Six-Sigma Class 1 - 12

SIX-SIGMA
For Operations
What is Quality?
• Deming: Quality is defined from the customer’s point of view as anything
that enhances their satisfaction.
• Juran: Fitness for use. Those product features which meet the needs of
customers & thereby provide product satisfaction. Freedom from
deficiencies.
• ASQC: The totality of features & characteristics of a product or service that
bear on its ability to satisfy stated or implied needs.
• COPC: Quality is defined as knowledge of agents that would enable them to
provide accurate & consistent solutions to the customer at the very first
attempt.
• ISO: Degree to which a set of inherent characteristics of a product or
service, fulfil requirements.
Quality Defined
• Simply stated, Quality comes from meeting customer
expectations and occurs as a result of four activities:
- Understanding customer requirements

- Designing products & services that satisfy those
customer requirements
- Developing processes that are capable of producing
those products & services
- Controlling and managing those processes so that they
consistently deliver to their capabilities
Quality Defined
Quality is all about

doing it right the
first time!!!!
Quality Evolution
Quality Traditional View Quality Six-Sigma View
“ Errors are inevitable” “ Errors can be eliminated”
Focus on cost Focus on customer requirements

Quality not important Increased awareness of Quality
Culture of Radical change Culture of Continuous Improvement
Little analysis performed Data driven analysis
Cut middle management Includes suppliers
Sporadic labor productivity Continuous productivity
BRAND EROSION! BRAND ENHANCEMENT!!
Cost improvements to Quality improvements to

Result in Net Income result in Market Share
A paradigm shift!!!!
Quality Evolution
T.Q.M. B.P.R. Six-Sigma
Business Drivers Globalisation & Recession & changing Internet
Product Quality markets
Goals Defect Reduction Streamlining Process Alignment

Tools SPC Chart Process Map DOE/Simulation
Method Add value to existing Challenge Process Prioritize by COPQ &
processes Fundamentals Capability
Deployment Bottom-up Top-Down Top, Middle & Bottom
Key Feature Quality Circles Outside Consultants Internal Experts

Impact Long Term Short & Medium term Short, Medium & Long
term
Role of Technology Support Enabler Enabler
Risk / Return Low/Low High/Low Medium / High
Continuous Radical Redesign Align & Maintain

Improvement
Quality Mindset
• Quality focuses on achieving “Operational Excellence”
- Our survival is dependent upon continued growth of the business

- Our business growth is largely determined by customer satisfaction
- Customer satisfaction is governed by quality, price & delivery
- Quality, price & delivery are controlled by process capability
- Our process capability is greatly limited by the process variation
- Process variation leads to increased defects, costs & cycle times
- To eliminate variation, we must apply the right knowledge at work
- In order to apply the right knowledge we must first acquire it
- To acquire new knowledge means willingness to get trained!
Growth, Customer satisfaction, quality, knowledge and

training….all tied together!!!!
Quality Mindset
Excellence is not an act but a habit!

- Aristotle
Six-Sigma History
• Motorola (1987)
• Texas Instruments (1988)
• Digital Equipment (1989)
• IBM Computers (1990)
• Asea Brown Boveri (1993)
• AlliedSignal & Kodak (1994)
• General Electric (1996)
Six-Sigma Results
Profile of an Average Company:
• Profitable & growing
• Market prices declining
• Has a quality assurance program
• Spending over 25-35% of sales dollars on correcting, or reworking
product before it ships
• Unaware that best in class companies have similar processes that
are greater than 100X more defect free
• Believes that a zero-defects goal is neither realistic nor achievable
• Has 10X the number of suppliers required to run the business
• 5-10% of the firms customers are dissatisfied with product, sales or
service and will not recommend that others purchase products or
services
• Competitors are increasing!!
Businesses operate roughly around 3 σ

Six-Sigma Results
• Motorola: Over $16 bn savings and won MBNQA’89
• General Electric: Over $10-15 bn savings!
• AlliedSignal: Over $2.2 bn savings!

Six-Sigma Defined
What is Six-Sigma….
Six-Sigma Defined
“ Past definitions of quality focused on

conformance to standards, as companies
strived to create products and services
that fell within certain specification
limits…”
- Mikel Harry & Richard Schroeder
Six-Sigma Defined
“…this Six-Sigma journey will change the

paradigm from fixing products so that they
are perfect to fixing processes so that they
produce nothing but perfection, or close to
it”
- Jack Welch
Six-Sigma Defined
Vision
Metric
Philosophy
Six Methodology
Tool
Sigma Culture
Value
Symbol
Definition
Goal
Six-Sigma Defined
• Six-Sigma as a symbol – “Brand Enhancement”
6σ
Measure of Capability – the values Measure of Variation - The

to σ represent the symbol σ represents the
performance and consistency standard deviation of any data
level of any process population
The symbol σ is also the 18th letter in the Greek Alphabet!!!

Six-Sigma Defined
• Six-Sigma as a Philosophy – “Measurements”
Measure
Everything
That
Results
In
Customer
Satisfaction
Six-Sigma Defined
Sigma PPM Factor
80
70 1 690000 -
60
Improvement Factor
2 308500 2.24
50
40
3 66800 4.62
30
20 4 6210 10.76
10
5 233 26.65
0
1 2 3 4 5 6
Sigm a Value 6 3.4 68.53
6σ is not 3 times but 90735 times better than 2σ !!!

Six-Sigma Defined
Is 99.00% in
our daily lives
good enough?
Six-Sigma Defined
• Six-Sigma @ Daily Life!
Common Activity Defects @ 99.0% Defects @ 99.9997%
(3.80 Sigma) (6.00 Sigma)
Mail Delivery 20,000 lost articles of mail 7 lost articles of mail per hour
per hour
Drinking Water Unsafe drinking water for 15 Unsafe drinking water for 2
mins per day mins per year
Hospital Surgery 5000 incorrect surgery 2 incorrect surgery

procedures per week procedures per week
Air Travel 2 abnormal landings at most 1 abnormal landing every 5

airports each day years
Sometimes 99% is just not good enough!!!

Six-Sigma Defined
• Six-Sigma @ Daily Life!
Defects @ 99.9997%
(6.00 Sigma)
500 incorrect surgery procedures per week
22,000 cheques deducted from wrong bank accounts each hour
50 newborn babies dropped at birth by doctors each day
32,000 missed heartbeats per person, per year
One hour of unsafe drinking water every month
Even 99.9% is just not good enough!!!

Six-Sigma Defined
“ Consistently delighting customers with

increased Quality & decreased costs
through people and technology”
Six-Sigma Defined
• Six Sigma as a Culture – “Metrics”
- We don’t know what we don’t know

- If we cant express what we know in the form of numbers,
we really don’t know much about it
- If we don’t know much about it, we can’t control it
- If we can’t control it, we are at the mercy of chance!
Measure everything that results in customer

satisfaction as measurements get attention!!!
Six-Sigma Defined
• Six Sigma as a Culture – “Metrics”
Focus
On
Customer
Understanding &
Satisfaction
Variation may be the spice of life…but for a customer,

consistency is the king!!!
Six-Sigma Defined
• Six Sigma as a Methodology
Always look
for the low
hanging
fruits!!
Six-Sigma Defined
A journey, not a destination!

Attitude & Discipline

The winning combination for a Quality
centric Business!
• Behaviour of any individual is the way in which he
responds to situations in his day to day life and is a
function of values imbibed over a period of time
B = f (V)
• Values being a complex of beliefs, ideals or standards,

which characterizes any individual in the society.
Our values drive the behaviour that we demonstrate!!!

• Behaviours & Values
Behaviors
Attitudes
Beliefs
Values
Our environment drives everything above

• Attitude is a complex mental orientation involving
values, beliefs, feelings and dispositions to act in certain
ways at any given point in time
• Discipline is to get accustomed to a regular and

systematic action in order to act together under orders
and form a habit of continued obedience
• Attitude in Quality is all about being focused on

the customer. We need to view Quality
externally from the customer’s perspective.
Customers have all the votes concerning the
extent of satisfaction and value!
• Discipline in Quality deals with the process. It’s about

attending to every detail methodically, with no room for
error. By meeting challenges everyday, head-on, we
strive for continuous improvements in solving whatever
issues may arise and close-looping continuous
improvements in a systematic, scientific and fact based
manner.
• Attitude & Discipline in Quality is all about being customer centric
and meeting customer expectations every time with a process focus
and through a continuous improvement cycle which is:
Systematic
Scientific
Fact-Based
Data-driven
The right attitude & discipline are called for on both the
management and associate levels!!!
“ A positive attitude is contagious and like a

fire, unless you continue to add fuel, it
goes out and stops spreading!”
- Author unknown
Business Statistics
Statistical Thinking
Business Statistics
Measurement tools and statistical analysis
necessary to drive a metrics culture!
Statistics
• Data, Information and Statistics
- Data is a set of numbers which do not convey any logical meaning
- Information is a set of numbers which when represented correctly

conveys meaning and helps us take logical decisions
- Statistics is the science that compiles, analyses and interprets

numerical data. Statistics has two branches:
 Descriptive Statistics : This branch of statistics organises, summarises

and describes the data we want to study (mean, median, mode, range,
variance, standard deviation, etc)
 Inferential Statistics : This branch of statistics generalises to a whole
population the results obtained from a given sample (sampling,
confidence and precision levels, hypothesis testing, etc)
Statistics
• Data, Variables and Observations
 Data as defined by the Dictionary is facts, figures or
values from which a logical conclusion may be drawn as
per requirement.
 Variable is any quantity that varies; i.e. an aspect or
characteristic of a person, object or situation that can
assume different values
 Observation is any data measurement value for a
defined variable; for eg. the scores for boys and girls in
a history exam:
Boys 56 67 46 58 57 65 52 67 78 53 45 69 51 52 77 66
Girls 47 43 78 64 49 46 55 70 61 56 63 62 59 51 55 58
Statistics
• Observations and Variation
When observations, i.e. measurements for a variable, are

repeated for a variable we usually end up getting different
answers…which is nothing but variation in the measured data.
This variation can be classified into two possible categories:
 Common cause variation: Differences in the measurements

which are expected and predictable; also known as white noise
 Special cause variation: Differences in the measurements

which are NOT expected and NOT predictable; also known as
black noise
Measurement variation is natural, expected and is the foundation of statistics!!

Statistics
• Data Classification
Data Types
Numeric Attribute
Discrete Continuous Binary Non-Binary

Statistics
• Data Classification – Numeric Data
- Discrete: The data is counted. These counts can be in the form of event
occurences (Poisson, e.g. the number of errors in the form, number of
calls in the queue, etc.) or can be count-based proportions (binomial, e.g.
percentage pass/fail, etc.; pass/fail being binary data). Therefore discrete
data can have count-based values and/or count-based proportions as a
possibility and involves counting.
- Continuous: The data is measured. For e.g. the measurement of length &
breadth of a table in the classroom. This type of measurement can result
in an output like 5 feet, or 4 feet 5 inches, etc. Therefore continuous data
can have any decimal (decimal sub-division are meaningful) / non-decimal
measurement values as a possibility and involves measurement
Statistics
• Data Classification – Attribute Data
- Binary: A set of data arising out of a logical decisioning which is

either of two possible outcomes
For eg. “…is the door of the classroom open?” (Y/N). Therefore
binary data output have only two finite options
- Non-Binary: A set of data arising out of a logical decisioning which

is either of multiple possible outcomes (these outcomes can either
be ordered or random. Ordered non-binary is called as ordinal data
while random non-binary is called as nominal data)
For e.g. “…what is the color of the wall in the classroom?
(white, brown, green, ….etc, is nominal; while light, medium, dark is
ordinal)”. Therefore non-binary data output can have infinite options.
Statistics
• Data Classification
Summary: Data Types
Numeric Attribute
Discrete Continuous Binary Non-Binary
Can take a count Can take any Can take only Can take more than
based value or a numerical two values two values
derived proportion value (1, 3.2,
value (3, 5, 78, 34, 5, 7.35, 9, 15) Eg: Colors –
10 out of 100….) Nominal: red, green,
Eg: Height, Eg: blue,…
Eg: Number of
length, Yes / No Color intensity –
people,
temperature, Boy / Girl Ordinal: strong, light
percentage
etc. Day / Night moderate
present, etc.
Statistics
• Data Representation
Numeric Attribute
Histogram Bar Diagram

Time Plot Pie Chart
Statistics
• Data Representation – Histogram
A Histogram is a graphical representation of numerical data. It is
constructed by placing the class intervals on the horizontal axis of a
graph and the frequencies on the vertical axis
Age Interval Frequency 14

12
10
Employees
8
21-25 13 6
4
2
26-30 10 0
2 1 -2 5 2 6 -3 0 3 1 -3 5 3 6 -4 0
31-35 5
36-40 7
Statistics
• Data Representation – Time Plot
A time plot is a graphical representation of numerical data. A time
plot is a two dimensional graph used to detect trends in time.
Month Tickets Tic k e ts Is s u ed
50
Jan 32 40 35 36 40
30 32
Tickets
33
Feb 35 20
28
10
Mar 28 0
Ja n F eb M ar A pr M ay Jun
Apr 36
May 33
Jun 40
Statistics
• Data Representation – Bar Diagram
A Bar diagram is a graphical representation of attribute data. It is
constructed by placing the attribute values on the horizontal axis of a
graph and the counts on the vertical axis.
Social Status No. Of People 30 26

25
20
15
People
15
9
High 9 10
5
0
Medium 26 H ig h Me d iu m Low
Low 15
Statistics
• Data Representation – Pie Chart
A pie chart is a graphical representation of attribute data. The “pieces”
represent proportions of count categories in the overall situation. Pie charts
show the relationship among quantities by dividing the whole pie (100%)
into wedges or smaller percentages
No. Of People
Social Status No. Of People 18%
30%
H igh
High 9 Medium
Low
Medium 26
52%
Low 15
Statistics
• Population Attributes
LOCATION
S P R E A D
SHAPE
CONSI TENCY
S
Statistics
• Population Attributes
Four important process output characteristics that need

to be determined to fully profile any process
performance, are:
- Location: where the data tends to concentrate or
cluster?
- Spread: How much variation exists around the clustering
- Shape: What is the frequency pattern in the clustered
data?
- Consistency: Is the current snapshot useful for future
decisions?
Statistics
• Descriptive Statistics – Measures of Location & Spread
Measures of Central
Measures of Dispersion
Tendency
Standard Deviation
Mean
Variance
Median
Range
Mode Span
Stability Factor
Quartiles
Statistics
• Measures of Location – Mean
It is the arithmetic average computed by summing all the values in the dataset and
dividing the sum by the number of data values
For a finite dataset with measurement values X1, X2, ….., XN (a set of N numbers), it
is defined by the formula:
The sample mean is represented by

The population mean is represented by Greek Letter μ
Statistics
• Measures of Location – Median
- If we put all data in rank order (low to high) then the median is the
ordered value located at n/2, i.e. in the middle.
- If there are odd number of observations, then median is the
(n+1)/2th ordered value
- If there are even number of observations then median is the
average of the two middle values i.e. the n/2 and the (n/2)+1
ordered value
Statistics
• Measures of Location – Mode
- It is defined as the most frequently occuring value in the dataset

- In a histogram, it is the corresponding value to the highest bar
- The mode may not exist; and if it does exist, it may not be unique
Statistics
• Measures of Location – Quartiles
- Quartile 1: (Q1 or P25) is defined as the ordered value below which

25% of the data points fall. If we put all the data in rank order (low to
high), then Q1 is the value located at n/4
- Quartile 3:(Q3 or P75) is defined as the ordered value below which

75% of the data points fall. If we put all the data in rank order (low to
high), then Q3 is the value located at 3n/4
Statistics
• Inferential Statistics
- Population: It is the total group of elements we want to study. Population
would mean each and every employee of the organisation.
 μ – The mean, or average calculated for a population
 Σ – The standard deviation calculated for a population
- Sample: It is the subgroup of the population we actually want to study, as

to study each and every of the elements of the population is usually not
possible. Sample would mean a group of 20 employees chosen randomly
from the organisation population.
 – The mean, or average calculated for a sample
 S – The standard deviation calculated for a sample
Statistics
• Measures of Spread – Standard Deviation
- Standard Deviation can be interpreted as the average distance of the
individual observations from the group mean; it is a measure of
distance with a positive value
- Standard deviation for the population is represented as σ
Statistics
- Standard deviation for a Sample is represented as s
Statistics
Example:
• Let's say we wanted to calculate the standard deviation for the
amount of gold coins that the pirates on a pirate ship have.
• There are 100 pirates on the ship. In statistical terms, this means we
have a population of 100. If we know the amount of gold coins that
each of the 100 pirates has, we use the standard deviation
equation for an entire population
• What if we don't know the amount of gold coins that each of the
pirates has? For example, we only had enough time to ask 5 pirates
how many gold coins they have. In statistical terms this means we
have a sample size of 5 and in this case we use the standard
deviation equation for a sample of a population:
Statistics
• The rest of this example will be done in the case where we have a
sample size of 5 pirates, therefore we will be using the standard
deviation equation for a sample of a population.
• Here are the amounts of gold coins the 5 pirates have:

4, 2, 5, 8, 6.
• Now, let's calculate the standard deviation:

1. Calculate the mean:
Statistics
2. Calculate for each value in the sample:
3. Calculate :
Statistics
4. Calculate the standard deviation:
The standard deviation for the amounts of gold coins the pirates have
is 2.24 gold coins.
Statistics
• Measures of Spread – Variance
- Variance is defined as the spread of the standard deviation from the
group observations; it is a measure of area with a positive value
- Variance for the population is represented as
- Variance for a sample is represented as

Statistics
• Measures of Spread – Range
Range is the difference between the smallest value and the largest
value in a data set; it is influenced by extreme values
Range for a given dataset is calculated as:

Range = Highest Value – Lowest Value
Example: 34 50 61 77 99 26 15 62 44 74 88
is 99 - 15 = 84
- As range is calculated from extreme values it does not tell how

much is the scatter of other values around the center; in such
scenarios the interquartile range (IQR) is helpful
- Interquartile range for a given dataset is calculated as:
IQR = Value Q3 - Value Q1
Statistics
• Example of Range
To find the range in 3,5,7,3,11
Step 1: Arrange the numbers in ascending order.
3,3,5,7,11
Step 2:
In the above distribution
The largest number is 11
The smallest value is 3
Formula = largest number - smallest number
Range = 11-3 = 8
Statistics
• Example of Interquartile Range: (temperatures)
Statistics
• Measures of Spread – Span
- Span measures how successful we are in meeting our customer
needs and requirements; and is NOT influenced by extreme values
On Time
But how early?

But how late?
- Span for a given dataset is calculated as under:

Span = P95 value – P5 value
Statistics
• Measures of Spread – Stability Factor
- Stability Factor captures the spread or the variation of the output

- Stability factor is calculated as:
SF = Q1
Q3
Statistics
Advanced Statistics
Statistics
• Distributions – Measures of Shape
Probability Frequency
Statistics
• Distributions – Measures of Shape
The two possible distribution models for our statistical study are:
- Probability distribution – A theoritical data distribution which
represents futuristic data; values are between -1 & 1
- Frequency distribution – An actual data distribution which
represents historical data; values are actual measurements
Probability Distribution Frequency Distribution
Futuristic data Historical data

Values are between -1 & 1 Values are actual measurements
Theoritical distribution Actual distribution
Statistics
• Probability Distributions – Measures of Shape
Discrete Continuous
Poisson Exponential
Binomial Gaussian
Statistics
• Probability Distributions – Measures of Shape
- A probability distribution of a random variable is an assignment of

probabilities to each of the possible values the random variable
might assume at any given point of time
- Each probability is a value between zero and one and the sum of all
probabilities must equal 1
- The bins of a histogram keep getting smaller and smaller as the
number of data points gets larger and larger
- With enough data a discrete distribution also approaches as
continuous distribution
Statistics
• Probability Distributions – Gaussian
- A gaussian distribution also known as the normal distribution
gives us a bell shaped curve for the frequency distribution which is
symmetrical about the mean
- This continuous distribution is used to determine the probability
that a transaction would take between x & y range
Examples: Used to represent distribution of the time it takes to

process a certain type of application measured in seconds; etc.
Normal distribution results in a symmetrical

bell shaped curve!!!
Statistics
• Probability Distributions – Gaussian
- This statistically most important probability distribution exhibits the
following vital characteristics:
- The probability curve indicates random or chance variation
- The mean, median, mode of the probability curve are the same
- The probability curve peak represents the center of the process
- The probability curve theoretically does not reach towards zero
- The probability curve can be divided in half with equal pieces falling
either side of the most frequently occurring value
- The total area under the curve represents virtually 100% of the
product or service that the process is capable of producing
Statistics
• Measures of frequency – Normality
• Normality is a measure of distribution of frequently occurring values

around the average, and the other values tailing off symmetrically in
both directions, from -∞ to +∞
- Anderson-Darling test is used to compare the actual frequency
distribution with a theoretical normal distribution calculated by using
sample estimates for μ and σ of the frequency distribution
The test calculates a Anderson-Darling statistic (AD) and a
Critical-value statistic (CV) and if AD < CV then we assume the
frequency distribution to be exhibiting characteristics of a normal
distribution
Statistics
• Measures of Consistency – Stability
- Stability is a measure of differences in the frequency distribution
which are expected and predictable for and over a period of time
- For the frequency distribution to be stable the probability of
differences being un-expected and un-predictable is less than 0.05
Stability presumes only common cause variation!!!

Statistics
• Inferential Statistics
- Estimation: Estimation is the process by which sample data are

used to indicate the value of an unknown quantity in a population
- Results of estimation can be expressed as:

 A single value known as a point estimate, or
 A range of values known as a confidence interval
- How well the sample represents the population is gauged by two

important statistics – the sampling confidence level (confidence
coefficient) and precision level (margin of error). They tell us how
well the samples represent the entire population!!
- The confidence coefficient statistic is a variable dependent of

confidence intervals and confidence limits
Statistics
• Inferential Statistics – Confidence Interval
- Confidence Interval: A confidence interval gives an estimated range

of values which is likely to include an unknown population
parameter, the estimated range being calculated from a given set of
sample data; these intervals are useful in assessing the practical
significance of an given result
- The width of a confidence interval is related to sample size and

measurement variability in the observations
The width is decreased by increasing the sample size, but is
increased with the increasing variability in our processes
Statistics
• Inferential Statistics – Confidence Limit
- Confidence Limits: The lower and upper limits of a confidence

interval that define the interval within which a population parameter
being estimated presumably lies. These limits are computed from
sample data and have a known probability that the unknown
parameter of interest is contained between them
- Confidence limits, which define the range of a confidence interval,
are usually annotated as:
 LCL – Lower confidence limits
 UCL - Upper confidence limits
Statistics
• Inferential Statistics – Confidence Level
- Confidence Coefficient: The confidence coefficient of a confidence

interval for a parameter is the probability that the interval will contain the
value of the parameter of interest. It is the percentage of intervals that can
be expected to include the actual value of the parameter being estimated
- The confidence level (CL) is the probability value associated with a

confidence interval and is often expressed as percentage
Statistical experiments are often carried out at a confidence

level of 95%!!!
Statistics
- Confidence coefficient for running statistical experiments can take

different values depending on the confidence level required for the
estimation of results
- Coefficient has a value of 1.64 for tests conducted at 90%-CL

The most commonly used confidence coefficient values for

many statistical experiments!!
Statistics
Example: Confidence level in sampling for survey respondents of

ABC firm from the entire consumer base in a metropolitan city
A 95% level of confidence means that 5% of the surveys will be off

the wall with numbers that don’t make sense. Therefore for
example, if 100 surveys are conducted using the same customer
service question, five of them will provide results that are somewhat
wacky. Normally researchers do not bother about this 5% because
they are not repeating the same question over and over so the odds
are that they will obtain results from among the 95%.
Statistics
• Inferential Statistics – Precision Level
- Margin of Error: The precision level refers to the spread of an

estimate of a parameter, and/or the quality associated with a set of
measurements by which repeated observations approximate to the
true value of a parameter
- A precise measurement may not be accurate because of the
unrecognised bias or the other errors in the process of sampling
Statistical experiments are typically carried out at a

“precision level” of 5%!!!
Statistics
• Inferential Statistics – Precision Level
Example: Precision level in sampling for survey respondents of

ABC firm from the entire consumer base in a metropolitan city
A survey may have a margin of error of plus or minus 5% at a 95%

level of confidence. These terms (treat it as precision) simply mean
that if the survey were conducted 100 times, the data would be
within a certain number of percentage points above or below the
percentage reported in 95 of the 100 surveys.
Statistics
• Inferential Statistics – Confidence & Precision Levels
- Summary: Confidence & Precision Levels in sampling for survey

respondents of ABC firm from the entire consumer base in a
metropolitan city
A customer research firm surveys customers and finds that 50%

of the respondents say that its customer service is “very good”.
The confidence level is cited as 95% plus or minus 5%. This
information means that if the survey were conducted 100 times,
the percentage who say service is “very good” will range
between 45% & 55% most of the time at a 95% confidence level.
Normal Curve
A theoretical probability distribution

which represents variation in any data
futuristically on a bell shaped curve!!
Normal Curve
• Normal Curve
– A probability distribution where the most frequently occuring
value is in the middle and the other probabilities tail off
symmetrically in both directions from
-∞ to +∞
-6σ -5σ -4σ -3σ -2σ -1σ 0 +1σ +2σ +3σ +4σ +5σ +6σ
68.26 Percent
95.46 Percent
99.73 Percent
99.9937Percent
99.999943 Percent
99.9999998 Percent
Normal Curve
68.26% of the area falls within +/- 1 sigma

95.5 % of the area falls within +/- 2 sigma
99.73% of the area falls within +/- 3 sigma
Normal Curve
Normal Curve
Normal Curve
2 Sigma Process
308,500 Defects per million

High inspection required
More correction and re-work
Clumsy and delayed process
High cost per transaction
Normal Curve
6 Sigma Process 3.4 Defects per million
No inspection required
No correction and re-work
Agile and On-time process
Optimum transaction cost
Normal Curve
Six-Sigma indicates
how much of the data falls
under a normal curve within
the customers’ requirements!!!
Process Focus
The business Process

A Process is a collection of activities that
takes one or more kind of inputs and
creates output that is of value to the
customer!
Process Focus
• Levels of maturity of a process
- Any business process at a given point in time, in its life cycle, can
be operating at any one of the following maturity levels:
 Level 1 - Anecdotal (ancient)
 Level 2 - Informal
 Level 3 - Formal
 Level 4 - Managed
 Level 5 - Optimised
Understanding the level of maturity for our business

process(es) is vital to our project success!!
Process Focus
• Defining Process and Process Capability
- A process focused business constantly realigns processes to optimised levels

in order to remain capable of meeting changing market demands. Only by
gaining predictability, can an enterprise truly maintain capable processes to
changing customer demands
- Three key terms that help us define process capability are:

 Process Baseline
 Process Entitlement
 Process Benchmark
Six Sigma focuses on understanding variation in our business

processes!!!
Process Focus
• Process Baseline
- Process Baseline is the average, long term performance level of a

process when all the input variables in the process are running in
an unconstrained fashion
- It doesn’t mean that we let our processes run out of control.
Instead, it means that we let all input variables vary across their
entire range of expected values
- The result of allowing all possible variables to vary across their full
range helps us determine our average, long term performance
Process Focus
• Process Entitlement
• Process entitlement is the best case, short term

performance level of a process when all the input
variables in the process are centered and in control
• It is effectively the opposite of the process baseline
• The result of process entitlement helps us determine
what we are really capable of. This allows us to set
realistic goals, not wishes
Process Focus
• Process Benchmark
- Process benchmark is the performance level of the

process deemed by comparison to the best process
possible
- It takes us to the best that anyone has ever done. In
practical terms this means researching and finding the
best that has ever been done in the industry
- Knowing the benchmark is very valuable because it
allows us to establish competitive and realistic objectives
Process Focus
• Secret truths about Processes
• Processes are everywhere

• We often don’t know that we are part of a process until something
goes wrong!
• Processes grow and change without anybody realising it!
• Left alone, processes tend to get more complex!
DMAIC
DMAIC
Methodology
DMAIC
DMAIC
A continuous improvement methodology
under the Six-Sigma Framework!
DMAIC
• A data based Improvement methodology
- You need baseline data to understand the process

- Don’t trust historical data! If it’s all that’s
available….question it!
- Analysis and improvement plans must be based on data
- Improvements must be validated with data
- Monitoring and control plans require data
DMAIC
• The Approach
DMAIC
A
PD C
DEFINE
CONTROL
MEASURE
Act Plan
ANALYZE
Study
(Check) Do
IMPROVE
DMAIC
• DEFINE
This is the first step that refers to defining the goals of the
project. Process improvement goals may be aimed at
increasing market share, the output of a particular
department, bringing about improved employee satisfaction
as well as customer satisfaction and so on.
The goal has to align the customer demands and the strategic
goals of the organization. Data mining methods can be used
to find prospective ideas for project implementation.
In other words, businesses are designing a road map for
achieving the targets and goals of the organization.
DMAIC
• Measure
This phase refers to the analysis of the existing system with

various measurement techniques for the defects and levels of
perfection that exist. In this step, accurate metrics have to be
used to define a baseline for further improvements.
This helps Six Sigma team leaders understand if progress has

been achieved when process improvements are implemented.
DMAIC
• Analyze
This phase is extremely important in order to determine any

disparity that may exist in the goals set and the current
performance levels achieved. Various statistical tools are
available to undertake such an analysis.
The understanding of the relationship between cause and

effect is necessary to bring about any improvements, if
needed.
DMAIC
• Improve
Improvements in existing systems are necessary to bring the
organization towards achievement of the organization goals.
Creative development of processes and tools brings about a
new lease on life for the organization’s processes and takes
them nearer to organizational objectives.
Various project management and planning tools can be used
to implement these new techniques and processes.
Appropriate usage of statistical tools is important to measure
the data, which is necessary to understand improvements
done and any shortcomings that may exist.
DMAIC
• Control
The last phase of DMAIC is the Control phase. It helps ensure
that variations in the processes are rectified before they have
a negative effect. Controls can be used to ensure sustained
improvements in new processes and operating procedures.
The new project components should become a part and
parcel of existing processes. Once all the factors are
performing to satisfaction, transfer of ownership should be
done to process owners and process teams.
The DMAIC model is extremely beneficial in bringing about
the change which the Six Sigma project aims at and has the
ability to show good business results
DMAIC
• Define – Identify the business problem that needs to
be fixed and the scope that will be addressed by the
project team in a stipulated time frame
- Step 1 – Customer and Project CTQ Identification

- Step 2 – Project chartering and story boarding
- Step 3 – Project acceptability and success factors
- Step 4 – As-Is process mapping and SIPOC
Project alignment!!!
DMAIC
• Measure – Quantify the problem at hand as it
currently exists and gather data that will help you
understand the issues completely
- Step 5 – Define performance Measurements

- Step 6 – Identify and prioritize possible Project X’s
- Step 7 – Develop data collection strategy
- Step 8 – Calculate Process Capability
Establish baseline!!
DMAIC
• Analyse – Use the gathered data to understand the
problem statistically and identify the area of focus
for improvement
- Step 9 – Define Performance Objectives

- Step 10 – Identify and validate Vital X’s
Determine Y = f(X)
DMAIC
• Improve – Find possible solutions, plan for new
measurements and implementation strategies for fixing the
quantified problem
- Step 11 - Generate possible solutions

- Step 12 - Select and pilot solutions
Optimise Y = f(X)
DMAIC
• Control – verify that implemented improvements
have fixed the problem and put procedure in place to
continue the verification on a periodic basis
- Step 13 – Process Control & Risk Analysis

- Step 14 – Document and execute Control Plan
- Step 15 – Institutionalize and Sign-Off
Sustain long term!!

DMAIC
• The DMAIC Philosophy
Define Project Alignment

Measure Establish Baseline
Analyse Determine Y = f(X)
Improve Optimise Y = f(X)
Control Sustain Long term
DMAIC
DMAIC Projects
DMAIC
Improvement Strategy:
Cost
esses
Cycle
DMAIC
Improvement Strategy:
Project
Qual
Result
Lower
DMAIC
• Cost, Quality or Cycle?
- Goal – I want to reduce the cycle time

Most projects start this way because our view is that this is the problem
and it is impacting customer satisfaction!
- Whats the defect – “The process takes too long”
Which leads to an erroneous conclusion that the defect is …too much of
time. Therefore the tendency is to spend our energy in reducing it by
going faster!
- Where is the re-work?
If we focus more on identifying the rework loops, bottlenecks, etc., we
get closer to the real cause of “why does this take so long?”
- What is causing it?
If we now take the approach that an incorrect procedure (for example) is
the real cause, then we focus more on the process and drive quality!
- Is your project centered around cost, quality and cycle?
Improving quality in the process will eliminate rework loops/bottlenecks
that improve cycle and drive costs down!
DMAIC
• Variation – Discrete vs. Continuous Thinking
- Discrete Thinking: e.g.; Payment posted on time? (yes / no)
- Continuous Thinking: e.g.; When was payment posted?

(number of hours); how much payment was posted (value of
transaction)
Projects with Con
tinuous
data are preferre
d as
more data to meas
ure!!
DMAIC
• Variation – the prediction Equation Thinking
Variation in product characteristics (Output Variation) =

Function of variation in process parameters (Input and
process variation)
Y = f (X)
DMAIC
• The prediction Equation and the Project Philosophy
Define The Problem

Measure 20 -25 Inputs
Analyse 10 – 12 X’s
Improve 4 – 8 Vital X’s
Control 3 – 6 Vital X’s
DMAIC
• Lifecycle of a Project – The DMAIC Philosophy
Define What is important to the customer?
(Voice of customer / Surveys & Interviews / Transaction Behaviours /
….)
Measure What is the frequency of defects?
(Measurement system / Process mapping / FMEA / Output Measure
baseline / ….)
Analyse When, Where and why do defects occur?
(Descriptive statistics / Pareto / Fishbone / Hypothesis testing /
FMEA / ….)
Improve How can we improve the process?
(Expert brainstorming / benchmarking / modeling / Design of
experiments…)
Control How can we sustain the improvement?
(Measurement feedback & control / Procedural transparency / ….)
Define Phase
Define
• Key deliverables

Identify the customers; identify customer needs & requirements;
identify customers CTQs & interpret Project CTQs; define Defects
- Step 2 – Project chartering and story boarding
Define the business case; develop the problem statement; develop the
goal statement; assess the project scope; decide project milestones; project
team selection; define team roles & responsibilities
- Step 3 – Project acceptability and success factors
Conduct threat vs. opportunity Matrix; identify critical success factors;
conduct stakeholder analysis
- Step 4 – As-Is process mapping and SIPOC
Identify key suppliers, inputs & outputs to the process; Conduct As-Is
Process mapping; Document the SIPOC
Define
Step 1
DMAIC
Define
Identify the customers;

Identify customer needs & requirements;
Identify customers CTQs & interpret Project CTQs;
Define Defects
Define
Customer Identification
Define
• A word from your customer
- I am your customer…..and if you satisfy my wants with personal

attention and a friendly touch, I will become a walking advertisement
for your products & services. Ignore my wants, show carelessness,
inattention and poor manners, and I will simply cease to exist – as
far as you are concerned
- I am sophisticated…much more than I was a few years ago. I have
become accustomed to better things. I have better money to spend.
- I am an egotist…and I am sensitive. My ego needs the
nourishment of a friendly, personal greeting from you. It is important
to me that you appreciate my business. After all, when I buy your
products & services, my money is feeding you.
Define
• A word from your customer
- I am a perfectionist …and I want the best that I can get for the
money that I spend. When I criticize your products or services – and
I will, to anyone who will listen – take heed. The source of my
discontent lies in something you or the products that you sell have
failed to do. Find the source and eliminate it or you will lose my
business and that of my friends as well.
- I am fickle …and other businesses continually beckon me with
offers for my money. To keep my business, you must offer
something better than they. I am your customer now, but you must
prove to me again and again that I have made a wise choice in
selecting you, your products and services, above all others!
Define
• Customer Identification
- Customer is one who buys or uses your products and/or

services
- Customer is one who receives the process output time-to-time
- We broadly classify customers into two categories:
 Internal Customer – Management, employees, any
functional department in your organisation, internal
shareholders, etc
 External Customer – Clients, end-customers, external
shareholders, etc.
Define
• Customer Identification
Internal External
VOICES
Needs and Requirements

Define
Needs and
Requirements
Define
• Customer Needs & Requirements
- Need is a desire or an expectation of a customer from a given

product or service; Customers have many stated needs which
are often vague and generally are the ‘wants’ from a product
or service
- Requirement is an attribute of the product or service which

fulfills the needs of a customer; customer defines these
requirements and are the “musts” of a product or service
Needs are stated while requirements are defined!!

Define
• Understanding Key Customer Needs
- In order to understand key customer requirements it is
important to obtain information on varied customer needs
from customer voices
- There are various ways of customer research to obtain
information on customer needs:
Interviews
Customer Observation
Be a Customer
Customer Complaints
Focus Groups
Market Research
Surveys
Define
Voice
Of Customer
Define
• Voice of Customer
- Voice of customer methodology is used as a framework to capture

and translate the varied customer needs – both current and latent
- VOC methodology helps capture the needs of a customer through
verbatim comments i.e. voices
- VOC helps translate verbatim comments (customer voices) into
wants (customer needs) to product or service output characteristics
that are musts (customer requirements)
- VOC does not rewrite the customer need, but merely translates it!
Voices are infinite, needs are many while requirements are few!!
Define
• Voice of Customer – The Framework
At an organisation level we broadly classify the Voice of Customer

framework into four different and distinct categories:
 Voice of Customer: The feedback from our customers on

fulfillment
 Voice of Business: The feedback from our Management; what is
fundamentally required to be done in a business. Has to go with
profitability, growth, strategies, etc.
 Voice of Process: The feedback from our CTPs & our CTQs. Has
to go with what we control as an input while the customer may not
see the same; adjectives like capable, stable, deteriorating, etc
 Voice of Employee: The feedback from our employees. Has to do
with freedom to speak, opportunity to grow, supporting environment,
etc.
Define
• Voice of Customer – The Framework
Voice of Customer: VOC

Voice of Business: VOB
Voice of Process: VOP
Voice of Employee: VOE
Define
• Voice of Customer and Customer Segmentation
- Not all of your customers have the same needs, requirements

or priorities. Segmenting your customers allows you to more
clearly understand the requirements of different groups and
focus your improvement efforts accordingly
 If you have many types of customers, breaking them
down into different subgroups allows you to develop targeted
measurement indicators and strategies for each group
 All customers are important, but some customers will use
your services more frequently, or could be more critical than
others and hence would require more time and resources
being devoted!
Define
• Voice of Customer and Customer Segmentation
- A segmentation of the customer base is recommended
to focus customer research on the most important
customers as different customer segments often have:
 Different functions
 Different causes
Segments can be based on many different variables like:

 Geography
 Customer relationship tenure
 Type of business or industry, etc
Define
• Voice of Customer and Problem containment
- A critical problem is a problem that’s revealed through VOC capture and

requires immediate attention for an urgent corrective action
 Critical problems require containment to avoid severe, costly

damage to your customer and business need. If a critical problem
is not contained, your team misses the opportunity to help the
business avoid unnecessary risks
- An identified problem requires immediate containment if it:

 Could cause harm to the customer concerned
 Impacts the customers ability to deliver the output
 Results in a significant financial loss to business
 Results in significant risk of loosing the customer
 Has controllership and compliance issues
Define
• Problem containment – The Method
The steps for the containment of a critical problem are:
 Assess financial impact of a critical problem containment
 Identify what action will result in an immediate fix and stop
bleed
 Ensure that customer and business will be benefiting from
the action
 Identify consequences of the action on the project, business
and customer
 Define how effectiveness of the containment action will be
measured
 Establish target date for conclusion of the containment
strategy
Define
• Voice of Customer – The Method
- In order to manage customer expectations throughout VOC

data collection, we need to:
 Select the target customer audience carefully
 Explain your intent for gathering the information
 Clarify your ability to act on the information gathered
 Communicate next steps to the customer in advance
Asking for information does not translate to a promise to act!!!

Define
• Voice of Customer – The Method
- Three key methods that can be developed to capture VOC

are:
 Customer focus groups
 Customer Interviews
 Customer surveys
- One key method that can be deployed to classify VOC is:

 Affinity Diagram
Define
• VOC Capturing method – Customer Focus Groups
- Customer focus groups is a technique in which we capture or
organise perceptions i.e. the collective viewpoint, of a group of
customers on a pre-defined parameter or a particular need
segment
 Participants share their ideas, concerns, likes or dislikes
 Focus groups have an environment which is conducive
 Focus groups are moderated by a facilitator
- The objective or use of a focus group during customer research is:

 To clarify and define the customer needs
 Gain insights into the prioritization of needs
 To test for and get feedback on concepts
Define
• Focus Groups – The method
- A focus group being a carefully planned discussion , designed to obtain

perceptions on a defined area of interest in a non-threatening environment,
needs to meet the following rules:
 Compose a team of 7- 14 participants who share characteristics
that relate to the focus group topic (a pre-defined parameter or a
particular segment)
 Often ask only two to three vital questions during the focus
group discussion
 Participants should be asked to thoroughly discuss on the vital
questions
 A typical focus group should be planned to last for at least two or
four hours
 Usually try to offer an incentive for better team participation
Define
Planning Tips
Pull together existing data
verbatim complaints, com
Administer the unstructure
Define
Executing tips
Move straight to understan
requirements of the custom
Demonstrate by examples
provide good clarity and a
Define
• VOC Capturing method – Customer Interviews
- Customer interviews are forums in which we capture or learn about a

customer’s perception on a specific product or service
 The interview process can be either informal or structured
 The interviewer can use both open-ended and close ended questions to
gather information
- The objective or use of an interview during customer research is:

 At the beginning: To learn what is important to customers, which
supports the development of hypotheses about customer needs and
priorities
 In the middle: To clarify points or to better understand why a particular
issue is important to customers
 At the end: To clarify findings, to get ideas and suggestions, or to test
ideas with customers
Define
• Customer Interviews – The method
- Some of the important interviewing requirements are:

 When listening to customers record exactly what they say
as you hear and what they do as you see!
 When asking questions try and get maximum facts and not
just the opinions
 Don’t come head-on, but approach the interviewees from
different angles using open ended questioning and
relevant examples
Define
• VOC Capturing Method – Customer Surveys
- Customer surveys is a technique in which we take the feedback

from a customer sample representing the entire population to
measure the importance and/or performance of customer needs.
- The Survey focuses on collecting prompt and accurate information
on:
 Customer preferences, needs & behaviors
 Performance against their needs
- Different question types that can be used in a survey are:

 Open ended: Text
 Likert Scale: Rating from 1 – 10
 Binary: Yes/No
Define
• Customer Surveys – The Method
- A customer survey as a technique should be used to analyse

goals, processes and possible cause & effect relationships
upfront
- Widely scaled rating questions should be used as they allow
lots of analytics on the results and help keep survey sample
sizes low
- We should have only few written questions so as to address
any unforeseen situations or problems
Surveys should have an action oriented focus to generate

solutions, not more questions!!!
Define
A good survey questionnaire has the following three

sections:
o The demographic section – Captures the categories into
which the respondents need to be split along with their basic
particulars
o The numerical section – captures the ranking of root causes
using a continuous rating scale, like “agreement’ or
“satisfaction”
o The written question section – captures root causes that did
not come out of the stated questions, providing some clues, in
case none of the expected rot causes could hit home. It
should be short & focused!
Define
• A rating scale is an important factor towards the success of
any survey:
• A poor rating scale – Discrete scales: 1 to 5 or Disagree to
Agree. This provides limited analysis and requires big sample
sizes for administering
• A good rating scale – Continuous scales: 1 to 9 or Strongly
Agree to Strongly Disagree. This provides a strong analysis
and require small sample sizes
Define
• Customer Surveys – The DMAIC linkage
Linking the B
Define
Simple survey
Linking the P
Measure
Define
• Comparison of VOC Data Sources
Focus G roups Interview s Surveys
G roup Interaction helps us to generate Can help tackle com plex issues w ith relative
better inform ation ease & com fort Low er cost/ tim e approach
M ore in-depth responses can be solicited
Facilitates use of visual aids tow ards data
from participants capture Higher rate of response
s
G ood w hen people are not w anting to

Pluse
respond w illingly or accurately via phone/

Require least am ount of trained &
Excellent for defining CTQ s m ail experienced resourcses to adm inister
Inputs can be easily sought on vital & Results can be produced m uch faster &
com plex issues quicker
Learnings only apply to those w ho get Requires trained & w ell experienced W e can get incom plete results from the
asked & w e cant just generalise for the interviw
rest ers to get the desired inform ation
respondents
s
Long cycle tim e to cover all respondentsM ail

& based response rate is generally 20%
Delta
W e get only anecdotal data to com plete data collection exercise 30%
data collected is qualitative v/s quantitative Interview er can influencethe interview ee in
requirem ent a tele-survey
Define
• VOC Capturing Methods – Key Considerations
- How could the collectors bias affect what is heard?

- What contact or relationship do you have with the customer?
- What are your time constraints to complete data capture?
- What budget is available to facilitate data gathering?
- How much certainty do you need to move forward with the
project?
- How will you ensure that customer expectations are aligned
with your expectations or actions?
Define
• VOC Classification Method – Affinity Diagrams
- The Affinity diagram tool used to organise & present large

amounts of data (ideas, issues, solutions, problems, etc) into
logical categories based on user perceived relationships and
conceptual frame-working
- The freely collected ideas, one written on cards or sticky
notes, are lined up & organised to show mutual, exclusive or
natural relationships, i.e. their affinity.
Define
• Affinity Diagrams – The Objective
We should use Affinity Diagrams in situations like:
 Where a breakthrough to new ideas is desired

 Where a group consensus is necessary
 When issues seem too large & complex to grasp
 Where facts or thoughts are in chaos
 When the teams creativity & intuition is to be tapped
Define
• Affinity Diagrams – The Method
The key steps towards making an affinity diagram are:

 Capture all the voices i.e. ideas, as they come
 Clarify on ideas discussed if so be the requirement
 Let ideas emerge & do not have predefined groups
 Have each team member write one idea per Post-it note and post
on a wall randomly.
 Group ideas that seem to belong together or “affinities”
 See if smaller idea sets belong to a larger group
 Build consensus on the grouping of ideas
 Create a 'header' card for each general category of ideas below it.
Define
• Affinity Diagram: Example
The personnel department of a food supermarket chain store

identified a high resignation rate of good checkout staff. There was
information available from exit interviews about their reasons for
leaving, but this was disorganized and there was no clear area that
they felt they could address. They decided to use an Affinity
Diagram to try to better understand why these people were leaving
(see as illustrated).
As a result, the checkout process was investigated further and
eventually completely redesigned. This included a redesigned booth
and hourly breaks for operators. Consequently, there were
significantly fewer leavers (and as a bonus, customer satisfaction
increased).
Diagram……
Define
Affinity
Diagram
Define
• Translating Voices to needs – The objective
The objective of translating infinite voices to the customer

needs is to meet the following:
 Differentiate the needs, or the wants, from infinite voices

 From the voices establish the service or product requirements
 To capture the stated needs, or wants of the customers
 To help get the final customer requirement parameters
Define
• Translating Voices to needs – The Method
Some of the guidelines for documenting “needs” are:
 Must be written from the customers perspective

 Write the need and NOT the potential solution
 Write the need in complete sentences; examples help
 Use measurable terms & be concise
 Write from a positive perspective
 Validate the need with the customer
Define
• Translating Needs to Requirements – The Method
Some of the guidelines for establishing “requirements” are:

 Ask questions & clarify on need verbatim
 Probe for deeper understanding & better clarity
 Translate into terms that make sense to you & to your process
 Review with customers & process participants
 Refine & re-state customer requirements
Define
• Prioritising Key Customer Requirements
 Basic Requirements
 One-dimensional Requirements
 Delighter Requirements
 Indifferent Requirements
 Reverse Requirements
Define
• Basic Requirements
 Are “must-be’s”
 Are required & expected to be there
 Are needs the customer assumes will be met
 When are fulfilled, the customer is greatly satisfied
 When fulfilled, the customer is neutral (i.e. they do not produce additional
satisfaction)
For example, if a restaurant is very clean, it will not bring additional

satisfaction to the customer because cleanliness is regarded as a must-be
requirement. If a restaurant does not meet the minimum requirement for
cleanliness, customers will not go to that restaurant at all. Customers usually
do not explicitly demand must-be requirements.
Define
• One-dimensional Requirements
 A direct positive co-relation exists b/w satisfaction levels & the degree of
presence
 The more One-dimensional elements needs are met, the better
For example, for a given model of car, the higher the mileage per gallon, the
higher the customer satisfaction. If the mileage per gallon is under a certain
level, customers will be dissatisfied. We can say that the lower the mileage
per gallon, the higher the dissatisfaction regarding this requirement. In this
example, the level of neutral satisfaction is the industry average for that
class of car. Usually, customers explicitly demand one-dimensional
requirements.
Define
• Delighter Requirements
 Are “Attracters”
 Their presence in a product or process is unexpected & fulfills the latent
needs of a customer
 Leads to great satisfaction if found present
 When delighters are absent, the customer still is neutral (& not dissatisfied)
For example, if at the end of a dinner a restaurant gives a complimentary

gift to its customers, it will bring satisfaction. If the gift is not offered, it will
not bring dissatisfaction to customers. Attractive requirements are neither
explicitly expressed nor expected by the customer.
Define
• Indifferent Requirements
 Indifferent elements are needs that result in neither satisfaction nor

dissatisfaction whether they are present / met or not
Examples of neutral/ indifferent characteristics are those product features

that are never or rarely used by the customer
Define
• Reverse Requirements
Reverse elements are needs that result in:

 Dissatisfaction when they are fulfilled
 Satisfaction even when they are not fulfilled
They may indicate that the perception of that question in the

marketplace is the opposite of the perception of the survey’s
creator.
Define
CTQ Identification
and Interpretation
Define
• Translating Requirements to Project CTQs
A Critical-to-Quality (CTQ) parameter is a specific attribute

or quality of the output that is a key requirement for customer
satisfaction
Once the key requirements have been determined &
classified into basic, performance & excitement needs, the
same need to be converted into product or service CTQs
Two commonly used tools to convert customer requirements
into project CTQs are:
 Quality Function Deployment
 CTQ Drilldown tree
Define
• Quality Function deployment
Quality Function deployment is a structured approach to:
 translate customer voices into requirements (CTQs)
 translate specific wants into service requirements
 translate service requirements to service processes
 translate service processes to process controls (CTPs)
Many Six Sigma practitioners use the Quality Function Deployment

(QFD) tool to translate the voice of the customer (VOC) into product
specifications. QFD offers a customer-oriented approach, supporting
design teams in developing new products based on an assessment of
customer needs. Customer needs are translated into design attributes,
which are then deployed in process and quality requirements.
Define
Benefits derived from using the QFD tool include:
• The creation of work teams that include multiple skills and

experiences
• The determination of specific work aims
• A display of a wide variety of important design information in
one place in a compact form
• Reduced overall costs from realizing a reduction in design
changes
• Reduced production costs by eliminating redundant features
and over-design
Define
What Is QFD?
QFD is composed of four stages:

• Complete the house of quality (HOQ).
• Design the product – Determine tolerance of each part of the
product so that it satisfies the target value identified from the
HOQ.
• Design the process – Determine the necessary production
process that will satisfy tolerances established during product
design.
• Control the process – Determine quality standards for the
new product design.
Define
• House of Quality - HOQ Room 2 of the HOQ, is
sometimes completed by the
marketing department
because of its relationship
with customers, although
other areas of the company
with customer touch points
also may contribute. The
information contained in
this section represents the
VOC and is obtained from
various sources such as:
Searching the market for
industry standards /
Customer surveys /
Analyzing customer
complaints, etc
Define
• Kano Model
Created in the 1980's by Professor Noriaki Kano, it's main

objective is to help teams uncover, classify, and
integrate 3 categories of Customer Needs and
Attributes into the Products or Services they are developing.
Missing any of these needs will jeopardize the success of the
offering.
The Benefits
 Provides a systematic, data-based method for deeper
understanding of customer needs by prioritising them
 Helps focus efforts on meeting the vital few needs
Define
• Kano Model –
Define
• Kano Results and the HOQ
An example of designing a pencil is used to illustrate the process of
completing the HOQ. Step 1 for this example is shown below,
where results from the Kano survey are entered into the customer
requirements and needs section of the HOQ
Customer Needs and Requirements
Define
Step 2 involves asking customers to rate the importance of each
requirement using a numerical ranking (from 1 to 10). The ranking is then
entered into the HOQ as shown:
n k ed ents
Ra irem
u
Re q
Define
The benefit of Step 3 is viewing situations where a company’s product is
weaker than its competitors products as assessed by customers. This step
also uncovers which specific requirements to focus on for improvement. To
display this comparison, A represents the company and X, Y and Z the
competitors:
Com
pare
d R
equi
rem
ents
Define
Step 4: This section of the HOQ is typically completed by the product
design or engineering department. This information represents the design
elements that correspond to customer-stated needs. This process
transforms customer requirements into specific characteristics to be
designed into the product
Eng
ine
eri
ng
Cha
rac
t er
isti
cs
Define
Step 5: Determining the Relationship Between Customer Requirements and
Engineering Characteristics
Different indicators are used to depict the level of relationship between the
customer requirements and the design requirements. In the example below,
the following indicators are used:
• Strong relationship = 9 points
(reflected on HOQ as =)
• Moderate relationship = 3 points
(reflected on HOQ as O)
• Weak relationship = 1 point
(reflected on HOQ as X)
Define
Step 6: Comparing Competitor Characteristics with Design Engineering
Characteristics
In this step the engineering design
characteristics are compared
with those of the competitor
products.
Define
Step 7: Completing the HOQ Roof
Sometimes two product characteristics have a negative influence on each
other. For example, when one is increased the other is decreased. This is
recorded by placing an X in the
roof of the HOQ chart where
influence is anticipated
Expected Negative Influence Marked in HOQ Roof

Define
Step 8: Weighing the Engineering Characteristics
• It is important to weigh the characteristics in order to identify the amount of
importance each characteristic has on the level of customer satisfaction desired. The
weight of each characteristic is derived using the following formula:
Wj = sum(Wi x dij), where
• Wj = weight of characteristic j,
• Wi = rate of importance of requirement
• i (determined in Step 2)
• dij = point of relationship between
characteristic j and requirement i
Define
Step 9: Determining New Target Values for Characteristics
At this point in the completion of the HOQ, it is
possible to set new target values for the product.
These values are determined based
on the following items:
• Relative importance (Step 2)
• Ranking of competitors relative to
the satisfaction level of customers
(Step 3)
• Competitor product value (Step 6)
• Component influences at the roof
of house (Step 7)
Define
Kano Results and the HOQ
• The new target values in the last row of the chart provide
direction to set a strategy that matches the best competitor on
the most important customer requirements. To achieve the
goal, the design must consider the following challenges:
• Technical difficulty of designing and deploying the new
product
• Cost of changing to a new design
• Negative relationship shown in the roof of the chart indicating
potential difficulty in implementing specific requirements
Define
• The Kano Model of customer satisfaction is a useful tool to

help determine attractive or must-be customer requirements.
Results from the Kano survey feed the customer requirements
and needs section of the HOQ matrix.
Define
• CTQ Drilldown Tree
- Superficial CTQs (e.g. “easy business) are too vague to be made
actionable
- QFD requires that improvement take place at the most fundamental
level whereas CTQ drilldown requires in-depth analysis of customer
CTQs & as a tool can be used effectively to:
 Convert customer needs & requirements to measurable product or

service characteristics
 Establish linkage between Project Y & Business Y
 Make the Project manageable
Define
• CTQ Drilldown Tree – The Need
Situations in which we can make use of a CTQ Drilldown Tree are:
 CTQ Drilldown Tree is used to identify Process CTQs v/s Project
CTQs
 Frequently used in the Define Phase of the Project & helps to
integrate Project CTQs with business strategy
 Often helps us identify what we need to work & measure in order to
meet our CTQs consistently
 Using the CTQ Drilldown Tree we first measure where we are &
then identify targets to where we want to be
Define
• Defect Definition
- A Defect is defined as any variation of a required

characteristic in a product or its parts; or of a service, which is
far enough away/ removed from its target value
- Any variation of a required characteristic within a product or
service, that prevents that product or service from fulfilling the
specific requirements of the customer or internal quality
standards
- Anything requiring rework!
Define
Now we need to understand what an opportunity for making defects is.

Lets look at the possible ‘opportunities’ on a Credit Card Statement:
 The Credit Card Statement
 Transactions made on the card
 Customer Information
 Every word / letter / number
Potential definitions for a defect are:
 At least one error (any error) on a Statement
 Any error on the transactions made on the Card
 Any piece of incorrect customer information
 Any incorrect word / letter / number
Define
Some important guidelines for defining an opportunity are:

 It should be defined as how the customer is judging you
 Track what the customer actually feels throughout the process
 Determine whether you are measuring DPU or DPMO
Defect & Opportunity should be clearly

defined in the Problem Statement!!
Define
• Defect – Various Examples
Purchase Order Revisions Maintenance errors
Discrepancy reports Quality errors
Late job order releases Shortages
Ledger entry errors Wrong parts
Time card errors
Reworks
Audit findings
Safety errors
Late responses to customer requests
Define
Step 2 – Project chartering and story boarding
 Define the business case

 Develop the problem statement
 Develop the goal statement
 Assess the project scope
 Decide project milestones
 Project team selection
 Define team roles & responsibilities
Define
• Project Charter
Project Charter is one of the most important things necessary to get

a project team started on a solid footing. It helps to:
 Set expectations & obtain buy-in on scope, goal & resources (It is
documented, so that there is no misunderstanding)
 Accelerate the on-boarding of new resources (It is documented, you
don’t need as much time explaining your project)
 Avoid future scope from creeping into a Project (Refer back to the
Project Charter at all times to ensure that you are on the right track)
Define
• Project Charter Elements
 Business Case
 Problem Statement
 Goal Statement
 Project Scope
 Milestones
 Team Roles & Responsibilities
Define
• Business Case
A business Case helps the Management understand how the

Project is linked with the overall business objectives. It also
helps them make a decision about investing in the Project.
It should be able to address a few vital questions

Define
• Problem Statement
Quantitatively describes the current pain in the process &

answers the following questions:
 What is the pain?
 When & where is the pain?
 How much is the pain?
 What has the pain led to?
However it should not:

 Assign a cause or blame
 Include a solution
Define
• Goal Statement
The goal statement specifies the target that the Project team is
seeking to achieve collectively. It must:
 Start with a verb
 Have focus of the Project
 Have a measurable target (e.g. by 25%, by 70%)
 Have a completion date
It should not:
 Assign a cause or blame
 Include a solution
Define
. Project Scope
- One of the characteristics of a good Project is that it is
manageable & can be finished within 60-90 days
- While selecting a Project keep in mind the start & end point /
resource availability / “regular jobs” / other constraints
Define
• Milestones
- The project Charter should clearly mention the dates on which

each phase of the DMAIC cycle will be completed
- Should be aggressive
- Must include a detailed project plan along with a documented
communication plan
Define
• Project Team & Roles
Sponsor and/ or Champion Project Coach
Drive “vision” by sharing needs Drive “mission” by deploying teams
Define project need & scope Prioritise & strategise projects
Engage MBB for initiating Project Facilitate project management
Provide strategy & roadmap Train & mentor project teams
Orient teams towards support Coach on successful closures
Timely project progress reviews
Project Leader Team Member

Drive “results” by project execution Drive “tasks” to logical closures
Train & mentor teams on tools Provide subject matter expertise
applicability & usage Participate in project meetings
Plan & conduct team meetings help with administrative tasks
Ensure meeting of deadlines
Drive change & create buy-in
Define
• Project Team & Roles – ARMI
Used to assess each persons role in the Project during
various phases of the Project
Approver
Resource
Member
Informed
Define
• Project Storyboarding
- Identify the customer
- Define customers needs & requirements
- Specify deliverables tied to those expectations
- Identify CTQs for those deliverables
- Map the process
- Determine where in the process the CTQs can be most
seriously affected
- Evaluate which CTQs have the greatest opportunity for
improvement
- Define the defect to be attacked
Define
• Project Storyboarding & Risk Management
 Must be written with care & precision to avoid

misinterpretation by any third party
 Must reflect the effort of improving a good business
 Avoid blame of the current status & focus on improvement
 Don’t speculate…let facts / data speak
Define
Step 3 – Project acceptability and success factors
Conduct threat vs. opportunity Matrix
Identify critical success factors
Conduct stakeholder analysis
Define
• Project acceptability
Effectiveness of a solution =
Quality of the solution * Acceptability of the Solution
E=Q*A
Define
- Improvements bring about change & it is a commonly known

fact that change is always resisted
- It is important for all of us to realise that just finding a solution
is not good enough
- The identified solution should be understood & implemented
WIIFM!!
Define
• Project acceptability – Key Benefits
- Enable projects to be started & completed more quickly

- Help ensure that solutions are supported
- Helps ensure that customers & suppliers are getting involved
appropriately
- Team involvement in ensuring change sustenance
- Reinforcing change
- Helps drive change on a global basis
Define
Three commonly used tools to help create a shared need are:

 Threat vs. Opportunity Matrix
 Critical Success Factors
 Stakeholder analysis
Define
• Threat vs. Opportunity Matrix
The Recession
Threa
Or
Short Commonwealth Games

Scandal
Term
Define
• Critical Success Factors
- These are factors that are critical for the success of the
project & needs to be considered & tracked. Some of them
are:
 Appropriate metric
 Data availability
 Resource availability
 Proximity to Champion
 Degree of difficulty
 Scalabilty
 Benefits
Define
• Stakeholder Analysis
This enables the team to answer questions like:

 Who are the key stakeholders?
 Where do they currently stand on issues associated with this
change initiative?
 Where do we need them to be in terms of their level of support?
Define
• Stakeholder Analysis
Stakeholders Strongly M
Name Against A
Define
Step 4 - As-Is Process Mapping and SIPOC
Identify key suppliers, inputs & outputs to the process;
Conduct As-Is Process mapping;
Document the SIPOC
Define
• Process Mapping
 Process mapping is a graphic display of steps, events &

operations that constitute a process
 A pictorial illustration which identifies the steps, inputs &
outputs and other related details of a process by providing a
step-by-step picture of the process “as-is”
 It is a team effort & is documented by everyone
 A process map is a picture of the current steps in the process
targeted for improvement.
Define
• Process Mapping – Key Benefits
• Provides a structure for thinking through a complex process
• Creates a means for communicating knowledge across the
business
• Helps understand communication disconnects
• Helps clarify process reality from perceived / documented
plan
• Helps reveal unnecessary / redundant steps
• Magnifies “insignificant” / “overlooked” areas
Define
• Process Mapping – SIPOC
SIPOC as a tool displays a cross-functional set of activities in

a single & simple diagram which:
 helps us identify process inputs (Xs) & outputs (Ys)
 helps us identify the process owner, customers & suppliers
 helps us identify & establish boundaries for the process
Define
• SIPOC
The five key elements of this tool are:

Supplier – Whoever provides the input to your process
Input – The product / data that a process delivers
Process – Activities that you must perform to satisfy your
customers requirements
Output – Product / data that results from the successful
operation of a process
Customer – Whoever receives the output of your process
Define
• SIPOC Example:
S I
(Supplier) (Input
1. Scanned
1. Client docum ents
Define
• SIPOC
• A process map has five major categories of work from the
identification of the suppliers of the process, the inputs the
suppliers provide, the name of the process, the output of the
process, and the customers of the process. Each of these
steps is summarized as SIPOC to indicate to the team the
steps that must be conducted to complete a process map.
Measure
Measure
• The role of Measurement:
- A Measure describes the dimension, quantity, performance,

capacity or characteristic of a population. In our context, it
would now represent organised data.
Measure
Key questions at this Stage:
• What is the Process How does it function?

• Which Outputs affect CTQ’s most?
• Which Inputs affect Outputs (CTQ’s) most?
• Is our ability to measure/detect sufficient?
• How is our current process performing?
• What is the best that the process was designed to do?
Measure
• What is the Measure phase ?
The Measure phase is the second phase of DMAIC

methodology. The measure phase defines the defect(s),
gathers baseline information about the product or process, and
establish improvement goals.
• Objective
Assure that the current situation is understood in detail from a

variety of perspectives so that strategies can be developed to
address it.
Measure
• Why is the Measure phase important ?
The Measure phase allows you to understand the present

condition of the process before you attempt to identify
improvements. Because the Measure phase is based
upon valid data, it eliminates guess work about how well
your process is working.
Measure
• A Project should address some or all of the
following on Measures:
 What measures are used to assess the state of the problem &
the success of the Project?
 Are these measures already in place or they now being
defined / established?
 Is the measurement system understood in terms of its
accuracy, repeatability & reproducibility?
 What is the baseline data for these measures? What is the
extent of the problem? What is the gap between actual &
desired performance?
 Has the process been mapped, identifying key measurable
outputs & inputs per step?
Measure
• Types of Output Measures
Effectiveness Measures
- The degree to which customer needs & requirements are met & exceeded
- E.g. “You need to complete a transaction in 10 minutes and the overall
accuracy must be 95%”. In this case effectiveness is: Completing each
transaction in 10 minutes with 95% overall accuracy”
Efficiency Measures
- The amount of resources allocated in meeting & exceeding customer
requirements
- E.g. “You need to complete a transaction in 10 minutes and the overall
accuracy must be 95%”. In this case efficiency is: Completing each
transaction in less than 10 minutes with 95% overall accuracy”
If the TAT remains 10 minutes but your accuracy goes up; it is not efficiency!!
Measure
Relation of Measure to DMAIC
Collect (1) facts

Define
(2)VOC inform
Establish a bas
Measure
understand ho
Measure
Define Unit, Opportunity & Defect for the Process; Set Performance Targets
& identify specification limits; determine the Project Y
- Step 6 – Identify and prioritize possible Project X’s
Identify & Prioritise possible X’s driving defect occurrences
- Step 7 – Develop data collection strategy
Understand data types; develop data collection plan; develop data
sampling strategy; validate measurement system; execute the data
collection plan
- Step 8 – Calculate Process Capability
Understand variation to figure out Special & Common causes; interpret &
analyse the data graphically; establish Defect rate & Process Yield;
Establish process capability & calculate Baseline process Sigma
Measure
 Define Unit, Opportunity & Defect for the Process;
 Set Performance Targets & identify specification limits;
 Determine the Project Y
Measure
• Unit, Opportunity & Defect
 Unit
The item produced or processed relative to the Project Y as
reviewed by the customer
 Defect
Any Project Y measurement value that does not meet the “Y”
performance standard
 Opportunity
Any Y measurement event which provides a chance of not
meeting the performance standard
 Defective
A unit with one or more defects
Measure
• Performance Targets & Specifications
- A performance target is the requirement (s) and

specification is the requirement range, imposed by the
customer on a specific CTQ. It addresses the following:
 What does the customer want?
 What is a good process / product?
 What is a customer defect?
Measure
The goal of a performance Target & Specification is to
translate the customer requirements into a measurable
characteristic which has a defined:
 Operational definition
 Target value
 Specification limit
 Defect definition
Measure
Operational definition
 An operational definition is a precise description that tells
how to get a value for the characteristic (CTQ) you are
trying to measure
 It includes “what something is” & “how to measure it”
Target value
 A target value is the level or output characteristic of a CTQ
where the customer satisfaction is maximised in terms of
his requirements being met by the product or process
Measure
Specification limit
 A specification limit is the range on the target for output
characteristic of a CTQ within which the customer is
satisfied satisfactorily
 Is defined by the customer & is also called as
“performance standards” and/or “tolerance limits”
 Defect definition
 A defect is a customer experience that results in an
unacceptable level of customer satisfaction by the usage
of the product / process
Measure
• Determine the Project Y
Project Y is defined as
 An output characteristic of a process, as is felt by the
customer
 Determines how well the process meets the CTQs
 Is dependent on the inputs (Xs) to the process
Measure
• Performance metrics can be gauged from:
– Material shortage
– Downtime
– Absenteeism
– Quality level
– Setup time
– Transport time
– Delivery on time
Measure
. Key deliverables
Step 6 – Identify and prioritize possible Project X’s

 Identify & Prioritise possible X’s driving defect occurrences
Measure
• Three key process analysis tools that are used to
identify the casual Xs are:
 Cause & Effect Diagram
 Failure Modes & Effects Analysis
 Process Map Analysis
• Two key tools that are used to prioritise the casual

project Xs are
 Cause & Effect Matrix
 Control Impact Matrix
Measure
The Five Whys

Measure
• The Five Why’s
The standard rule of thumb is that if you ask why 5 times

you will usually get to the root cause of the problem
Why? Why? Why? Why? Why?
Problem Root cause

Measure
• The Five Why’s
Problem Statement: You are on your way home from work and
your car stops in the middle of the road.
1. Why did your car stop?
- Because it ran out of gas.
2. Why did it run out of gas?
- Because I didn't buy any gas on my way to work.
3. Why didn't you buy any gas this morning?
- Because I didn't have any money.
4. Why didn't you have any money?
- Because I lost it all last night in a poker game.
5. Why did you lose your money in last night's poker game?
- Because I'm not very good at "bluffing" when I don't have a
good hand.
Measure
Cause & Effect Diagram

Measure
• Cause & Effect Diagram
• A method of systematically identifying all of the potential causes
that may be contributing to the problem (i.e. Effect)
• A Cause & Effect Diagram also known as a fishbone or an
ishikawa diagram, classifies all the sources of variation into 6
categories (5M / 1P)
 Machine
 Material
 Mother Nature
 Measure
 Method
 People
Measure
• Cause & Effect Diagram – The method
• Call the team for a fishbone brainstorming session into a room & provide
the members with adequate Post-its
• Create a “Fishbone” structure on a white board
• The head of the fish contains the “problem”; & the bones of the fish
contain the possible “causes”
• The 5 M’s & 1P is a commonly used nomenclature
• Use the Affinity Model to get all the discussion findings into similar
groups
• Every member sticks post-its onto the relevant 5M’s & 1P
Measure
•Cause & Effect Diagram – The method
•Get all the Post its together & remove the duplicates. Discuss for
any conflicts of placing the post-it in wrong categories
•Check if the points identified are measurable or not; identify
measures for each of the factors (X’s) identified; relate the factors
(X’s) to the Project Y.
Measurement Methods
Machine
Potential
Effect “Y”
Causes (X’s)
Materials Mother Nature People

Measure
• Cause & Effect Diagram – Key Benefits
• It helps summarise potential causes for the problem

• Helps provide a visual display of possible potential causes
• Helps stimulate the identification of other potential causes
Measure
• Failure Modes & Effects Analysis
• An FMEA is a structured analysis tool for identifying ways &
methods in which a product or process can fail and then
attempts to plan an action to prevent those failures
 Failure Mode – It is a manner in which a part or a process
can fail to meet customer specifications. It is usually
associated with defects or non-conformities
 Cause – Causes are sources of variation which are
associated with key process inputs. Cause can be best
defined as a deficiency which results in a failure mode
 Effect – Effect is the impact on the customer if the failure
mode is not prevented or corrected
Measure
• FMEA Timing
The FMEA is a living document. Throughout the product
development cycle change and updates are made to the product
and process. These changes can and often do introduce new failure
modes. It is therefore important to review and/or update the FMEA
when:
 A new product or process is being initiated (at the beginning of the
cycle).
 Changes are made to the operating conditions the product or
process is expected to function in.
 A change is made to either the product or process design. The
product and process are inter-related. When the product design is
changed the process is impacted and vice-versa.
 New regulations are instituted.
 Customer feedback indicates problems in the product or process
Measure
• Benefits of FMEA
FMEA is designed to assist in improving the quality and reliability of
design. Properly used, the FMEA provides us with several benefits.
Among others, these benefits include:
• Improve product/process reliability and quality
• Increase customer satisfaction
• Early identification and elimination of potential product/process failure
modes
• Prioritize product/process deficiencies
• Capture engineering/organization knowledge
• Emphasizes problem prevention
• Documents risk and actions taken to reduce risk
• Provide focus for improved testing and development
• Minimizes late changes and associated cost
• Catalyst for teamwork and idea exchange between functions
Measure
• Failure Modes & Effects Analysis – The Method
• Select the process FMEA team & call them for a brainstorming
• Refer to the process map & list down all the process steps
• List all the value added process requirements that the process adds
• For each process step, list process inputs & rank the inputs
according to importance (C&E Matrix)
• Start the process FMEA; ask how each process step can fail (i.e.
the potential failure mode)
• For each process input, list ways that it can vary (causes) & identify
associated failure modes & effects
• List all other cause & associated failure modes & effects
Measure
Measure
• Identify Failure Modes. A failure mode is defined as the manner in
which a component, subsystem, system, process, etc. could potentially
fail to meet the design intent .
• A failure mode in one component can serve as the cause of a failure
mode in another component. Each failure should be listed in technical
terms. Failure modes should be listed for function of each component or
process step. At this point the failure mode should be identified whether
or not the failure is likely to occur. Looking at similar products or
processes and the failures that have been documented for them is an
excellent starting point.
• Describe the effects of those failure modes. For each failure mode
identified the engineer should determine what the ultimate effect will be.
A failure effect is defined as the result of a failure mode on the function
of the product/process as perceived by the customer. They should be
described in terms of what the customer might see or experience should
the identified failure mode occur. Keep in mind the internal as well as
the external customer
Measure
• Establish a numerical ranking for the severity of the effect. A
common industry standard scale uses 1 to represent no effect
and 10 to indicate very severe with failure affecting system
operation and safety without warning. The intent of the ranking is
to help the analyst determine whether a failure would be a minor
nuisance or a catastrophic occurrence to the customer. This
enables us to prioritize the failures and address the real big
issues first.
• Identify the causes for each failure mode. A failure cause is
defined as a design weakness that may result in a failure. The
potential causes for each failure mode should be identified and
documented
• Identify the causes for each failure mode. A failure cause is
defined as a design weakness that may result in a failure. The
potential causes for each failure mode should be identified and
documented
Measure
• Identify Current Controls (design or process). Current Controls
(design or process) are the mechanisms that prevent the cause of
the failure mode from occurring or which detect the failure before it
reaches the Customer. We have to now identify testing, analysis,
monitoring, and other techniques that can or have been used on the
same or similar products/processes to detect failures. Each of these
controls should be assessed to determine how well it is expected to
identify or detect failure modes. After a new product or process has
been in use previously undetected or unidentified failure modes may
appear. The FMEA should then be updated and plans made to
address those failures to eliminate them from the product/process
• Determine the likelihood of Detection. Detection is an assessment of

the likelihood that the Current Controls (design and process) will
detect the Cause of the Failure Mode or the Failure Mode itself, thus
preventing it from reaching the Customer.
Measure
• Review Risk Priority Numbers (RPN). The Risk Priority Number is a
mathematical product of the numerical Severity, Probability, and Detection
ratings:
RPN = (Severity) x (Probability) x (Detection)
The RPN is used to prioritize items than require additional quality planning or
action.
• Determine Recommended Action(s) to address potential failures that have a
high RPN
• Assign Responsibility and a Target Completion Date for these actions. This
makes responsibility clear-cut and facilitates tracking
• Indicate Actions Taken. After these actions have been taken, re-assess the
severity, probability and detection and review the revised RPN's. Are any further
actions required?
• Indicate Actions Taken. After these actions have been taken, re-assess the
severity, probability and detection and review the revised RPN's. Are any further
actions required?
Measure
• Process Map Analysis
With the SIPOC & as-is process maps in place, we

conduct a process map analysis to identify casual X’s.
Four key tools that can be used are:
 Nature of Work
 Moments of Truth
 The flow of work
 Disconnects Matrix
Measure
• Nature of Work
Nature of Work evaluates each and every step in the process
map in terms of it being a :
 Value Added – Steps that are considered essential to produce
and deliver the product or service to meet the customer’s needs
& requirements. Customer is willing to pay for this.
 Non Value Added - Steps that are considered non-essential to
produce and deliver the product or service to meet the
customer’s needs & requirements. Customer is not willing to pay
for this.
 Value Enabling - Steps that are not essential, but that allow the
value adding tasks to be done better / faster.
Measure
• Nature of Work
The main focus in Nature of Work analysis is on non value added work
& one must be on the watch out for all kinds of process steps beginning
with the prefix “Re”
Revise
Recheck Remake Reevaluate
Return Rework Redo
Redesign
Reject Repeat
Remeasure
Reship Reissue Retest
Recall
Retype
Measure
Moments of Truth
This is defined as the critical judgement of a customer about

the service delivery and can be drawn anytime based upon
a service experience (or lack thereof)
Measure
• The Flow of Work
o Helps us understand & identify what in our process s

causing a service delay that results in a customer wait
o Evaluates the cycle time of the process under two broad
categories:
 Process time
 Delay time
Process time
+Delay time
Cycle Time
Measure
• Disconnects Matrix
Speaks of identifying different types of disconnects which are
an integral part of the process at a given point in time & helps
prioritise disconnects for initiating improvement.
Speaks of identifying process disconnects in the form of:
 Process Gaps
 Conflicting objectives
 Redundancies
 Unclear requirements
 Tricky hand-offs
 Common problem areas
Measure
• Cause & Effect Matrix
Is a method of systematically prioritising all of the potential

causes that may be contributing to the problem (possible
project X’s)
It is a grid structure, also known as C&E matrix, which
numerically relates X’s to the Y’s
•
Cause & Effect Matrix
Measure
Contact to contact
with no recognition
Short order-to-
delivery time
Accurate order
fulfillment
High quality
components &
software
Rating of Importance
Measure
• Control Impact Matrix
Helps Prioritise Possible X’s
Measure
Step 7 – Develop data collection strategy
 Understand data types
 Develop data collection plan
 Develop data sampling strategy
 Validate measurement system
 Execute the data collection plan
Measure
Data collection plan:
Pre-Data Collection Steps
1. Clearly define the goals and objectives of the data collection
2. Reach understanding and agreement on operational
definitions and methodology for the data collection plan
3. Ensure data collection (and measurement) repeatability,
reproducibility, accuracy, and stability
During Collection Steps
4. Follow through with the data collection process
Post-Data Collection Steps
5. Follow through with the results
Measure
Step 1: Define Goals And Objectives
A good data collection plan should include:
• A brief description of the project
• The specific data that is needed
• The rationale for collecting the data
• What insight the data might provide (to a process being
studied) and how it will help the improvement team
• What will be done with the data once it has been
collected
• Being clear on these elements will facilitate the accurate
and efficient collection of data.
Measure
Step 2: Define Operational Definitions and
Methodology
• How many observations are needed

• What time interval should be part of the study
• Whether past, present, and future data will be collected
• The methodologies that will be employed to record all
the data
Measure
Step 3: Ensuring Repeatability, Reproducibility, Accuracy and
Stability
The data being collected (and measured) will be repeatable if the
same operator is able to reach essentially the same outcome
multiple times on one particular item with the same equipment. The
data will be reproducible if all the operators who are measuring the
same items with the same equipment are reaching essentially the
same outcomes. In addition, the degree to which the measurement
system is accurate will generally be the difference between an
observed average measurement and the associated known
standard value. The degree to which the measurement system is
stable is generally expressed by the variation resulting from the
same operator measuring the same item, with the same equipment,
over an extended period.
Measure
Step 4: The Data Collection Process
• Once the data collection process has been planned and
defined, it is best to follow through with the process from start
to finish, ensuring that the plan is being executed consistently
and accurately. Assuming the Black Belt or project lead has
communicated to all the data collectors and participants what
is to be collected and the rationale behind it, he or she might
need to do additional preparation by reviewing with the team
all the applicable definitions, procedures, and guidelines, etc.,
and checking for universal agreement. This could be followed
up with some form of training or demonstration that will further
enhance a common understanding of the data collection
process as defined in the plan.
Measure
• Step 5: After The Data Collection Process
• Referring back to the question of whether or not the data
collection and measurement systems are reproducible,
repeatable, accurate, and stable, the Black Belt or project
lead should check to see that the results (data and
measurements) are reasonable and that they meet the
criteria. If the results are not meeting the criteria, then the
Black Belt or project lead should determine where any
breakdowns exist and what to do with any data and/or
measurements that are suspect. Reviewing the operational
definitions and methodology with the participants should help
to clear up any misunderstandings or misinterpretations that
may have caused the breakdowns.
Measure
• Sampling Strategy
• Population is the total group of elements that we want to study

• Sample is the subgroup of the population that we actually
study, as to study each & every element is usually not
possible,
Measure
• Population & Sampling – The Purpose
Sampling is carried out to estimate & to test characteristics of
distributions and to develop explanatory or predictive models
showing the relationship between variables of the distribution.
Estimates can be categorised as:

• Point Estimates – Single numbers, unlikely to be exactly
correct of the parameter being estimated
• Confidence Intervals – ranges of values with a known
likelihood of containing the correct value of the parameter
being estimated
Measure
• Sampling Approach
Four key techniques that can be used to sample are:

• Random Sampling
• Stratified random sampling
• Systematic sampling
• Sub group sampling
Measure
• Random Sampling
• Here each unit of the population has the same chance of being selected while
drawing out a sample
• Random samples are used in population sampling situations when reviewing
historical or batch data. The key to random sampling is that each unit in the
population has an equal probability of being selected in the sample. Using
random sampling protects against bias being introduced in the sampling
process, and hence, it helps in obtaining a representative sample.
• In general, random samples are taken by assigning a number to each unit in
the population and using a random number table or Minitab to generate the
sample list. Absent knowledge about the factors for stratification for a
population, a random sample is a useful first step in obtaining samples.
• For example, an improvement team in a human resources department wanted
an accurate estimate of what proportion of employees had completed a
personal development plan and reviewed it with their managers. The team
used its database to obtain a list of all associates. Each associate on the list
was assigned a number. Statistical software was used to generate a list of
numbers to be sampled, and an estimate was made from the sample.
Measure
• Stratified Random Sampling
• Like random samples, stratified random samples are used in
population sampling situations when reviewing historical or batch
data. Stratified random sampling is used when the population has
different groups (strata) and the analyst needs to ensure that those
groups are fairly represented in the sample. In stratified random
sampling, independent samples are drawn from each group. The
size of each sample is proportional to the relative size of the group.
• For example, the manager of a lending business wanted to estimate
the average cycle time for a loan application process. She knows
there are three types (strata) of loans (large, medium and small).
Therefore, she wanted the sample to have the same proportion of
large, medium and small loans as the population. She first
separated the loan population data into three groups and then
pulled a random sample from each group.
Measure
• Systematic Sampling
• Systematic sampling is typically used in process sampling situations
when data is collected in real time during process operation. Unlike
population sampling, a frequency for sampling must be selected. It
also can be used for a population study if care is taken that the
frequency is not biased.
• Systematic sampling involves taking samples according to some
systematic rule - e.g., every fourth unit, the first five units every
hour, etc. One danger of using systematic sampling is that the
systematic rule may match some underlying structure and bias the
sample.
• For example, the manager of a billing center is using systematic
sampling to monitor processing rates. At random times around each
hour, five consecutive bills are selected and the processing time is
measured.
Measure
• Rational Sub-Grouping
• Rational sub-grouping is the process of putting measurements into meaningful
groups to better understand the important sources of variation. Rational sub-
grouping is typically used in process sampling situations when data is collected
in real time during process operations. It involves grouping measurements
produced under similar conditions, sometimes called short-term variation. This
type of grouping assists in understanding the sources of variation between
subgroups, sometimes called long-term variation.
• The goal should be to minimize the chance of special causes in variation in the
subgroup and maximize the chance for special causes between subgroups.
Sub-grouping over time is the most common approach; sub-grouping can be
done by other suspected sources of variation (e.g., location, customer, supplier,
etc.)
• For example, an equipment leasing business was trying to improve equipment
turnaround time. They selected five samples per day from each of three
processing centers. Each processing center was formed into a subgroup.
• When using sub-grouping, form subgroups with items produced under similar
conditions. To ensure items in a subgroup were produced under similar
conditions, select items produced close together in time.
Measure
• Sampling Bias
• Bias – Systematic differences between sample & the
population; occurs when these differences are introduced into
the sample as a result of the selection process resulting in a
non-representative sample and the subsequent results will not
be reliable.
Bias can enter at two levels:
• At a strategic level: while developing the sampling plan –
e.g. environmental sampling bias, convenience sampling bias,
systematic sampling bias, etc
• At a tactical level: while carrying out the sampling plan e.g.
measurement bias, non-response bias, etc
Measure
• Measurement System Analysis
• In order to ensure that the measurement system is sufficient
we formulate a plan to test for measurement system
variability:
• Data Repeatability – Variability exclusive of time, people
differences or other changing conditions
• Reproducibility – Variability inclusive of time, people
differences or other changing conditions
Measure
• Measurement System Analysis
Key parameters:
 Bias
 Repeatability
 Reproducibility
 Stability
 Linearity
 Correlation
 Tolerance
 Distinct categories
Measure
• Bias – If a standard scale measures the distance b/w point A &
point B as 10 ft, then the scale that the appraiser is using must
also measure the distance b/w point A & point B as 10 ft
• Repeatability – If an appraiser measures the distance b/w point
A & point B as 10 ft with a scale; given the same scale to the
same appraiser, he/she should measure the distance b/w A & B
as 10-ft again
• Reproducibility - If an appraiser measures the distance b/w
point A & point B as 10 ft with a scale; given the same scale to
another appraiser, he/she should measure the distance b/w A &
B as 10-ft again
• Stability - If an appraiser measures the distance b/w point A &
point B as 10 ft with a scale; given the same scale to the same
appraiser after a long interval of time, he/she should measure
the distance b/w A & B as 10-ft again
Measure
• Gauge Repeatability & Reproducibility (GR&R)
Is used to analyze the variation of components of
measurement systems to minimize any variation in the
measurement systems.
Measure
Step 8 – Calculate Process Capability
 Understand variation to figure out Special & Common causes

 Interpret & analyse the data graphically;
 Establish Defect rate & Process Yield;
 Establish process capability & calculate Baseline process
Sigma
Measure
• Process Variation – The types
• Common Cause variation is free of any assignable

causes and the observed variation represents the effect
of only random causes
• Special Cause variation has assignable causes present
and the observed variation represents the effect of non-
random causes
Measure
• Data Interpretation
 Stability (Run Charts)

 Normality (Anderson-Darling test)
 Shape (Skewed data)
 Spread
 Centering
Measure
• Stability is a measure of differences in the frequency
distribution which are expected & predictable for and
over a long period of time.
• Run Chart is an important tool for understanding data
stability and simple time ordered plots of process output
data
• Their patterns indicate special causes of variation
Measure
• Run Chart – Special Cause patterns
 Same Value Plot

 Clustering or too few runs Plot
 Mixtures or too many runs Plot
 Oscillations Plot
 Trends Plot
 Shifts Plot
Measure
Measure
• Centering – Spread, Shape & Normality
 Symmetrical Distribution
 Modal (Bi & Multi) Distribution
 Exponential (Skewed & Long Tailed) Distribution
Anderson Darling test is performed on the data to

understand its normality characteristics.
Measure
Negatively
skewed
Measure
• Process Capability
Process capability is the inherent variability of a

customer quality characteristic that the process is
capable of maintaining, when in a state of statistical
control, under a given set of conditions. Of 2 types:
 Short term capability

 Long term capability
Difference b/w both these is known as the Process Shift.

Measure
Normal Distribution, with Curve
Normal Distribution with Curve
45
40
35
30
y
c
n 25
e
u
q 20
e
r
F 15
10
-3 -2 -1 0 1 2 3
Norm al
Measure
• Measures provide information on
– current process performance.
– impact of changes to the process.
– signals of potential problems.
• Without measures, it is difficult to

– establish priorities and set realistic goals.
– assess whether changes in a process result in
improvement.
– identify causes of problems.
– determine if changes in the process are maintained over
time.
Measure
Examine the current situation Measure
Objectives: Questions:
Examine the • What can we learn from a
View the process from the detailed process flow ?
process flow problem’s • Are there obvious
in detail perspective by weaknesses ?
detailing how the
overall process
works.
• When does the problem
Describe the Analyze process occur ?
current data to find major • Where does the problem
situation with data sources of occur ?
variation. • Who is involved when the
problem occurs ?
A • What are the symptoms of
the problem ?
• How will the data be
gathered ?
ANALYSE
Analyse
• Use the gathered data to understand the problem statistically
and identify the area of focus for improvement
• Key Deliverables
- Step 9 – Define Performance Objectives
Baselining, Entitlement & Benchmarking
- Step 10 – Identify and validate Vital X’s
Establish Segmentation & Stratification Plan; Conduct process door
analysis; prioritise vital few Xs; conduct data door analysis; quantify
opportunity; conduct hypothesis testing; summarise significant Xs
Analyse
Step 9 – Define Performance Objectives
 Baselining, Entitlement & Benchmarking of a Process
Analyse
• Baselining is quantifying the goodness (or badness) of the current
process before ANY improvements can be made, using a sample data.
(Average long term performance of a process)
• Know how the best do it

• Benchmarking – "World class in process“
• Competitive comparison
• Best practices
Analyse
Step 10 – Identify and validate Vital X’s
 Establish Segmentation & Stratification Plan
 Conduct process door analysis
 Prioritise vital few Xs
 Conduct data door analysis
 Quantify opportunity
 Conduct hypothesis testing
 Summarise significant Xs
Analyse
Analyse
• Quantify Opportunity
– Pareto
 Histogram is useful for continuous data same way when the data is
discrete, most teams create a Pareto chart. Discrete data is counted
data - go/no-go, off/on, yes/no, and defect/no defect type data.
 An Italian economist Vilfredo Pareto,in the sixteenth century proved
mathematically that 80 percent of the world's wealth was controlled by
20 percent of the population. This 80-20 rule eventually proved
applicable in arenas other than economics.
 When dealing with discrete data, the project team should create reason
codes for why a defect occurs
and count and categorize the
data into these reason codes
and a pareto chart should be
prepared.
Analyse
• How To Construct A Pareto Chart
A pareto chart can be constructed by segmenting the range of the data into groups (also
called segments, bins or categories). For example, if your business was investigating the
delay associated with processing credit card applications, you could group the data into the
following categories:
• No signature
• Residential address not valid
• Non-legible handwriting
• Already a customer
• Other
• The left-side vertical axis of the pareto chart is labeled Frequency (the number of counts
for each category), the right-side vertical axis of the pareto chart is the cumulative
percentage, and the horizontal axis of the pareto chart is labeled with the group names of
your response variables.
• You then determine the number of data points that reside within each group and construct
the pareto chart, but unlike the bar chart, the pareto chart is ordered in descending
frequency magnitude. The groups are defined by the user.
• What Questions The Pareto Chart Answers
• What are the largest issues facing our team or business?
• What 20% of sources are causing 80% of the problems (80/20 Rule)?
• Where should we focus our efforts to achieve the greatest improvements?
Analyse
• Box Plot
A Box Plot is a graphical representation of a five number summary. The
ends of the "box" are the first and third quartiles, with the median inside the
"box". Points are used to represent the
maximum and minimum value, with lines (called
whiskers) connecting the points to the box.
Analyse
• Scatter Plot
Y Y Y
r =0 X r = -1.0 X r = 0.8 X
Y Y Y
r = 1.0 X r = 0.6 X r = 0.0 X

Analyse
• Hypothesis Testing
 Discuss and clarify the potential root causes.
 Clean-up the potential root causes by removing duplicates.
 Prioritize the cleaned-up potential root causes by using a
voting or ranking process. The team should reach consensus
on the final list of root causes.
 Develop a formal hypothesis for each root cause on the final
list.
Analyse
• Conclusion:
• Analysis can take several forms. Some Six Sigma
programs like to use a lot of diagrams and worksheets,
and others prefer discussion and list making. There are
many tools which can be used to perform analysis like
Box Plot, Cause and Effect Diagram, Progressive
Analysis, Ranking, Pareto Analysis, Prioritization Matrix,
Value Analysis etc. The proper procedure is the one that
works best for your team, provided that the end result is
successful.
IMPROVE
Improve
• Improve – Find possible solutions, plan for new measurements and
implementation strategies for fixing the quantified problem
Key Deliverables
Define Solution parameters; Generate possible solutions & refine solutions;
screen possible solutions against musts & wants; conduct cost benefit
analysis for potential solutions; assess risks for potential solutions
Select implementable solutions; implementation of solutions through a
pilot; validate measurement system; calculate improved process sigma; set
new interval team targets; conduct new process mapping
Improve
Key Deliverables
 Define Solution parameters
 Generate possible solutions & refine solutions
 Screen possible solutions against musts & wants
 Conduct cost benefit analysis for potential solutions
 Assess risks for potential solutions
Improve
• Solution Generation:
(1) Re-Express from Someone Else's Perspective - say, with the

mentality of a five-year-old or of an alien from another planet. With
this technique, team members drop all pre-conceived notions,
experience and knowledge about the problem and simply ask:
"What would I expect of a new mobile phone?" or "How would I
overcome long waiting hours at the medical clinic on Monday
mornings?" It might take the team some time to warm up to this
approach, but wait and see. By allowing everybody to say "stupid"
things, a project leader may start some great solution generation. As
Albert Einstein said, "If at first an idea is not absurd, then there is no
hope for it."
Improve
• (2) Related Worlds - Never assume that no one has ever faced an
issue like the one the team is facing now. And never assume there
is nothing valuable to learn from the immediate world.To harness
related worlds, the project leader should encourage every member
of the team to go out and see what can be learned from others. The
creative act occurs when someone applies what they discover to the
team's own challenge, but in a unique way.
• Roll-on deodorant illustrates this technique. The related world was
the ballpoint pen. Inventors of the roll-on deodorant "discovered" the
similarity between two situations where a liquid had to be spread
equally thin across a surface.
• Related worlds is a great technique for demystifying the creative
process. All one has to do is ask, "Where in the world has my
challenge been faced before?" and "What can I learn and steal from
that?" Yes, steal. "Originality is nothing but judicious imitation,"
according to Voltaire.
Improve
(3) Revolution - Revolution is creativity at its most provocative. Revolution is a
deliberate challenge of existing rules and assumptions. Some of the great creative
leaps of our time - "What if we could fly?" - have come from revolution. To use this
technique, start by writing down all the "rules." What is the shape, use, feel, touch,
application, process, etc.? Somebody, somewhere started with the rule that "Human
beings cannot fly." And then other people - among them, Leonardo da Vinci and the
Wright brothers - imagined breaking that rule.
• Take, for example, shampoo:
Table 1: Rule
• As illustrated in Table 1, once the rules are written down, the team can start to play
around with them: Try exaggerating, opposing, reducing and reversing as many of the
assumptions as possible.
Improve
(4) Random Links - Though no more effective than the other techniques,
random links may feel like the most creative one. It is the simple art
of selecting a random piece of stimulus that has nothing to do with
the creative challenge and then deliberately forcing a connection.
• This technique was used in a brainstorming session some years
ago with engineers in the mobile phone industry. Participants
collected ideas on new features, wrote them on cards and put them
on a pin wall. After some time the team realized it was stuck. To get
ideas flowing again, the session facilitators put the pin wall out of
sight, set up another one and asked the team to brainstorm ideas on
an advertisement randomly chosen from a magazine.
• The brainstorming produced words like "friendship," "eternity," and
"vacation." After a couple of minutes, the facilitators stopped this
process, brought back the first pin wall and asked the team to make
a connection between the ideas on the advertisement regarding the
original topic - new features for mobile phones. Some of the "crazy"
ideas that appeared on the pin wall in that session became reality!!
Improve
Key Deliverables
 Select implementable solutions
 Implementation of solutions through a pilot
 Validate measurement system;
 Calculate improved process sigma
 Set new interval team targets;
 Conduct new process mapping
Improve
• Improve
Improvements in existing systems are necessary to bring the
organization towards achievement of the organization goals.
Creative development of processes and tools brings about a new
lease on life for the organization’s processes and takes them nearer
to organizational objectives.
Various project management and planning tools can be used to
implement these new techniques and processes. Appropriate usage
of statistical tools is important to measure the data, which is
necessary to understand improvements done and any shortcomings
that may exist.
Improve
• If the project team does a thorough job in the root causation
phase of Analysis, the Improve phase of DMAIC can be quick,
easy, and satisfying work.
• The objective of Improve Phase is to identify improvement
breakthroughs, identify high gain alternatives, select preferred
approach, design the future state, determine the new Sigma
level, perform cost/benefit analysis, design dashboards/
scorecards, and create a preliminary implementation plan.
CONTROL
Control
• Control – Verify that implemented improvements have
fixed the problem and put procedure in place to continue
the verification on a periodic basis
Key Deliverables
- Step 13 – Process Control & Risk Analysis
Control Charts Deployment & Monitoring; Risk
assessment & mistake proofing
- Step 14 – Document and execute Control Plan
- Step 15 – Institutionalise and Sign-Off
Control
• (1) Quality control:
• The ultimate purpose in control is overall assurance that a high
standard of quality is met. The customer's expectations depend on
this, so control is inherently associated with quality.
• Since the purpose to Six Sigma is to improve overall process by
reducing defects, quality control is the essential method for keeping
the whole process on track; for enabling us to spot trouble and fix it;
and for judging how effectively the project was executed and
implemented.
• Quality is at the heart of the Six Sigma philosophy. Reducing
defects has everything to do with striving for perfection. Whether we
reach perfection or not, the effort defines our attitude toward quality
itself.
Control
• (2) Standardization:
• One feature of smooth processing is to enable processes to
go as smoothly as possible. This usually means
standardization. In a manufacturing environment, the value of
standardization has been proven over and over.
• We need to devise a control feature to processes so that the
majority of work is managed in a standardized manner.
Control
• (3) Control methods and alternatives:
• The development of a new process of any change to an
existing process requires the development of procedures to
control work flow.
• When a process cannot be managed in the normal manner,
we need to come up with alternatives short of forcing
compliance to the standardized method.
Control
• (4) Responding when defects occur:
• The final step in a control process is knowing how to respond
once a defect is discovered. The weak links in the procedure
where defects are most likely to occur, can and should be
monitored carefully so that defects can be spotted and fixed
before the process continues.
• The response to a defect may be to prevent a discovered flaw
from becoming a defect at all. In the best designed systems,
defects can be reduced to near zero, so that we may actually
believe that Six Sigma can be attained.
Control
• Conclusion:
• The project team determines how to technically control
the newly improved process and creates a response plan
to ensure the new process maintains the improved
sigma performance.
Minitab
New Minitab techniques:
Opening Minitab, navigating,
entering data, generating
random data, naming columns,
displaying descriptive statistics,
time series plots, control charts
and options, dot plots,
histograms, box plots, tiling
graphics.
Calc > Random Data > Normal
Generate 100 rows of data store
in C1, mean of 100, standard
deviation of 3.
Name C1”Normal 100-3”
Window > Info
Stat > Basic Statistics > Display
Descriptive > Select C1
Stat > Basic Statistics > Display
Descriptive > Select Graphs >
Select “Histogram of data, with
normal curve” > Close Graph
Graph> time series plot> select
C1
Project Selection Guidelines
 Clarify the Big Picture
In sync with business objectives, who is
accountable, who will execute, etc
 Strategic alignment through Projects
Look at improving customer satisfaction, cost,
capacity & top end growth
 Identify & improve key performance metrics
Defects / cost savings / Quality / Performance / TAT

Six-Sigma Class 1 - 12

Transféré par

Informations du document

Description originale:

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Six-Sigma Class 1 - 12

Transféré par

Droits d'auteur :

Formats disponibles

SIX-SIGMA

- Understanding customer requirements

Quality is all about

Focus on cost Focus on customer requirements

BRAND EROSION! BRAND ENHANCEMENT!!

Cost improvements to Quality improvements to

Goals Defect Reduction Streamlining Process Alignment

Deployment Bottom-up Top-Down Top, Middle & Bottom

Key Feature Quality Circles Outside Consultants Internal Experts

Continuous Radical Redesign Align & Maintain

- Our survival is dependent upon continued growth of the business

Growth, Customer satisfaction, quality, knowledge and

Excellence is not an act but a habit!

Businesses operate roughly around 3 σ

• General Electric: Over $10-15 bn savings!

• AlliedSignal: Over $2.2 bn savings!

“ Past definitions of quality focused on

“…this Six-Sigma journey will change the

Measure of Capability – the values Measure of Variation - The

The symbol σ is also the 18th letter in the Greek Alphabet!!!

6σ is not 3 times but 90735 times better than 2σ !!!

Hospital Surgery 5000 incorrect surgery 2 incorrect surgery

Air Travel 2 abnormal landings at most 1 abnormal landing every 5

Sometimes 99% is just not good enough!!!

500 incorrect surgery procedures per week

22,000 cheques deducted from wrong bank accounts each hour

50 newborn babies dropped at birth by doctors each day

32,000 missed heartbeats per person, per year

One hour of unsafe drinking water every month

Even 99.9% is just not good enough!!!

“ Consistently delighting customers with

- We don’t know what we don’t know

Measure everything that results in customer

Variation may be the spice of life…but for a customer,

A journey, not a destination!

Attitude & Discipline

• Values being a complex of beliefs, ideals or standards,

Our values drive the behaviour that we demonstrate!!!

Our environment drives everything above

• Discipline is to get accustomed to a regular and

• Attitude in Quality is all about being focused on

• Discipline in Quality deals with the process. It’s about

“ A positive attitude is contagious and like a

- Data is a set of numbers which do not convey any logical meaning

- Information is a set of numbers which when represented correctly

- Statistics is the science that compiles, analyses and interprets

 Descriptive Statistics : This branch of statistics organises, summarises

When observations, i.e. measurements for a variable, are

 Common cause variation: Differences in the measurements

 Special cause variation: Differences in the measurements

Measurement variation is natural, expected and is the foundation of statistics!!

Discrete Continuous Binary Non-Binary

- Binary: A set of data arising out of a logical decisioning which is

- Non-Binary: A set of data arising out of a logical decisioning which

Discrete Continuous Binary Non-Binary

Histogram Bar Diagram

Age Interval Frequency 14

Month Tickets Tic k e ts Is s u ed

Social Status No. Of People 30 26

Four important process output characteristics that need

The sample mean is represented by

- It is defined as the most frequently occuring value in the dataset