five Why analysis is done to determine the root cause of the
Problem . It is a kind of brainstorming to reach the root cause of the Problem.
It is Observed that by the time you arrive at the 5th Why the solution of the
Problem is with you. It is not essential to ask why 5 times, you can locate the root
cause at the 3rd or 4th Why also..
Rolled Throughput Yeild
RTY : Rolled throughput Yeild : It is the Probability that the product will pass
through all the stages without any rejection / rework.

RTY is calculated by calculating the YFT’s of Individual stages.
RTY = YFT1 * YFT2 * ……… YFTn ,
where YFT’s are the First time yields of the Individual Stages/Process
connected in series.
YNA = Normalized Yield = ( RTY of the line )1/no. of stages in Line
YNA gives the average yield of line . This is used to calculate when we have
to compare the performance of two lines on the basis of RTY.

RTY is the probability of going through all the processes with zero-defect the first
time . Also, it provides an indication of opportunities to reduce the waste.
To increase productivity
Goal through Quality Improvement
Input
-Process 1: (Acceptance-
rate:99.0%) Overall process’s defect, m/c
Target trouble, No work,L/B,Model
-Process 2 (92.0%) Change Loss,Non-value added work
-Process 3 (97.0%)
Tool 6 Sigma
Final Inspn.
(97.0%) TDR, 6 Sigma, NWT,
Activity
One man one project
Un-controlled Total Process Final
Loss Defect-rate Product
- Process defect rate
- Self & sequential inspection
To improve all
RTY = 0.99×0.92×0.97×0.97= 85.7 % the hidden defects of all
* RTY : Rolled Throughput Yield the processes
Yield First Time
1000 Are 820

Painted Yes
the parts
Components
Good ?
No
100 Can 30 Rework &

the parts be Pass-on
Rework & Yes repaired ? Yes ( Paint Touch up)
Reprocess
( Rust, Chemical Wash) 50
No
Scrap
(Dent)
First Time Yield Rolled Throughput Yield Normalised Yield
YFT = S/U YRT = YFT1 x YFT2 x YFT3 YNA = (YRT)**1/Opp

YFT = First Time Yield YRT = Rolled Throughput YNA = Normalized Yield
S = No. of units that pass the yield Opp = Number of opportunities
first time YFT1,YFT2,… = First Time Yield of
U = No. of units tested each process
•YRT gives the •YNA us average

probability of going yield of processes.
YFT gives the through all the •YNA allows for
processes with zero
probability of going calculation of Z value
defects in the first time.
•YRT provides an of processes.
through one process •YNA allows for
indication of
with zero defects. opportunities to comparison
reduce waste. between processes.
QFD : Quality Function Deployment ( What Customer Wants )
It is defined in two steps :
a) Converting customer’s Voice into Engineers Voice
b) Converting Engineers voice into Technical CTQ’s & CTP’s
QFD is tool which is used to generate data in the form of taking feedback from the
customer through quality matrix, converting those requirements into Tech changes
in the Process through Quality Matrix.
• Identify key consumer cues by reviewing market, reliability requirements, general
requirements and current quality issues.
• Rank cues by importance and translate them into technical specifications required to meet
customer cues. Rank technical specifications by impact on customer cues and translate
them into potential part characteristics(CTQ’S).
• Rank part characteristics by impact on meeting technical specifications(CTQ’S)
QFD translates the Voice of the Consumer into the Voice of the Engineer.
QFD : Sub Process 1 : To Convert Consumer’s Voice into Engineer’s Voice
Engineer's Voice
Priroty
System for
Capacity of
HE Coils
Charging
Fins per
Ranking
Type of
Blower
/Scroll
Motor
Comp
Inch
Gas
Less Price 9 3 1 1 1 10
Low Noise 3 9 9 1 1 6
Requirement
Customer's
Air Flow 1 1 9 3 3 5
Less Power 9 9 1 3 1 8
More Cooling 9 3 3 9 9 3
Rating 212 170 126 82 66
QFD : Sub Process 2 : To Convert Engineer’s Voice into Potential CTP’s & CTQ’s
Potential CTP's & CTQ's
Comp Specifications
EER Specifications
HE Design Spec to
Variation less than
Gas Charging Qty
be maintained
KW Rating
Priroty
Ranking
5mg
Type of Comp 9 1 3 3 1 212
Engineer's Voice
Capacity of Motor 1 9 3 1 1 170
Blower /Scroll Design 1 1 9 1 3 126
System for Gas Charging 1 1 1 9 3 82
HE Coils Fins per Inch 1 1 1 3 9 66

3096 2016 2428 1848 1600
FMEA : Failure Mode Effect Analysis ( What Customer Doesn’t want )
It gives you possible reasons in which a given Process / Design of part of a

Product can Fail. To every Failure Mode we associate RPN Number
RPN : Risk Priority Number = Severity * Occurrence * detection
Rating Scale (1 ~ 10) (1 ~ 10) (1 ~ 10)

1 : If the Problem is Less Severe
Severity
10 : If the Problem is Life Threatening.
1 : If the Problem has chances of less occurrence
Occurrence
10 : If the Problem has more chances of occurrence
Detection 1 : If the Problem is easily detectable.

10 : If it is difficult to locate/detect the Problem
FMEA is used to proactively identify and rank risks in a product design and
assign appropriate actions to be taken to prevent the failure mode.
FMEA Process
• Brainstorm potential failures of the product design.
• Assign severity and probability (likelihood of occurrence) ratings to each
potential failure mode.
• Determine existing control measures being taken to eliminate significant
failure modes.
• Develop actions to be taken to eliminate or reduce risk on all remaining
significant failure modes.

Brainstorming : It is Discussion among the Process Experts.The basic rule of

brainstorming is no ideas are criticized.
Brainstorming is of three Types :
a) Freewheel
b) Round Robin
c) Card Method
In Freewheeling type of brainstorming, everybody participates in the

simultaneous discussion
In Round Robin type of brainstorming each Individual in the group is given a

chance to give his opinion
In Card type method the Individuals write their Ideas on the Card
Measure Understanding Six Sigma
Measure Phase
• Gage R & R
• Types of Sampling
• Process Capability
• Four Block Diagram
The Aim of the measure stage is to :

a. To Establish the validity of the measuring system & operator
b. It tells the present Level of the Process.

Six Sigma is based on the measured data. There will be unfavorable consequences
from analysis using statistical tool if we have a problem with measuring system.
What’s more, the process gets worse, then experiment will end up in failure.
Therefore, we do better secure correct measurement system before the project.
Overall Variation
Part to Part Variation Measurement System Variation
Variation due to gage Variation due to Operator
Repeatability Reproducibility
Operator Operator by Part

Gage R & R - Introduction Understanding Six Sigma
σ2Total = σ2Part-Part + σ2R&R
Total variation Variation due to Measurement error variation

differences among the parts.
σ2R&R = σ2Repeatibility + σ2Reproducibility
Measurement error variation Variation due to Variation due to

Gage Operator
Gage R & R - Introduction Understanding Six Sigma
What is Gage R&R
Gage R & R is Gage Repeatability & Reproducibility.
◆ Repeatability = EV(Equipment Variation)

It is the variation observed in the system when one Operator measures the same part twice
using the same gage. Repeatability is the variation due to equipment variation.
Example : Consider one Operator who successively measures the Thickness of paint coating
on Ref L/R part using the same gauge.
Reading 1 = 12.5 , Reading 2 = 12.0 , Difference = 0.5 , variation due to Gage (Repeatability)
◆ Reproducibility = AV(Appraiser Variation)
It is the variation observed in the system when two different operators measures the same part
using the same gage. Reproducibility is the variation due to change in operator.
Example : Consider two Operators who successively measures the Thickness of paint coating
on same Ref L/R part using the same gauge.
Operator 1 = 14.8 , Operator 2 = 14.2 , Difference = 0.6 , variation due to Operator
(Reproducibility )
Total Gage R&R = E.V2 +A.V2

Gage R &R - Long Study Method Understanding Six Sigma
Repeatability ? Repeatability :
“Getting consistent results”
☞ Variation observed with one measurement
device when used several times by one operator
while measuring the identical characteristic on
the same parts.
Measure/Re-measure variation
Reproducibility ?
Operator B
Variation obtained from different operators
using the same device when measuring the Operator C
identical characteristic on the same parts.
Operator A
Reproducibility
Gage R & R - Purpose Understanding Six Sigma
Gage R&R
• Gage R & R is used to ensure that the measured data used
for statistical tests is valid.
• Selection of the most appropriate gage for the task.
• When we want to compare the performance of each Gage.
• When we want to exclude the Gage error from results.
• Maintenance of measurement system ( Calibration )
• For measurement training for existing and New Staff.

Measure ~ Gage R & R Understanding Six Sigma
Two types of Gage R&R Study

◆ Short study method
• Requires minimum 2 operators and minimum 5 parts with each part measured at
least once.
• This method cant separate the total variation Observed through Gage R &R into
repeatability & reproducibility
• Permits speedy acceptance for adapting Gauge.
◆ Long study method
• Requires minimum 2 Operators, minimum 10 parts with each part measured at

least twice.
• This method can divide the total variation observed in the system through Gage
R & R into repeatability & reproducibility, so that we can get to know what we have
to improve Operator or Gage
Gage R & R ~ Short Study Understanding Six Sigma
Short Study Method

Example : The height of a component has specifications given by 5.0 ± 0.5. (tolerance
= 1.0 )
Solution : The Measurements taken by the two Operators for the Five Parts are listed
below Part Operator 1 Operator 2 Ranges (1-2)
1 4.9 4.8 0.1
2 4.7 4.7 0
3 5.2 5.1 0.1
4 5 5.1 0.1
5 4.8 4.7 0.1
Range Sum 0.4
• Average Range ( R-bar ) = Σ R / n = 0.4 / 5 = 0.08

• Gauge Error = ( 6.0 /d) ( R-bar) = (5.15 /1.19) (0.08 ) = 0.3464
5.15 indicates the Confidence Level of 99 % ; 6.0 indicates a Confidence Level of 99.73
%
•Gauge R & R as % of Tolerance = (0.3464 x 100) /1.0 = 34.64 %
Gauge Error is calculated by multiplying the average range by a constant d ( to be
taken from the Table )
Gage R & R ~ Short Study Understanding Six Sigma
d* values for distribution of the average range

Number of parts Number of operators
2 3 4 5
1 1.41 1.91 2.24 2.48
2 1.28 1.81 2.15 2.40
3 1.23 1.77 2.12 2.38
4 1.21 1.75 2.11 2.37
5 1.19 1.74 2.10 2.36
6 1.18 1.73 2.09 2.35
7 1.17 1.73 2.09 2.35
8 1.17 1.72 2.08 2.35
9 1.16 1.72 2.08 2.34
10 1.16 1.72 2.08 2.34
Gage R & R ~ Guidelines Understanding Six Sigma
Pre Requisites for Gage R & R
★ Blind Test :
• The Operator should not be aware that Gage R&R is going On.
• The Previous Readings should not be conveyed while taking Next Reading.
★ Gage selection(Resolution)
•The Gage must have a resolution of less than or equal to 10% of the one sided
specification or process variation.
• Resolution is the smallest unit of measure the gage is able to read.
•Ex) In case of part feature tolerance equals +/-0.020, Gage must have resolution
0.002 and Gage R&R ≤20% to be recommended.
★ Intentional Sampling :
• The samples must not be randomly selected, the sampling must be proceeded
by a plan, so the total range of variation and specification are covered.
• Most values should lie near the LSL/USL , because the chances of discrepancy
are more near these limits
Gage R &R ~ Long Study Method Understanding Six Sigma
An acceptable value for a Gage R&R Study (Continuos Data - )
≤ 20% : Acceptable
% Study Variation & % Study Tolerance 20% to 29% : Conditional
≥ 30% : Unacceptable
An improvement plan to lower the gauge R&R variation should be implemented. If there
is no improvement , consideration should be made for the risks associated with high
Gauge R &R
Number of Distinct Categories >4
P Value of the Operator * Part > 0.25

Input : Parts, Operator

Long study method (using Minitab) & Measurement data
Select: ANOVA
Why ANOVA method is more accurate than X(bar) R Method…..????????
X - R Method ANOVA
Part to Part Variation Part to Part Variation
Repeatability Repeatability
Reproducibility Reproducibility
Operator by Part
Operator
ANOVA Method further breaks the Variation due to Operator(Reproducibility)

into Operator & Operator by Part
Long study method (using Minitab)

If significant, P-value < 0.25 indicates
Gage R&R Study - ANOVA Method
that an operator is having a problem
measuring some the parts. Hence Gage
Two-Way ANOVA Table With Interaction R&R is not acceptable.
Source DF SS MS F P
Parts 9 81.6 9.06667 204.000 0.000
Operator 1 0.1 0.10000 2.250 0.168
Parts * Operator 9 0.4 0.04444 0.889 0.552
Repeatability 20 1.0 0.05000
Total 39 83.1
Two-Way ANOVA Table Without Interaction
Source DF SS MS F P
Parts 9 81.6 9.06667 187.810 0.000
Operator 1 0.1 0.10000 2.071 0.161
Repeatability 29 1.4 0.04828
Total 39 83.1
Long study method (using Minitab)

% Study Variation < 20 % and % Study
Source VarComp (of VarComp)
Total Gage R&R 0.05086 2.21
Tolerance> 20%, Gage R&R is acceptable
Repeatability 0.04828 2.09
Reproducibility 0.00259 0.11
Operator 0.00259 0.11
Part-To-Part 2.25460 97.79
Total Variation 2.30546 100.00
Study Var %Study Var %Tolerance

Source StdDev (SD) (6 * SD) (%SV) (SV/Toler)
Total Gage R&R 0.22553 1.35316 14.85 33.83
Repeatability 0.21972 1.31831 14.47 32.96
Reproducibility 0.05085 0.30513 3.35 7.63
Operator 0.05085 0.30513 3.35 7.63
Part-To-Part 1.50153 9.00919 98.89 225.23
Total Variation 1.51837 9.11024 100.00 227.76
For Gage R&R to be acceptable, number of distinct categories > 4

Number of Distinct Categories = 9
σ Gage R&R * 100

% Study Variation =
σ total variation
σ Gage R&R * 100

% Study Tolerance = 6.0 *
σ total variation
Gage R & R( Nested ) : Used for Destructive Testing

Gage R & R (Nested ) is used when each part is measured once only as
when measuring.
Ex. The torque release of a bolt during QC sampling, we cannot measure again
Accuracy ? True Accuracy

The degree of agreement of the (Reference)
value
measured value to the true magnitude
(unbiased values).
(Accuracy is typically expressed as 1-%Bias)
* Setting a true value is a one that Observed average

is measured by the most accurate measuring
device.
Stability ? Time 1
Stability is the total variation in the measurements obtained with a

measurement system on the same master or reference value
when measuring the same characteristic over an extended time
period. Stability
Time 2
Bias : It is a measure of the distance between the average value the

measurements and the "True" or "Actual" value of the sample or part.
Linearity ? LSL USL
Actual
values
Linearity is the difference in Actual
the bias values throughout the values
expected operating range of the
gage.
(Gage is less accurate at the low (No Bias)
Reference value
end of specification or operating
range than at the high end).
Reference values
Larger Bias Small Bias

Measure - 4 Block Diagram Understanding Six Sigma
Block Diagram Poor

2.5
Zshift
2.0 A B
(Process 1.5
Control)
1.0
C D
0.5
Good
1 2 3 Z.st 4 5 6
Poor Good
(Process Technology )
A : Poor control, poor technology

B : Must control the process better, technology is fine
C : Process control is good, poor technology
D : World Class
Gage R & R - Discreet Data Understanding Six Sigma
Pre-Requisites Gage R & R (discreet data)
• The Minimum Number of Samples should be at least 20
• Minimum Number of Operators should be at least 2
• Each Operator must take at least two readings of each Part.
Acceptability of Gage R & R (discreet data)

• % Gage R & R should be less than 5 %
Gage R & R - Discreet Data Understanding Six Sigma
Gage R & R (discreet data)

Visual Inspection Gage Study
•The gage is acceptable if both the
Appraiser "A" Appraiser "B"
1 2 1 2 Appraisers (four per part) agree..
1 G G G G • % Gage R&R = No. of
2 G G G G
3 NG G G G Disagreements/Total Opportunities X100
4 NG NG NG NG
18 Samples
5 G G G G = 3 / 20 x 100% = 15%
6 G G G G
7 NG NG NG NG
• If the results of checkers are different,
8 NG NG G G the gage
9 G G G G
10 G G G G • must be improved and re-evaluated.
11 G G G G
12 G G G G • If the gage cannot be improved, it is
13 G NG G G
14 G G G G
unacceptable and an alternate
15 G G G G measurement system should be found.
16 G G G G
17 G G G G
18 G G G G
19 NG NG NG NG
20 G G G G
Gage R & R - Importance Understanding Six Sigma
From the Gauge R & R Study we can determine the Following :
• Gage resolution is adequate.
• The Measurement System is stable over time.
• The measurement system error is small enough and

acceptable enough relevant to the process variation or
Specification
Gage R & R Indicates that whether the Measurement System is good

enough for the collection of Data.
Measure - Capability Analysis Understanding Six Sigma
Capability analysis is a set of calculations used to assess whether a system is

statistically able to meet a set of specifications or requirements. Specifications or
requirements are the numerical values within which the system is expected to operate, that
is, the minimum and maximum acceptable values. Specifications are numerical
requirements, goals, aims, or standards.

Cp = Specification Width Cp = USL-- LSL

Process Capability 6σwithin
Cp = Product Specification Cpk = Cp (k-1)

Manufacturing Variability
k = T-µ CPU = USL - X CPL = X - LSL

USL - LSL 3σwithin 3σwithin
2
Process Capability ( Cp) is the Tolerance width in Relation to the process

capability, expressed as the best short Term performance. Takes no account of
the process centering.
Capability Index (Cpk) accounts for the process centering. Considers

sample data variation & location simultaneously.
Measure - Central Limit Theorem Understanding Six Sigma
Central limit theorem states that as the sample size increases, the sampling
distribution of the mean will approach normality. Statisticians use the normal distribution as
an approximation to the sampling distribution, whenever the sample size is at least 30.
SEM : Standard Error of Mean
Standard error of mean gives the difference between
the standard deviation of Population & Standard
deviation of sample.
SEM *
SEM = σp Standard deviation of

(n)1/2 Population
Sample Size
n=8 n=30
Sample Size
Measure - Central Limit Theorem Understanding Six Sigma
From the graph shown on the previous slide, it is evident that for sample size 30 the difference
between the standard deviations of sample and population is very less. Even though this
difference reduces further by increasing sample size, but this reduction is negligible. Hence
while sampling, sample size of 30 is considered as the idle sample size.
Sample Size Difference b/t Standard deviation of

Population and Sample
• Less than 8 high variation

• Between 8 ~ 30 Moderate Variation
• 30 & above Minimum Variation
As the sample size increases, above 8 samples the difference between standard
deviation of Population & sample reduces drastically. At sample size 30 the difference
is minimum and it remains constant & beyond 30 it remains constant., so the curve
line representing the difference becomes parallel to X-axis.
Measure - Sampling Understanding Six Sigma
Types of Sampling
1.) Random Sampling : In this type of Sampling each data point of the Population
has an equal chance/ Probability of being selected.
Example : During the draw of lottery tickets each & every lottery ticket number has an
equal chance of winning the Prize.
2.) Stratified Sampling : In this type of Sampling , the sub group taken for sampling
has data points of same type.
Example : For determining the Quality of Food in the Canteen, if we take the Sample
group in which all Supervisors/Operators/Managers are there, then the difference in
the variation of taste within the sub group would be minimum but among the Sub
groups would be maximum.
3) Clustered Sampling : In this type of Sampling,each & every type of data point
present in the population would be covered in the Sample.
Example : In the above example if take the sample in such a way that in the subgroup
operator, supervisor and manager are taken so the difference in the taste would be
maximum within the sub group and minimum among the subgroup.
BLACK NOISE
(Signal)
RATIONAL
SUBGROUPS
PROCESS RESPONSE
WHITE NOISE
(Common Cause
variation)
TIME
White Noise
• White noise represents the variation present in every process. Also known as common
cause variation
• It is not controllable variation within the existing technology.
• Represents that best the process can be with the present technology(Inherent process
capability).
Black Noise
• Black Noise represents the outside influences on a process that cause average to shift and
drift. Also known as Special Cause or assignable cause variation.

• It is potentially controllable variation with the existing process technology.
• It represents how the process is actually performing over time(Sustained process capability).
From a statistics perspective, There are only two problems.
Problem with Spread Problem with Centering
Desired
Desired
Current
Current situation
situation
LSL T USL LSL T USL
Shift
Accurate but not Precise Precise but not Accurate

Process Capability Ratios
The greater the design margin, the lower the Total Defects Per Unit
Design margin is measured by the Process Capability Index (Cp)
Maxium Allowable Range of

Characteristic
Cp = USL - LSL
Normal variation of Process X Cp =
-3σ +3σ + 3σ
Process Width
Zst = 3 Cp Design Width
Zlt = 3Cpk
Is it Control or Technology?
Long Term Data Short Term Data
. Data taken over a period of time . Data taken over a short enough
long enough that external factors period of time that there are no
can influence the process. external influences on the
process Z st : Z lt Technology:
Control
. Z lt (σlt ) Z st (σst )
Zlt is always less tahn Zst,
because the long term value is
. Cpk . Cp reduced by the shift of the
process
. Defined by technology and process . Defined by technology
control . Process Capability
. Process Performance (Entitlement - The best process
can be)
6σ means Zlt=4.5 and Cpk=1.5 6σ means Zst=6.0 and Cp=2.0

Analyze Understanding Six Sigma
Analyze Phase
• Cause & Effect Diagram
• Hypothesis Testing
• Mean Testing
• Variance Testing
• Regression Analysis
The Aim of the Analyze Phase is to :

a. To List down all the Possible factors through Brainstorming.
b. Pick out the Vital few Potential Factors out of Trivial Many.
c. Statistical verification of Potential Factors by means of various type of Tests.
d. Ascertaining whether we have considered all factors & nothing is Left Out.
Analyze - Cause & Effect Diagram Understanding Six Sigma
Cause & Effect Diagram
Cause
Cause Cause Cause
Effect
Cause
Cause
Cause
The Purpose of this tool is to Find out the start of the collection of Data and analysis.
It list down all the Probable causes responsible for the main effect . Cause & Effect
Diagram is also known as Fishbone Diagram / 4M Diagram.
The Symptom or result is put under the Dark Box on the Right.. Lighter Boxes at the end of
the Large Bones are main groups in which ideas are classified. The Lighter Boxes may
consist of Five Ms - Man,Machine,Measurement & Method.(Money can be considered
wherever relevant) . The Middle Bones indicates the direction of path from cause to effect
Machine
Man
Cylinder
New casual Failure
Die
Handling Setting
problem M/C not
clean OUT CASE
DENT
Dented sheet
Chips on Piece
sheet Piece
Sheet unloading
check
thickness
Material Method
Analyze - Hypothesis Understanding Six Sigma
Hypothesis Testing
Hypothesis means something taken to be true for the Purpose of argument or
Investigation , an assumption.
Hypothesis Testing is defined as the comparison of two Populations (equality of mean/
variance ) by taking samples from those Populations. It is assumed in the beginning that the
two Populations are equal (Null Hypothesis ;µ1 µ2 ; σ1 = σ2 ) or not equal (Alternate
Hypothesis : µ1≠ µ2 ; σ1 ≠ σ2 ) . The equality is confirmed by actually conducting tests on
the sample. There is always a risk associated with the Hypothesis , in case the sample
taken for comparison from Population does not correctly represent the Population..
There are many types of hypothesis test. The test is selected depending on the type of
data or the comparison required.
Continuous Data Discrete Data
1) F-test : Compares Variances 3) Chi Square Test : Compares counts
• Levene’s Test • Goodness of Fit
• Bartlett’s Test • Contingency Table
2) t-test : Compares means
• 1 sample t-test • Paired t-test
• 2 sample t-test
Hypothesis Testing
Ho(Null Hypothesis) is assumed to be true .This is like the defendant being
assumed to be innocent.
Ha(Alternative Hypothesis) is alternatives the Null Hypothesis. Ha is the one
that must be proved.
Population
Ho Ha
In this case as the samples

In this case the samples correctly
Correct Type 2 does not correctly represent
represent the Population
so sample mean = Population mean.
Ho Decision Error the Population so sample
β mean ≠ Population mean.
Sample
Incorrect Decision
In this case as the samples does not In this case as the samples
correctly represent the Population Type 1 correctly represent the Population
Ha Error
so sample mean ≠ Population mean. so sample mean = Population
Incorrect Decision α mean.
Correct Decision
Important Terms
1.) Type 1 Error : This error gives us the probability of rejecting the Right Material .
This happens when a weird sample gets selected for the comparison of
mean/variance. It is also known as α − Error or Producer’s Risk. Generally It’s value
lies around 5 %.
2. ) Type 2 Error : This error gives us the probability of accepting the wrong
material. This also happens when a weird sample is selected for comparison. It is also
known as β − Error or Consumer’s Risk. It’s value generally lies around 10 %.
3 .) 1-α = Confidence of the Test
The probability that can be determined as a right thing when the Null
Hypothesis is correct.
4) 1-β = Power of the test

The rejecting probability when null Hypothesis you want to test is not
right.
It is not possible to simultaneously commit a Type 1 and Type 2 decision
error.
Analyze - Tests used for Comparison Understanding Six Sigma
Analyze
Mean Testing Variance Testing
Continuous Data Discrete Data Test for Equal

Variance
1 Sample 1 Proportion
Z Test Test 2 Variance
Test
1 Sample 2 Proportion
t Test Test
2 Sample Chi-Square
t Test Test
ANOVA
Testing
1 Sample Z Test :computes a confidence interval or performs a hypothesis test of the

mean when the population standard deviation, σ is known. This procedure is based upon
the normal distribution. This test compares the mean of the sample with some test
Population with known standard deviation.
Example : Measurements were made on nine widgets. You know that the distribution of
measurements has historically been close to normal with s = 0.2. Because you know s, and
you wish to test if the population mean is 5 and obtain a 90% confidence interval for the
mean, you use the Z-procedure.
Solution : Values
4.9
1 Open the worksheet enter the values.. 5.1
4.6
2 Choose Stat > Basic Statistics > 1-Sample Z. 5
5.1
3 In Samples in Columns, enter Values.
4.7
4 In Standard deviation, enter 0.2. 4.4
4.7
5 In Test mean, enter 5. 4.6
6 Click Options. In Confidence level, enter 90. Click OK.

7 Click Graphs. Check Individual value plot. Click OK in each dialog box.
One-Sample Z: Values Test Mean doesn’t lie

within the confidence p value < 0.05,
Test of mu = 5 vs not = 5 Interval Hence Ha, alternate
Hypothesis
The assumed standard deviation = 0.2
Variable N Mean StDev SE Mean 90% CI Z P
Values 9 4.78889 0.24721 0.06667 (4.67923, 4.89855) -3.17 0.002
Interpreting the results

The test statistic, Z, for testing if the population mean equals 5 is -3.17. The p-value, or the
probability of rejecting the null hypothesis when it is true, is 0.002. This is called the attained
significance level, p-value, or attained α of the test. Because the p-value of 0.002 is smaller
than commonly chosen α-levels, there is significant evidence that m is not equal to 5, so you
can reject H0 in favor of m not being 5.
A hypothesis test at α = 0.1 could also be performed by viewing the individual value plot. The
hypothesized value falls outside the 90% confidence interval for the population mean
(4.67923, 4.89855), and so you can reject the null hypothesis.
1 Sample t test : computes a confidence interval or performs a hypothesis test of the

mean when Population standard , σ is unknown. This procedure is based upon the t-
distribution, which is derived from a normal distribution with unknown σ.
Example : Measurements were made on nine widgets. You know that the distribution of
widget measurements has historically been close to normal, but suppose that you do not
know σ. To test if the population mean is 5 and to obtain a 90% confidence interval for the
mean, you use a t-procedure.
Solution : Values
4.9
1 Open the worksheet enter the data. 5.1
2 Choose Stat > Basic Statistics > 1-Sample t. 4.6
5
3 In Samples in columns, enter Values. 5.1
4.7
4 In Test mean, enter 5. 4.4
4.7
5 Click Options. In Confidence level enter 90. Click OK in each dialog box 4.6
One-Sample T: Values
Test of mu = 5 vs not = 5
Variable N Mean StDev SE Mean 90% CI T P
Values 9 4.78889 0.24721 0.08240 (4.63566, 4.94212) -2.56 0.034
Result Interpretation :
The p-value < 0.05 , also “ 0 “ does not lie within the Confidence Interval so Null Hypothesis
is rejected and Alternate Hypothesis is accepted. It confirms that the sample mean is not
equal to Population Mean ).
2 Sample t test : computes a confidence interval and performs a hypothesis test of the
difference between two population means when σ 's are unknown and samples are drawn
independently from each other. This procedure is based upon the t-distribution, and for small
samples it works best if data were drawn from distributions that are normal or close to
normal. You can have increasing confidence in the results as the sample sizes increase.
Example : 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). The energy
consumption data (BTU.In) are stacked in one column with a grouping column (Damper)
containing identifiers or subscripts to denote the population. Suppose that you performed a
variance test and found no evidence for variances being unequal .Now you want to compare
the effectiveness of these two devices by determining whether or not there is any evidence
that the difference between the devices is different from zero.
BTU.In Damper
7.87 1
9.43 1
Solution :
7.16 1
8.67 1
12.31 1
9.84 1
1 Open the worksheet , enter the data. 16.9

10.04
1
1
12.62 1
7.62 1
2 Choose Stat > Basic Statistics > 2-Sample T. 11.12

13.43
9.07
1
1
1
6.94 1
3 Choose Samples in one column. 10.28

9.37
7.93
1
1
1
13.96 1
4 In Samples, enter 'BTU.In'.

6.8 1
4 1
8.58 1
8 1
5.98 1
5 In Subscripts, enter Damper. 15.24
8.54
1
1
11.09 1
11.7 1
6 Check Assume equal variances. Click OK. 12.71

6.78
1
1
9.82 1
12.91 1
10.35 1
9.6 1
9.58 1
9.83 1
9.52 1
18.26 1
10.64 1
6.62 1
5.2 1
12.28 2
7.23 2
2.97 2
8.81 2
9.27 2
11.29 2
8.29 2
9.96 2
10.3 2
16.06 2
14.24 2
11.43 2
10.28 2
13.6 2
5.94 2
10.36 2
6.85 2
6.72 2
10.21 2
8.61 2
Minitab Output :
Two-Sample T-Test and CI: BTU.In, Damper
Two-sample T for BTU.In
Damper N Mean StDev SE Mean
1 40 9.91 3.02 0.48
2 50 10.14 2.77 0.39
Difference = mu (1) - mu (2)
Estimate for difference: -0.235250
95% CI for difference: (-1.450131, 0.979631)
T-Test of difference = 0 (vs not =): T-Value = -0.38 P-Value = 0.701 DF = 88
Both use Pooled StDev = 2.8818
Result Interpretation :
Minitab displays a table of the sample sizes, sample means, standard deviations, and
standard errors for the two samples.
Since we previously found no evidence for variances being unequal, we chose to use the
pooled standard deviation by choosing Assume equal variances. The pooled standard
deviation, 2.8818, is used to calculate the test statistic and the confidence intervals.
A second table gives a confidence interval for the difference in population means. For this
example, a 95% confidence interval is (-1.45, 0.98) which includes zero, thus suggesting that
there is no difference. Next is the hypothesis test result. The test statistic is -0.38, with p-
value of 0.701, and 88 degrees of freedom.
Since the p-value is greater than commonly chosen a-levels, there is no evidence for a
difference in energy use when using an electric vent damper versus a thermally activated
vent damper.
ANOVA : is a tool with which we can compare several means. It is a tool used to search for
the significant X factors that have an influence on the response variable Y. In effect, analysis
of variance extends the two-sample t-test for testing the equality of two population means to
a more general null hypothesis of comparing the equality of more than two means, versus
them not all being equal.
Example : You design an experiment to assess the durability of four experimental carpet
products. You place a sample of each of the carpet products in four homes and you measure
durability after 60 days. Because you wish to test the equality of means and to assess the
differences in means, you use the one-way ANOVA procedure (data in stacked form) with
multiple comparisons. Generally, you would choose one multiple comparison method as
appropriate for your data. However, two methods are selected here to demonstrate Minitab's
capabilities.
Solution :
1 Open the worksheet enter the data
2 Choose Stat > ANOVA > One-Way.
3 In Response, enter Durability. In Factor, enter Carpet.
4 Click Comparisons. Check Tukey's, family error rate. Check Hsu's MCB, family error rate
and enter 10. Durability Carpet
18.95 1
5 Click OK in each dialog box. 12.62 1
11.94 1
14.42 1
10.06 2
7.19 2
7.03 2
14.66 2
10.92 3
13.28 3
14.52 3
12.51 3
10.46 4
21.4 4
18.1 4
22.5 4
Results for: EXH_AOV.MTW
One-way ANOVA: Durability versus Carpet
Source DF SS MS F P
Carpet 3 146.4 48.8 3.58 0.047
Error 12 163.5 13.6
Total 15 309.9
S = 3.691 R-Sq = 47.24% R-Sq(adj) = 34.05%
Individual 95% CIs For Mean Based on

Pooled StDev
Level N Mean StDev ---------+---------+---------+---------+
1 4 14.483 3.157 (-------*-------)
2 4 9.735 3.566 (-------*--------)
3 4 12.808 1.506 (-------*-------)
4 4 18.115 5.435 (-------*-------)
---------+---------+---------+---------+
10.0 15.0 20.0 25.0
Tukey 95% Simultaneous Confidence Intervals

All Pairwise Comparisons among Levels of Carpet
Individual confidence level = 98.83%
Carpet = 1 subtracted from:
Carpet Lower Center Upper ------+---------+---------+---------+---
2 -12.498 -4.748 3.003 (------*-------)
3 -9.426 -1.675 6.076 (------*-------)
4 -4.118 3.632 11.383 (-------*------)
------+---------+---------+---------+---
-10 0 10 20 Carp
= 2 subtracted from:
3 -4.678 3.073 10.823 (-------*-------)
4 0.629 8.380 16.131 (------*-------)
------+---------+---------+---------+---
-10 0 10 20 Carp
= 3 subtracted from:

4 -2.443 5.308 13.058 (------*-------)
------+---------+---------+---------+---
-10 0 10 20
Tukey's comparisons
Tukey's test provides 3 sets of multiple comparison confidence intervals:
Carpet 1 mean subtracted from the carpet 2, 3, and 4 means: The first interval in the first set
of the Tukey's output (-12.498, -4.748, 3.003) gives the confidence interval for the carpet 1
mean subtracted from the carpet 2 mean. You can easily find confidence intervals for entries
not included in the output by reversing both the order and the sign of the interval values. For
example, the confidence interval for the mean of carpet 1 minus the mean of carpet 2 is
(-3.003, 4.748, 12.498). For this set of comparisons, none of the means are statistically
different because all of the confidence intervals include 0.
Carpet 2 mean subtracted from the carpet 3 and 4 means: The means for carpets 2 and 4
are statistically different because the confidence interval for this combination of means
(0.629, 8.380, 16.131) excludes zero.
Carpet 3 mean subtracted from the carpet 4 mean: Carpets 3 and 4 are not statistically
different because the confidence interval includes 0.
By not conditioning upon the F-test, differences in treatment means appear to have occurred
at family error rates of 0.10. If Hsu's MCB method is a good choice for these data, carpets 2
and 3 might be eliminated as a choice for the best. When you use Tukey's method, the mean
durability for carpets 2 and 4 appears to be different.
1 Proportion Test : Performs a test of one binomial proportion.

Use 1 Proportion to compute a confidence interval and perform a hypothesis test of the
proportion. For example, an automotive parts manufacturer claims that his spark plugs are
less than 2% defective. You could take a random sample of spark plugs and determine
whether or not the actual proportion defective is consistent with the claim. For a two-tailed
test of a proportion:
H0: p = p0 versus H1: p ≠ p0 where p is the population proportion and p0 is the hypothesized
value.
Example : 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. You need to test H0: p = .65 versus
H1: p > .65.
As her campaign manager, you collected data on 950 randomly selected party members
and find that 560 party members support the candidate. A test of proportion was
performed to determine whether or not the proportion of supporters was greater than the
required proportion of 0.65. In addition, a 95% confidence bound was constructed to
determine the lower bound for the proportion of supporters.
1 Choose Stat > Basic Statistics > 1 Proportion.
2 Choose Summarized data.
3 In Number of trials, enter 950. In Number of events, enter 560.
4 Click Options. In Test proportion, enter 0.65.
5 From Alternative, choose greater than. Click OK in each dialog box.

Session window output
Test and CI for One Proportion
Test of p = 0.65 vs p > 0.65
95%
Lower Exact
Sample X N Sample p Bound P-Value
1 560 950 0.589474 0.562515 1.000
The p-value of 1.0 suggests that the data are consistent with the null hypothesis (H0: p =
0.65), that is, 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 : Performs a test of two binomial proportions.

Use the 2 Proportions command to compute a confidence interval and perform a
hypothesis test of the difference between two proportions. For example, suppose you
wanted to know whether the proportion of consumers who return a survey could be
increased by providing an incentive such as a product sample. You might include the
product sample with half of your mailings and see if you have more responses from the
group that received the sample than from those who did not. For a two-tailed test of two
proportions:
H0: p1 - p2 = p0 versus H1: p1 - p2 ≠ p0
where p1 and p2 are the proportions of success in populations 1 and 2, respectively, and
p0 is the hypothesized difference between the two proportions.
Example : 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.
1 Choose Stat > Basic Statistics > 2 Proportions.
2 Choose Summarized data.
3 In First sample, under Trials, enter 50. Under Events, enter 44.
4 In Second sample, under Trials, enter 50. Under Events, enter 42. Click OK.
Test and CI for Two Proportions

Sample X N Sample p
1 44 50 0.880000
2 42 50 0.840000
Difference = p (1) - p (2)
Estimate for difference: 0.04
95% CI for difference: (-0.0957903, 0.175790)
Test for difference = 0 (vs not = 0): Z = 0.58 P-Value = 0.564
Since the p-value of 0.564 is larger than commonly chosen a levels, the data are consistent
with the null hypothesis (H0: p1 - p2 = 0). That is, 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.
You can make the same decision using the 95% confidence interval. Because zero falls in
the confidence interval of (-0.096 to 0.176) you can conclude that the data are consistent
with the null hypothesis. If you think that the confidence interval is too wide and does not
provide precise information as to the value of p1 - p2, you may want to collect more data in
order to obtain a better estimate of the difference.
Chi Square Test : It is a measure of the Observed & expected frequencies. Chi
Square test is a statistical test which consists of three different type of Analysis.
1) Goodness of Fit
2) Test for Homogeneity
3) Test for Independence
The test for Goodness of fit determines if the sample under analysis was drawn
from a population that follows some specified distribution .
Test for Homogeneity answers the proposition that several populations are
homogenous with respect to some characteristic.
Test for Independence is for testing Null hypothesis that two criteria of
Classification
Example : You are interested in the relationship between gender and political party
affiliation. You query 100 people about their political affiliation and record the number of
males (row 1) and females (row 2) for each political party. The worksheet data appears as
follows:
Column 1 Column 2 Column 3
Democrat Republican Other
28 18 4
22 27 1
1 Open the worksheet EXH_TABL.MTW.
2 Choose Stat > Tables > Chi-Square Test (Table in Worksheet).
3 In Columns containing the table, enter Democrat, Republican and Other. Click OK.
Chi-Square Test: Democrat, Republican, Other
Expected counts are printed below observed counts
Chi-Square contributions are printed below expected counts
Democrat Republican Other Total
1 28 18 4 50
25.00 22.50 2.50
0.360 0.900 0.900
2 22 27 1 50
25.00 22.50 2.50
0.360 0.900 0.900
Total 50 45 5 100
Chi-Sq = 4.320, DF = 2, P-Value = 0.115
2 cells with expected counts less than 5.
Chi-Square Test: Democrat, Republican, Other
Expected counts are printed below observed counts
Chi-Square contributions are printed below expected counts
Democrat Republican Other Total Observed Values
1 28 18 4 50 Expected Values
25.00 22.50 2.50
0.360 0.900 0.900 Chi Square Values
2 22 27 1 50
25.00 22.50 2.50 Row Totals
0.360 0.900 0.900

Total 50 45 5 100 Grand Total
Chi-Sq = 4.320, DF = 2, P-Value = 0.115
Column Totals
2 cells with expected counts less than 5.
No evidence exists for association (p = 0.115) between gender and political party affiliation.
Of the 6 cells, 2 have expected counts less than five (33%). Therefore, even if you had a
significant p-value for these data, you should interpret the results with caution. To be more
confident of the results, repeat the test, omitting the Other category.
Formulae’s :
a) Expected Value = Row Total * Column Total

Grand Total
b) Chi Square Value = ( Observed Value - Expected Value )2

Expected Value
c) Degrees of Freedom(DF) = (No. of rows - 1) * ( No. of Columns - 1)

2 Variance Test : Use to perform hypothesis tests for equality, or homogeneity, of

variance among two populations using an F-test and Levene's test. Many statistical
procedures, including the two sample t-test procedures, assume that the two samples
are from populations with equal variance. The variance test procedure will test the
validity of this assumption.

Example : 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). The energy
consumption data (BTU.In) are stacked in one column with a grouping column (Damper)
containing identifiers or subscripts to denote the population. You are interested in comparing
the variances of the two populations so that you can construct a two-sample t-test and
confidence interval to compare the two dampers.
1 Choose Stat > Basic Statistics > 2 Variances.
2 Choose Samples in one column.
3 In Samples, select the column which contain the values
4 In Subscripts, enter Damper. Click OK.

Test for Equal Variances for BTU.I n

F-Test
Test Statistic 1.19
1 P-Value 0.558
Damper
Levene's Test
Test Statistic 0.00
P-Value 0.996
2
2.0 2.5 3.0 3.5 4.0

95% Bonferroni Confidence I ntervals for StDevs
1
Damper
5 10 15 20
BTU.I n
Test for Equal Variances: BTU.In versus Damper
95% Bonferroni confidence intervals for standard deviations
Damper N Lower StDev Upper

1 40 2.40655 3.01987 4.02726
2 50 2.25447 2.76702 3.56416
F-Test (normal distribution)

Test statistic = 1.19, p-value = 0.558
Levene's Test (any continuous distribution)

Result Interpretation
The variance test generates a plot that displays Bonferroni 95% confidence intervals for the
population standard deviation at both factor levels. The graph also displays the side-by-side
boxplots of the raw data for the two samples. Finally, the results of the F-test and Levene's
test are given in both the Session window and the graph. Note that the 95% confidence level
applies to the family of intervals and the asymmetry of the intervals is due to the skewness of
the chi-square distribution.For the energy consumption example, the p-values of 0.558 and
0.996 are greater than reasonable choices of a, so you fail to reject the null hypothesis of the
variances being equal. That is, these data do not provide enough evidence to claim that the
two populations have unequal variances. Thus, it is reasonable to assume equal variances
when using a two-sample t-procedure.

Test for Equal Variance is used when comparing the variance of two or more than
two populations
Example : You study conditions conducive to potato rot by injecting potatoes with bacteria
that cause rotting and subjecting them to different temperature and oxygen regimes. Before
performing analysis of variance, you check the equal variance assumption using the test for
equal variances.
1Open the worksheet .

2 Choose Stat > ANOVA > Test for Equal Variances.
3 In Response, enter Rot.
4 In Factors, enter Temp Oxygen. Click OK.
Test for Equal Variances for Rot

Temp Oxygen
Bartlett's Test
2 Test Statistic 2.71
P-Value 0.744
10 6 Levene's Test
Test Statistic 0.37
P-Value 0.858
10
16 6
10
0 20 40 60 80 100 120 140

95% Bonferroni Confidence Intervals for StDevs
Test for Equal Variances: Rot versus Temp, Oxygen
95% Bonferroni confidence intervals for standard deviations
Temp Oxygen N Lower StDev Upper

10 2 3 2.26029 5.29150 81.890
10 6 3 1.28146 3.00000 46.427
10 10 3 2.80104 6.55744 101.481
16 2 3 1.54013 3.60555 55.799
16 6 3 1.50012 3.51188 54.349
16 10 3 3.55677 8.32666 128.862
Bartlett's Test (normal distribution)

Levene's Test (any continuous distribution)

Test for Equal Variances: Rot versus Temp, Oxygen

The test for equal variances generates a plot that displays Bonferroni 95% confidence
intervals for the response standard deviation at each level. Bartlett's and Levene's test
results are displayed in both the Session window and in the graph. Note that the 95%
confidence level applies to the family of intervals and the asymmetry of the intervals is
due to the skewness of the chi-square distribution.
For the potato rot example, the p-values of 0.744 and 0.858 are greater than reasonable
choices of a, so you fail to reject the null hypothesis of the variances being equal. That is,
these data do not provide enough evidence to claim that the populations have unequal
variances.
Analyze - Regression Understanding Six Sigma
One Variable Regression with Minitab

Example:
You are trying to optimize the performance of an paint cure oven.
One theory says that blower fan velocity affects evaporation of solvent in the paint.
You are trying to prove that such a relationship exists by analyzing the data below.
The Concept of Regression
A mathematical equation of describing a relationship between the ”Y”

and “X’s”
→Creating a Model of process
Y = b0 + b1x + error where b0 = constant b1 = slope
350
There appears to be a linear
relationship 300
Annual Sales
between floor space and annual
sales… 250
200
That is, Is the Annual sales reducing or
increasing according to change floor 50 100 150
space Floor Space

Regression analysis is used to investigate and model the

relationship between a response variable and one or
more predictors.
Use least squares procedures when your response variable is

continuous. Use partial least squares regression when your predictors
are highly correlated or outnumber your observations.· Use logistic
regression when your response variable is categorical.Both least
squares and logistic regression methods estimate parameters in the
model so that the fit of the model is optimized. Least squares regression
minimizes the sum of squared errors to obtain parameter estimates,
Example : You are a manufacturer who wants to obtain a quality measure on a product, but
the procedure to obtain the measure is expensive. There is an indirect approach, which uses
a different product score (Score 1) in place of the actual quality measure (Score 2). This
approach is less costly but also is less precise. You can use regression to see if Score 1
explains a significant amount of variance in Score 2 to determine if Score 1 is an acceptable
substitute for Score 2.
Score1 Score2
4.1 2.1
2.2 1.5
2.7 1.7
6.0 2.5
8.5 3.0
4.1 2.1
9.0 3.2
8.0 2.8
7.5 2.5
1 Choose Stat > Regression > Regression.

2 In Response, enter Score2.
3 In Predictors, enter Score1.
4 Click OK.
Regression Analysis: Score2 versus Score1

The regression equation is
Score2 = 1.12 + 0.218 Score1
Predictor Coef SE Coef T P
Constant 1.1177 0.1093 10.23 0.000
Score1 0.21767 0.01740 12.51 0.000
S = 0.127419 R-Sq = 95.7% R-Sq(adj) = 95.1%
The R2 value shows that Score 1 explains 95.7% of the variance in Score 2, indicating that the
model fits the data extremely well.
Analyze Phase
• 4M Diagram
• Hypothesis Testing
• Mean Testing
• Variance Testing
• Regression Analysis
The Aim of the Analyze Phase is to :

a. To List down all the Possible factors through Brainstorming.
b. Pick out the Vital few Potential Factors out of Trivial Many.
c. Statistical verification of Potential Factors by means of various type of Tests.
d. Ascertaining whether we have considered all factors & nothing is Left Out.
4 M Diagram
MACHINE
MAN
Cause
Cause Cause Cause
Effect
Cause
Cause
Cause
METHOD MATERIAL
4 M ( Man, Method, Machine & Material ) Diagram is used to list down all the Probable
factors (causes ) responsible for the Major Problem ( Effect ). After brainstorming the
Significant Factors are selected for further comparison ( Hypothesis Testing )
The Symptom or result is put under the Dark Box on the Right.. Lighter Boxes at the end of
the Large Bones are main groups in which ideas are classified. The Lighter Boxes consist
of Four Ms - Man,Method, Machine & Material. The Middle Bones indicates the direction of
path from cause to effect.
The test for equal variances generates a plot that displays Bonferroni 95% confidence
intervals for the response standard deviation at each level. Bartlett's and Levene's test
results are displayed in both the Session window and in the graph. Note that the 95%
confidence level applies to the family of intervals and the asymmetry of the intervals is
due to the skewness of the chi-square distribution.
For the potato rot example, the p-values of 0.744 and 0.858 are greater than reasonable
choices of a, so you fail to reject the null hypothesis of the variances being equal. That is,
these data do not provide enough evidence to claim that the populations have unequal
variances.
The Concept of Regression
A mathematical equation of describing a relationship between the ”Y”

and “X’s”
→Creating a Model of process
Y = b0 + b1x + error where b0 = constant b1 = slope
350
There appears to be a linear
relationship 300
Annual Sales
between floor space and annual
sales… 250
200
That is, Is the Annual sales reducing or
increasing according to change floor 50 100 150
space Floor Space

Regression analysis is used to investigate and model the relationship

between a response variable and one or more predictors.
Use least squares procedures when your

response variable is continuous.
Use partial least squares regression when
your predictors are highly correlated or
outnumber your observations..
Use logistic regression when your
response variable is categorical.
Both least squares and logistic regression
methods estimate parameters in the model
so that the fit of the model is optimized. Least
squares regression minimizes the sum of
squared errors to obtain parameter
estimates,
Example : Do regression and residual analysis for yield as shown in the table.Interpret the
output results. Please note that A,B,C are factors & yield is response.
S.No. A B C Yield
1 2 3 5 85
2 2 1 10 71
3 8 3 15 3129
4 6 4 20 1384
5 5 5 25 875
6 8 3 30 3159
7 5 1 35 823
8 3 2 40 254
9 2 2 45 150
10 1 8 50 298
11 9 7 55 4631
12 5 6 60 978
13 3 5 65 367
14 2 6 70 296
15 1 7 75 303
16 4 2 80 556
17 2 4 85 266
18 1 6 90 294
19 2 5 95 313
20 5 6 100 1058
Solution :
1.) Enter the columns A, B, C (Factors ) and Yield ( response ) in minitab Excel
sheet.
2) Go to stat> Regression > Regression.

3) Select Yield as Response and

A,B,C as Predictors by double
clicking on all.
4) Click OK.
Regression Analysis: Yield versus A, B, C
The regression equation is

Yield = - 1277 + 458 A + 136 B - 1.54 C
Predictor Coef SE Coef T P

R2 & R2 (adj) > 64 % indicating a strong
Constant -1277.3 360.6 -3.54 0.003 corelation between the Factors & the
A 457.70 47.64 9.61 0.000
B 135.72 61.28 2.21 0.042 Response (Yield)
C -1.544 4.579 -0.34 0.740
S = 487.537 R-Sq = 86.9% R-Sq(adj) = 84.5%

Analysis of Variance
Source DF SS MS F P
Regression 3 25308562 8436187 35.49 0.000
Residual Error 16 3803071 237692
Total 19 29111633
Source DF Seq SS
A 1 23966672
B 1 1314871
C 1 27018
Unusual Observations
Obs A Yield Fit SE Fit Residual St Resid

11 9.00 4631 3707 308 924 2.44R
R denotes an observation with a large standardized residual.

Minitab fits the regression Line using

the Least square method. As shown in
the diagram the least square method
Response
minimizes the sum of the squared
distances between the points and the
fitted Line.`
Predictors
Fitted Value : The predicted y or ; the mean response value for the given
predictor values using the estimated regression equation.
Residuals :The difference (ei) between the observed values and predicted or
fitted values (data minus fits). This part of the observation is not explained by the
fitted model. The formula for the residual of an observation is: ei = (yi - i)
YIELD = RESIDUAL + FITS

Three Regressions Models
Linear Y = bo + b1X
Quadratic Y = bo + b11X2
Cubic Y = bo + b1X + b11X2
Y is the response; X is the predictor; bo is the intercept; and b1, b11, and b111 are the
coefficients
Control Understanding Six Sigma
Control Phase
• Statistical Process Control
• Control Chart
AIM of Control Phase
Control Charts are used to track process statistics over Time and to detect
the presence of Special Causes.
• Provides structured closure of projects and re-allocation of resources
• Provides systematic changes to ensure the process continues in a new path of
optimization. It Transfers sustainability of the improvement to the appropriate
members of the Advocacy Team
• Provides communication of new procedures and systems to process owners
• Ensures that the new process conditions are documented and monitored
Statistical Process Control (SPC)?

(SPC)
Statistical
Statistical methods are used to monitor and analyze process variation from sample data
Process
Any repetitive (manual or automatic) task or steps
Control
Provides an early warning signal that a process has changed. The warning allows you to
make decisions about the process while there is still time to correct the problem before it
can be seen in the final output.
Six Sigma Quality focuses on moving control up stream in a process to leverage the input
characteristics for the Y response. If we can measure and control the vital few X’s, control
of the Y should be assured.
Statistical Process Control
Enables us to control our process using statistical methods to signal when process
adjustments are needed.
The Logic of SPC Desired Process Capability
Output
●
Upper Control Limit
Controller ●
Lower Control Limit
Samples
Input Process Output
A B C D E L M N O P
Controllable factors Uncontrollable factors
- Assignable causes - Common causes
- Adjustable - Noise
- Special - Inherent causes
SPC has traditionally been used to monitor and control the output of processes.
Six Sigma Quality focuses on moving control upstream to the leverage input
characteristic for Y. If we can measure and control the vital few X’s, control of Y
should be assured.
Control Charts
A Control chart is a graphical display of measurements ( usaually aggregated in the form
of means or other statistics) of an Industrial Process through time. By carefully scrutinizing
the chart, a quality engineer can identify any potential problem with the Production process .
The idea is that when a process is in control , the variable being measured - the mean of
every four Observations, for example - should remain stable through the time. the Mean
should stay somewhere around the middle line ( the grand mean for the process ) and not
wander off by more than the fixed standard deviations of the process . The required number
standard deviations is chosen so that there will be a small probability of exceeding them
when the process is in control . Addition and subtraction of the required number of standard
deviations ( three ) give us the Upper Control Limit ( UCL ) and the Lower Control Limit
( LCL) of the control chart. When the bounds are breached , the process is deemed out of
control.
A control chart is a time plot of a statistic, such as a sample mean, range, standard
deviation , or proportion, with a centerline and upper and lower control limits. The
limits give the desired range of values for the statistic. When the statistic is outside
the bounds, or when its time plot reveals certain patterns, the process may be out of
control.
Procedure of Control Chart Selection
Characteristic
definition
of control chart
No No No No
Variable Subgroup Average Median
Defect Faults of Control
Data Type? sampling? calculation
ratio parts chart
Yes Yes Yes
No Yes
u No
X
n =constant Control Xbar-R
Control n＞ 8 Control
chart
Yes
chart chart
Yes Yes
Yes
c No
Easy to Xbar-R
Control calculate
chart Subgroup Control
chart
Yes
No
p Xbar-R
n =constant Control Control
chart chart
Yes
pn
control This procedure is on the condition that data can be
chart collected after Gage R&R.
Control - Types of Control Charts Understanding Six Sigma
T y p e s o f C o n tr o l C h a r ts
V a r ia b le s C h a r ts A ttr ib u te s C h a r ts • There are basically two types of control

f o r m o n i t o r in g f o r m o n it o r i n g
c o n t in u o u s X 's d is c r e t e X 's charts:
–Variables charts - these charts are used
A v e ra g e & R a n g e F r a c tio n D e f e c tiv e
Xbar & R p C h a rt for monitoring X variables that are
n < 10, t y p i c a l ly n ≥ 5 0
t y p ic a ll y 3 - 5 tra c k s d p u /d p o
continuous, such as, a diameter or
consumer satisfaction rating.
A v e ra g e & N u m b e r D e fe c tiv e
S t d D e v ia t i o n n p C h a rt –Attribute charts - these charts are used
Xbar & σ n ≥ 5 0 ( c o n s ta n t)
n ≥ 10 tra c k s # d e f for monitoring discrete X variables, such
M e d ia n & R a n g e N u m b e r o f D e fe c ts
as, good/bad counts, or inventory levels.
X & R c C h a rt
n<10, c>5
t y p ic a ll y 3 - 5
• Refer to the diagram for a summary list of the
I n d iv id u a l & N u m b e r o f D e f / U n it specific control chart types
M o v in g R a n g e u C h a rt
Xm R n v a r ia b le
n=1
In order to select the appropriate control chart for monitoring your process, first determine if your key
process variables (X’s) are continuous or discrete. There are specific control charts for both
continuous data and discrete data.
Variable control chart : Customer Satisfaction Index
Example : A consumer services organization wants to monitor consumer satisfaction for their
company. Each week, a survey from each of the company’s ten regional service centers is
evaluated and the scores are tabulated. The following is an example of how an Xbar/R
control chart could be used to monitor consumer satisfaction. In this example, higher is better:
The vital information for creating an Xbar/R control chart :
Total subgroups = 25 Subgroup size, n = 10
Process average, X = 4.096 and R = 0.4504

File Open : S4 > Xbar - R
Calc > Random Data > Normal Distributions

- Generate : 10
- Store in column(s) : c1-c25
- Mean : 4.0
- Standard deviation : 0.6
( Only c7 Mean : 2.8, Standard deviation : 1.6
Manip > Stack > Stack Columns

- Stack the following columns : c1-c25
- Store the stacked data in : c26
Control Limit Formulas:
UCLX = X + A2 × R While it is acceptable to compute temporary

control limits after 5 to 10 subgroups, permanent
limits require at least 25 subgroups of data points
LCLX = X − A2 × R that are In-control for both the average and range
charts.
UCLR = D 4 × R
LCLR = D 3 × R
Actual Control Limiit Calculations for the Data
UCL = 4.096 + 0.308 x 0.4504 = 4.235
UCL = 4.096 - 0.308 x 0.4504 = 3.957
UCLR = 1.777 x 0.4504 = 0.8003
UCLR = 0.223 x 0.4504 = 0.1005

Control - Control chart Understanding Six Sigma
Constant used for Control chart

n A2 A3 D3 D4 B3 B4 d2 c4
1 2.660 3.760 - - - - - -
2 1.880 2.659 0 3.267 0 3.267 1.128 0.7979
3 1.023 1.954 0 2.575 0 2.568 1.693 0.8862
4 0.729 1.628 0 2.282 0 2.266 2.059 0.9213
5 0.577 1.427 0 2.115 0 2.089 2.326 0.9400
6 0.483 1.287 0 2.004 0.03 1.970 2.534 0.9515
7 0.419 1.182 0.076 1.924 0.118 1.882 2.704 0.9594
8 0.373 1.099 0.136 1.864 0.185 1.815 2.847 0.9650
9 0.337 1.032 0.184 1.816 0.239 1.761 2.970 0.9693
10 0.308 0.975 0.223 1.777 0.284 1.716 3.078 0.9727
Example of a variable control chart : Customer Satisfaction Index

Customer Satisfaction Index
4.5 3.0 S L= 4.5 4 6

Sample Mean
X= 3.9 3 8
3.5
-3.0 S L = 3.3 2 9
2.5 1
S u b g ro u p 0 5 10 15 20 25
4
3.0 S L = 3.5 0 9
Sample Range
3
2 R = 1.9 7 5
1
-3.0 S L = 0.4 4 0 6
0
• The weekly evaluation averages 7 and 16 fell below 3.957.
• This change in “consumer satisfaction” score was driven by some assignable cause
(either system-related or region initiated).
• The appropriate action would be to investigate, identify and fix the assignable
source of the variation.
• The variation among the regional centers for week 7 is larger than expected.
An “Out-of-control” indication can come from either chart, independently.
Control ~ Attribute data Understanding Six Sigma
Example : A local dental group wanted to know why a lot of their patients fail to keep their
appointments. A problem solving team was assembled and decided to use a p Chart to
track the percentages of “no shows”. The dental clinic began logging monthly percentages
of “no shows” for each month. Of the total appointments for each month, % “no shows” plus
% “shows” equal 100%. Since a “no show” is a defective appointment, the average total
fraction defective is called p.
Year 1996
Month Jul Aug Sep Oct Nov Dec
% Failed 40 36 36 42 42 40
Year 1997
Month Jan Feb Mar Apr May Jun
% Failed 20 26 25 19 20 18
Month Jul Aug Sep Oct
%Failed 16 10 12 12
p Chart Formulas:
p = 236/600 = 0.39333, where np =
40+36+36+42+42+40 = 236
np the fraction is based on 600,
p= total possible for 6 months
n
np UCL = .39333+3(.39333*.60667)/100)? =
p= 0.539
n LCL = .39333- 3(.39333*.60667)/100)? =
UCL = p + 3
(
p × 1− p ) and LCL = p − 3
(
p × 1− p ) 0.246
n n
Definition of Stability
• A process output is considered stable when it consists of only

common-cause variation.
• Stability also means all subgroup averages and ranges are between
their respective control limits and display no evidence of assignable-
source (special -cause) variation.
• If nonrandom patterns of data appear on the control chart, or when a
point is beyond the control limits, then this is a strong signal that
assignable-source (special-cause) variation is present in your process.
A stable process will rarely produce an output that lies outside of the +/- three
sigma stable process variation region.
Determination control limit of control chart

• The Empirical Rules emphasized that when a subgroup average falls outside
of the 3σ limits, it is a pretty rare event. Process stability is defined in terms of
these three sigma limits.
• Another way of visualizing how control charts work: Think about a sequential or
time-ordered hypothesis test for each new subgroup.
α/2
H o : µι = µ
Ha: µι ≠ µ
The hypothesis test provides the criteria for determining if

a difference exists between the subgroup mean and the
process average
α/2
The control limits are variation limits, not acceptance limits! Specification limits do Not appear on
SPC charts!
Example : The following samples were taken:

S1 S2 S3 S4 S5
22 26 20 18 19
28 22 21 16 22
15 21 20 20 21
17 22 24 19 24
16 27 19 20 22
Calculate the LCL & UCL for the X-R chart using Minitab. Comment on the
results. Find out whether process is in control or not.
Solution :
Step 1 ) Copy the data in Minitab
Worksheet.
2 ) Stack the data into one column
and subscript into other.
Go to data>stack> columns
3) Select columns from S1-S5. Stack columns
values into c7 and their subscript in c6.
4) As the sample size is less

than 10 so Xbar-R would be
used
Go to stat > control charts
>variable charts for sub
groups > X bar - R
5) Select the Option all observations in one

column and select c7 in which all the data is
stacked.
Go to Xbar-Options
6 ) Go to Estimate and choose Rbar
7 ) Go to S Limits and enter 1 2 3 in

the dialog box
8 ) Go to Tests and select all the eight

standard tests for special causes.
Test for Special Causes

• 1 point more than 3 standard deviations from center line
• 9 points in a row on same side of center line
• 6 points in a row, all increasing or all decreasing
•14 points in a row, alternating up and down
•2 out of 3 points > 2 standard deviations from center line (same side)
• 4 out of 5 points > 1 standard deviation from center line (same side)
• 15 points in a row within 1 standard deviation of center line (either
side)
• 8 points in a row > 1 standard deviation from center line (either side)
The Tests for special causes detects a special pattern in the data plotted on
the Chart. The occurrence of the pattern suggests a special cause for the
variation.
9 ) Click OK.
Xbar-R Chart of C7
40
+12SL=36.07
30
Sample M ean
+3SL=24.65
__
20 X=20.84
-3SL=17.03
10
-12SL=5.61
1 2 3 4 5
Sample
40
+12SL=36.02
30 10 ) All the Points are lying

Sam ple R ange
20 within +/- 3σ Limits, so the

10
+3SL=13.96
_
process is within the control
R=6.6
0 -12SL=0
-3SL=0
limits.
1 2 3 4 5
Sample
Example :Suppose you work in a plant that manufactures picture tubes for
televisions. For each lot, you pull some of the tubes and do a visual
inspection. If a tube has scratches on the inside, you reject it. If a lot has too
many rejects, you do a 100% inspection on that lot. A P chart can define
Rejects Sampled
when you need to inspect the whole lot.
20 98
Solution: 18 104
14 97
16 99
1) Open the worksheet and enter data. 13 97
29 102
2) Choose Stat > Control Charts >Attributes Charts > 21 104
P. 14 101
6 55
3) In Variables, enter Rejects. 6
7
48
50
7 53
4) In Subgroup sizes, enter Sampled. Click OK. 9 56
5 49
8 56
9 53
9 52
10 51
9 52
10 47
Minitab Output : P Chart of Rejects

0.35
UCL=0.3324
Sample 6 is outside the upper 0.30 1
control limit. Consider inspecting 0.25
Proportion
0.20
the lot. 0.15
_
P=0.1685
0.10
0.05
0.00 LCL=0.0047
2 4 6 8 10 12 14 16 18 20
Sample
Tests performed with unequal sample sizes

Understanding Six Sigma

Uploaded by

Copyright:

Available Formats

Understanding Six Sigma

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Understanding Six Sigma

Uploaded by

Copyright:

Available Formats

Understanding Six Sigma

Understanding Six Sigma Manual

Quality Management & Six Sigma

• Small group of sections

Most of service called products were

• Actual Practice Strategy by Dr. Michael Harry

• The Malcom Baldridge Award of 1988 of Motorola

• After Motorola, Texas Instruments, ABB, Allied Signal , GE, LG

Population Sample Data

The Group of Sample

Data Point : The single entity in the sample.

Population The Characteristic of populaton : parameter

C) Discreet Data : - The data which can be measured only in whole

Characteristics of Normal Curve

Measures of Central Tendency of Data

E.g 3, 5 , 1 , 8 , 2 , 7 , 1 , 4 . No. of Terms = 8, Arrange data in Increasing Order = 1 , 1 , 2 , 3 , 4 , 5 , 7, 8

Measures of Central Tendency of Data

C) Mode : The value Occurring maximum number of times

E.g Calculate the Median of the Following data :

Measures of Spread/Dispersion of Data

a.) Range : The Difference between Maximum & Minimum value.

E.g : The Process Specification = 10 ± 2

Measures of Spread/Dispersion of Data

What is Six Sigma ….???????

We measure defect rates in all the Processes through an expanding

We gain a competitive edges in Quality, Cost, Customer Satisfaction.

We should work smarter, not harder

The Percentage 68.3 %

The Area Under the Curve represents the Acceptance or Yield,

The 3 Sigma Company The 6 Sigma Company

• Spends 15-25% of sales dollars on • Spends 5% of sales dollars on cost

Guarantee for design completion

Quality Assurance in Manufacturing Stages

Maximizing Sales & Service

• 3/5 Why Analysis

The Process Mapping Method

Process Mapping [ex] Vehicle Purchasing

Meeting Look around Look around

Look around Trial No Shop around

Decide the Whether you No Visit another measurement

Process Mapping [ex] Refrigerator - R1 Line Rolled Throughput Yield

Door Ass’y 89.7%

D/Liner extrusion/mold Door forming Door assembly

I/Case extrusion/mold Case forming Cycle assembly

Front - CTQ, L painting O/Case, B/Plate LQC & appearance

Logic Tree (Structure Tree)

Rolled Throughput Yeild

through all the stages without any rejection / rework.

RTY = YFT1 * YFT2 * ……… YFTn ,

YNA = Normalized Yield = ( RTY of the line )1/no. of stages in Line

to compare the performance of two lines on the basis of RTY.

Yield First Time

1000 Are 820

100 Can 30 Rework &

First Time Yield Rolled Throughput Yield Normalised Yield

YFT = S/U YRT = YFT1 x YFT2 x YFT3 YNA = (YRT)**1/Opp

•YRT gives the •YNA us average

QFD : Quality Function Deployment ( What Customer Wants )

It is defined in two steps :