LEAN SIX SIGMA GREEN BELT TRAINING PROGRAM

ANALYZE PHASE
 Analyze Phase

Analyze Phase – Root Cause(s) are Identified

 Analyze Phase – Study about the
‘Function’

Y = f(X)
• •
• •
• •
• •

Outputs
Inputs Process:
Y1, Y2, Y3…Yn
X1, X2, X3…Xn What happens to
the inputs (f)?
 Cause and Effect Diagram
The Cause & Effect, or Fishbone diagram, was first used by
Dr. Kaoru Ishikawa of the University of Tokyo in 1943 –
hence it is also called as “Ishikawa Diagram“.

A Cause and Effect diagram is a visual tool to identify,

explore, and graphically display possible causes related to
a problem or condition.
 Cause and Effect Diagram

1
Problem/
Desired
Improve
Ask “Why?” to 3 ment
identify detailed Detailed
causes; add these Causes
details as small
branches off major
bones

Causes
Shows various
influences on a
2 process to identify
most likely causes
of problems
 Case Study: Fishbone Diagram
Step 1: Identify Effect

Example: The EFFECT is Poor Gas Mileage

 Case Study: Fishbone Diagram
Step 2: Identify cause influencing the Effect

Example: The possible CAUSES for Poor Gas Mileage

 Case Study: Fishbone Diagram
Step 3: Identify detailed levels [Potential Causes]

Example: The detailed CAUSES for Poor Gas Mileage

Example:
 Quality Tools
• Histogram

• Pareto Chart

• Run Charts

• Scatter Plot
 Histograms
• The Most common graphical method used to present quantitative data
is the histogram
• Data should be sorted into an appropriate number of classes
determined by the size of the data set
• Classes should have the same class width, except for the largest and
smallest classes, which may be left open to collect outliers
• Classes must be continuous and span all possible data values
• Although histograms are very useful for displaying large data sets they
are less useful for smaller sets, for which other methods should be
considered (such as Box Plot)
 Histogram – A Picture of Variation
Time to Fill up orders
45
40
35
30
Frequency: 25
Number of 20
Orders Filled 15
10
5
0
1 2 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31
Number of Weeks Orders Were Open

Helps In visualizing process variation for continuous Data.

 Pareto chart
Pareto Diagram is the Ideal Tool to select the Vital Few
from the Trivial Many Items

Pareto Diagram was First thought out by an Italian Economist,

Pareto (1843-1923), for National Income Analysis

After him Juran, (American Quality Control Authority) used it in

the field of Quality Control.

It is said that in reality, 20% of the Causes Contribute to

more than 80% of a Problems. These 20% causes are called
“Vital Few" and the remaining causes are Trivial.

This is called as the 80-20 Rule.

Exercise – Pareto Chart
 Pareto Diagram
Following are the data points for the no. of Prescription error over the period
Oct 2011 to Dec 2011

Prescription_Error Error Count

Draw a Pareto to find
Allergy drug prescribed 10
out the vital few
Diabetic advice not added in summary 8 causes for
Dosage form wrongly mentioned 3 Prescription Error.
Double time prescribed same brand 5
Drug duration not mentioned 9
Drug name wrongly prescribed 3
Drug frequency not mentioned 14
Drug not added in drug chart 16
Prescribed to wrong bed number 2
Strength not mentioned 13
Typing pool error 18
 Pareto Diagram
1. The first step is to arrange the data in descending order
and to find the Cumulative Figures
Cumulative
Prescription_Error Error Count Error % Error %
Typing pool error 18 18% 18%
Drug not added in drug chart 16 16% 34%
Drug frequency not mentioned 14 14% 48%
Strength not mentioned 13 13% 60%
Allergy drug prescribed 10 10% 70%
Drug duration not mentioned 9 9% 79%
Diabetic advice not added in summary 8 8% 87%
Double time prescribed same brand 5 5% 92%
Dosage form wrongly mentioned 3 3% 95%
Drug name wrongly prescribed 3 3% 98%
Prescribed to wrong bed number 2 2% 100%
101 100%
 Pareto Diagram
2. Next Scale the Left Axis & Draw Columns to mark the Individual figures
 Pareto Diagram
3. Next Mark the Cumulative Points in the Right Axis on a 100 Scale & Plot
the Line
 Pareto Diagram
4. Draw a Vertical Line where the Cumulative Percentage is 80%
 Run Charts
• Run chart as the name suggests typically is a plot which runs against time or
sequence
• A normal pattern for process in control is one of randomness. If only common
causes of variation exists in your process, the data will exhibit random behaviour.
• Why use Run Charts?
– To study observed data for trends or patterns over a specified period of time.
– To focus attention on truly vital changes in the process
– To track information for predicting trends
– To understand variation of process
– To compare a performance measure before and after implementation of a
solution
– To detect trends, shifts, and cycles in the process
 Run Charts

• Run charts show history and trends of variation

• After change, they show whether improvement is permanent
• X – axis is time or observation order
• The middle line is median.
• Requires minimum of 25 data points
Scatter Plot

• Scatter plots relate two variables

• They show whether there is any

association between them.
 Scatter Plots
• A scatter diagram is one of the best graphical methods for investigating a
possible relationship between two variables, especially when one is a
suspected cause and the other is the known effect

• Correlation can be of the following types:

– Positive Correlation
– Negative Correlation
– Mystery Correlation/No Correlation
 Strength and Direction of “+”
Correlation
Moderate positive correlation

110

100

90

Output
80

70

60

50

40

50 60 70 80 90 100
Weak positive correlation Input Strong positive correlation
110
85
100

90

Output
75 80
Output

70

60
65

50

40
55
30

40 50 60 70 80 90 100 110 120

40 50 60 70 80 90

Input Input
 Strength and Direction of “-”
Correlation
Moderate negative correlation
110

100

90

Output
80

70

60

50

40

50 60 70 80 90 100
Weak negative correlation Input Strong negative correlation
110
85
100

90

Output
75 80
Output

70

60
65

50

40
55
30

40 50 60 70 80 90 100 110 120

40 50 60 70 80 90

Input Input
Thank You!