What is data?

Data is a set of values of subjects with respect to qualitative or quantitative variables.

Types of Data:
Two types of quantitative data, which is also referred to as numeric data: continuous and
discrete.

Continuous data of Six Sigma measure phase is also known as variable data.

Discrete Data of six sigma is also known as Attribute data

The Poisson Distribution can be used to model discrete data.

The normal distribution is symmetric and has a skewness of zero. normal distribution is a

continuous probability distribution

If skewness is less than -1 or greater than 1, the distribution is highly skewed. If skewness is
between -1 and -0.5 or between 0.5 and 1, the distribution is moderately skewed.
If skewness is between -0.5 and 0.5, the distribution is approximately symmetric.

The skewness for a normal distribution is zero, and any symmetric data should have a
skewness near zero.

A positive mean with a positive skew is good, while a negative mean with a positive skew is
not good. If a data set has a positive skew, but the mean of the returns is negative, it means
that overall performance is negative, but the outlier months are positive.

If skewness = 0, the data are perfectly symmetrical.

Normality test can be performed:

During hypothesis testing, a normality test must be conducted against the data set for a 1
sample t-test or sets for a 2 sample t-test to determine which test to use during
hypothesis testing.
P-Value for Normality test:
After you have plotted data for normality test, check for P-value. P-value < 0.05 = not normal.
Note: Similar comparison of P-value is there in Hypothesis Testing. If P-value > 0.05, fail to
reject the H0.

Ordering in six sigma:

Stability and normality and capability

Cpk depends on your customer - the Voice of the Customer.

Cpk also requires input from the customer for the lower specification limit (LSL) and upper specification limit (USL).

The minimum value of "k" is 0 and the maximum is 1.0. A perfectly centered process will have Cp = Cpk.

Example One
Recall that the specification limits are set by the customer or your company. You may have only the USL and/or the
LSL depending on your situation. 

Let's assume the following information:

Lower Specification Limit (LSL): 5mm

Upper Specification Limit (USL): 10mm

The mean of your data set consisting of 50 observations is normal and the value is 7.85mm. The standard deviation
is 0.23mm

Find Cpk:

Find both values by substituting the values into the formula at the top of this page

1: (10mm - 7.85mm) / 3*0.63mm = 2.15mm / 1.89mm = 1.14

2. (7.85mm - 5mm) / 3*0.63mm = 2.85mm / 1.89mm = 1.51

The minimum value is used; therefore, the Cpk = 1.14

SMART stands for:

 S: Specific and Simple

 M: Measurable, Meaningful, and Motivational
 A: Achievable, Attainable, and Action-Oriented
 R: Relevant and Reasonable
 T: Time-based and Trackable

critical-to-quality (CTQ)
 A process that operates with “true six sigma” performance takes up 50% of
the specification if centered. This gives it a Cpk and a Cp of 2.0. A process such
as this will produce defects at a rate of only ~2 parts per billion.

DMAIC (Define—Measure—Analyze—Improve—Control)
DMADV (Define, Measure, Analyze, Design, Verify)

CPK value for the process to be stable:

Cpk = or >1.33 indicates that the process is capable and meets specification limits. 

Cp and Cpk are called Process Capability. Pp and Ppk are called Process
Performance. In both cases we want to try to verify if the process can meet to
meet Customer CTQs (requirements).

Cpl = (Process Mean – LSL)/(3*Standard Deviation)

Cpu = (USL – Process Mean)/(3*Standard Deviation)

Cpk is merely the smallest value of the Cpl or Cpu denoted:  Cpk= Min (Cpl, Cpu)

Stability can be measured:

Process stability can be easily determined using control charts.

Six sigma usage:

Six Sigma helps reduce the number of defects and aims to decrease waste. This helps lower
production costs. Improving Efficiency – Six Sigma increases process capacity by reducing
defects and material waste and enables organizations to do more with the same amount of
resources

Cp and Cpk, commonly referred to as process capability indices, are used to define the ability
of a process to produce a product that meets requirements

hypothesis testing, a normality test must be conducted

A normality test is a statistical process used to determine if a sample or any group of data fits a
standard normal distribution. A normality test can be performed mathematically or graphically

Anderson-Darling Normality Test: After you have plotted data for normality test, check for P-
value. P-value < 0.05 = not normal. ... If P-value > 0.05, fail to reject the H0.

The X-Y Matrix is a Six Sigma tool mostly used during the DMAIC measure phase and the
DMADV measure phase to show the relationship between X and Y factors.

Pareto chart identifies – cause and x factors

The Six Sigma expert uses qualitative and quantitative techniques to drive process
improvement.

Histogram - A bar chart that shows frequency distribution or variation in a data set.

 Which are the 7 QC tools?
7 Basic Quality Tool Templates
 Cause-and-effect diagram template (Excel)
 Check sheet template (Excel)
 Control chart template (Excel)
 Histogram template (Excel)
 Pareto chart template (Excel)
 Scatter diagram template (Excel)
 Stratification template (Excel)

A prioritization matrix is a tool used for determining the most important issues or solutions. 

The term Special Cause Variation was coined by W. ... Edwards Deming and is also
known as an “Assignable Cause.” unexpected variation that results from unusual occurrences

Common Cause Variation, also referred to as “Natural Problems, “Noise,” and

“Random Cause” was a term coined by Harry Alpert in 1947

Common cause variation can be predictable and controllable

Control charts are used to monitor two types of process variation, common-cause

variation and special-cause variation.

These are USL (Upper Specification Limit) and LSL (Lower Specification Limit).

Calculate cpk:

The formula for the calculation of Cpk is Cpk = min(USL - μ, μ - LSL) / (3σ) where USL and

LSL are the upper and lower specification limits, respectively. A process with a Cpk of 2.0 is
considered excellent, while one with a Cpk of 1.33 is considered adCTequate.
Control Charts for Variables
The control charts of variables can be classified based on the statistics of subgroup summary plotted
on the chart.

 X¯ chart
 R ChartCTQ
 S Chart
X¯ chart describes the subset of averages or means, R chart displays the subgroup ranges, and S
chart shows the subgroup standard deviations

Control and specification limit:

Control limits are calculated from process data for a particular control chart. An X-bar chart and an
Individual measurements chart will have different limits.
Specification limits are chosen in numerous ways. They generally apply to the individual items being
measured and appear on histograms, box plots, or probability plots.
The table below contrasts control limits and specification limits:

Control Limits Specification Limits

Voice of the process Voice of the customer
Calculated from Data Defined by the customer
Appear on control charts Appear on histograms
Apply to subgroups Apply to items
Guide for process actions Separate good items from bad
What the process is doingWhat we want the process to do

DPMO and Defective:

Defective: A defective unit; a unit of product that contains one or more defects with respect to
the quality characteristic(s) under consideration. Defects per million opportunities (DPMO): The
actual number of defects occurring divided by the total number of opportunities for a defect and
multiplied by 1 million
Defects allowed in six sigma:
3.4 defects per one million opportunities

93.3% accuracy – 3 sigma
99.99966% accuracy – 6 sigma
4 Sigma: 6.2K errors per million (99.4% accuracy). 
5 Sigma: 233 errors per million (99.97% accuracy)
99.7 % of accuracy can be maintained for + or -3 std deviation

..

What is an acceptable sigma level?

The right specification limit is at 4.5 sigma from the mean with a defect rate of 3.4 parts
per million (PPM).

Control limits are based on process variation.

the acceptable limits of variation equates to what one would expect to see in a random
process 99.73% of the time.

The standard deviation is calculated as the square root of variance

A Control Chart is also known as the Shewhart chart since it was introduced by

Walter A Shewhart.

SMED: Single Minute Exchange of Die

A p-value less than 0.05 (typically ≤ 0.05) is statistically significant. ... A p-value higher than
0.05 (> 0.05) is not statistically significant and indicates strong evidence for the null hypothesis.

Which statistical tool is use in Six Sigma?

Pareto Chart
1. Pareto Chart. The Pareto Chart stems from an idea called the Pareto Principle,
which asserts that about 80% of outcomes result from 20% of the causes

Selection of appropriate statistical method depends on the following three things: Aim and

objective of the study, Type and distribution of the data used, and Nature of the observations
(paired/unpaired). Selection of statistical test depends upon our aim and objective of the
study

The histogram is a popular graphing tool. It is used to summarize discrete or

continuous data that are measured on an interval scale.
Types of control chart:

Variable: (continuous)

X bar, Range R control, S Std deviation control charts

Attribute control charts: (discrete)

U and C chart
P and np chart
Control charts dealing with the number of defects or nonconformities are called c charts (for
count)

Control charts dealing with the proportion or fraction of defective product are called

p charts (for proportion).

There is another chart which handles defects per unit, called the u chart (for unit).

DMAIC – determine relationship b/w y and x

ProjectManager.com is the Perfect Tool for

Your Six Sigma Management
Relation between Control Limit and Specification Limit
1) When Specification Limit Falls between Control Limit.
As demonstrated in the given Graph. When Control limits exceeds the
specification Limit some part of the process will be functioning outside the
Specification range. That Means Whenever Range of control limits is greater
than range of specification limit; your process will be producing defects. Control
Chart
2) When Specification Limit Coincides with Control Limit
When Control limits will fall on Specification limits, all the variation in the
process which are common to the process will meet customer specification as
explained in the below graph. The serious challenge in this design will be to put
your system under statistical control because any variation from uncommon
source will produce Defect.
Control Chart 1
3) When Control Limit falls within Specification
This is probably the best design your process can have. In such design all the
variation in your process that comes from the common source will fall in
specification area. Also, Unlike Case no: 1 & 2, even if your process shifts over
time (i.e., control limits of your process shifts) then it will not have severe
impact on its ability to produce products/services that meets customer
specification
Control Chart 2
4) When there is no intersection at all between LSL and USL
Below graph shows that the complete process is working out of specification i.e.
complete process is defective. This situation appears often when your system is
not at all by virtue of its design able to meet customer specification. So in such
situation you need to diagnose your system. There can be only two possibilities.
Either the process is performing normal but your customer’s expectation from
the process is wrong or the system is not able to meet customer’s expectation
by huge deficiency (Design)

Yes, Control limits and Specification limits are different.

Control limit depends on inherent variation with in the process

Process capability index – cpk

Risk Priority Number, or RPN

Failure Modes and Effects Analysis (FMEA)

 The lower and upper con- trol lines, labelled LCL and UCL, are placed at a distance of three
standard deviations from the centre line.

Control limits are calculated by:

1. Estimating the standard deviation, σ, of the sample data.
2. Multiplying that number by three.
3. Adding (3 x σ to the average) for the UCL and subtracting (3 x σ from the
average) for the LCL

Control limits are calculated from your data. They are often confused with specification limits which
are provided by your customer.

A Cpk value less than 1.0 is considered poor and the process is not capable. A value between
1.0 and 1.33 is considered barely capable, and a value greater than 1.33 is considered capable.
But, you should aim for a Cpk value of 2.00 or higher where possible.

The Cpk is inversely proportional to the standard deviation, or variability, of a process

CPK <1.00 (Poor, incapable) 1.00< CPK <1.67 (Fair) CPK >1.67 (Excellent, Capable)

Cpu = (Process mean - LSL)/ (3 * Standard deviation) where the LSL is the lower specification
limit. Cpl = (USL - Process Mean)/ (3 * Standard deviation) where the USL is the upper
specification limit

Both Cpu and Cpl take into account where the process is centered. The value of Cpk is the
difference between the process average and the nearest specification limit divided by three
times the standard deviation
The key values are called a five-number summary, which consists of the minimum, first
quartile, median, third quartile, and maximum.

 It was introduced by American engineer Bill Smith while working at Motorola in 1986. Jack

Welch made it central to his business strategy at General Electric in 1995.