Professional Documents
Culture Documents
LSSGB Course Handout
LSSGB Course Handout
INTRODUCTION
00 Intro 1-2 © July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 2
2
Six!Sigma!Practically- A!Management!Philosophy
Yes or No?
A
12
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
So!Leaders!Must!Understand!WHY!It!Works
qAlignment!with!the!strategic!priorities!of!the!
business,!generally!with!a!sharp!focus!on!financial!
results.
D
² Top!Executive,!Champions
M
qDedicated,!well!trained!problem!solvers:!
A
² Black/!Green!Belts,!Green!Belts
I
C
qwell!architected!support!system
² Six!Sigma!Infrastructure!
qThe!powerful!Tool!Box
²RDMAICR!process!of!variable!reduction.
00 Intro 13 3
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Six!Sigma!Statistically?
Accuracy = Centering Precision = Spread
D
M
A
I
C
Spread Spread
Customer target
Customer target
00 Intro 1 - 15 © Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 15 4
Six Sigma Focus
REDUCE VARIATION
1 - 16 © Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 16
LSL USL
ss sss s
Process!Variation=!0.50!x!SpecificationTolerance
00 Intro 1 - 17 © Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved.17 17 5
Sigma Level- Statistically : A performance Measure
3 66,807 93.32%
4 6,210 99.38%
5 233 99.98%
6 3.4 99.99966%
1 - 18 © Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved.18 18
Six!Sigma!for!Black/!Green!Belt
Expectation
D
M
A
Performance
00 Intro 19 6
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Six!Sigma!for!Black/!Green!Belt
Products
Customer
CTQ – Critical to Quality
CTD – Critical to Delivery Expectation
CTP – Critical to Price
D
M
Price/
Quality Delivery
Value
A
Cycle
Defects Cost
I
Time
C
Processes
CTP1 – Critical to Process 1
Organization
CTP2 – Critical to Process 2 Performance
CTP3 – Critical to Process 3
20
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Six!Sigma!for!Black/!Green!Belt
q
(Define) (Measure-Analyze)
A
I
C
Capability
A
23
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Process!Performance:!Six!Sigma!Lingo
Benchmark
M
A
I
Entitlement
00 Intro 24 8
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Six!Sigma!Ultimately
$
D
(ROI)
M
Customer
Satisfaction
A
I
Yield Improvement
C
Defect Reduction
25
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Six!Sigma!:!From!where!it!came?
Period/ Year Event
00 Intro 26 9
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Key!roles!and!responsibilities
LEVEL ROLES TRAINING
Executive §Determine the scope ,identify goals 1 day
§Identify priorities, drive the use of 6s
D
§Support Exec./Champions
§Leader of Large scoped projects
Black Belt §Lead Project team/,Practitioner of DMAIC 4-5 weeks
27
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
END!OF!MODULE
31
00 Intro 10
Certified Lean Six Sigma Green Belt
DEFINE PHASE
Define Phase 11
Lean!Six!Sigma!– GB
Define!Phase
Define
Objectives 1. Define
01 Define phase 12
Define Road Map
Issues, Opportunities,
Strategic Goals
Initial Charter:
Problem/Opportunity,
Business Case, Objective
` Initiate Change
1. Define Management Efforts
S I P O
C
Document Targeted
Process
Clarify Customer Ensure Alignment with
& Requirements other Projects
01 Define phase 13
Identifying Potential Improvement Needs
Sources of Project Ideas & Examples
Project Filtering
Example
Low
A. Poor space utilization in buildings
B. Purchasing system errors
C. Accounting errors
D. Inventory systems do not
Level of Effort
High
Low High
Impact of Benefit
01 Define phase 14
Project Selection Criteria
1. Define
Meaningful
• Impact on key customers
• Impact on strategic plan
• Financial benefit — costs AND revenue
- Immediate
- Long-term
• Urgency or sequence
Manageable
• Resources needed
• Expertise available
• Organization “buy-in”
• Likelihood of success
• Learning experience opportunity
• Project duration
Key Handoff
From Leaders to Team 1. Define
01 Define phase 15
Solution “Parking Lot”
01 Define phase 16
Getting Started
1. Define
Project Charter
1. Business Case
2. Problem/Opportunity
3. Scope
4. Metrics
5. SMART Objective
6. Estimated Benefits
7. Constraints & Risks
8. Team Members & Time
9. Milestones
Project Charter
Project Title: <<short & clear name for the project, no technical terms>>
Problem/Opportunity Statement
Essential Description of “What’s Wrong”
Examples …
01 Define phase 18
Problem/Opportunity Statement Elements
Evidence
Initial/ Prospective customers have called to complain
Anecdotal that no one has contacted them.
Impact
Initial/
Anecdotal We risk losing business and hurting profits.
20
*This is still early and the Problem Statement will likely
evolve over the course of the project.
01 Define phase 20
Scope
What’s in? / What’s Out?
Scope Tool
In/Out of Frame
In-Scope Major IT
Reasons System
for Changes
Admits Format Cost/Effici
of Records ency of
Design of Record Design of Treatment
Forms Handling On-line Codes
Primary Tools
Care
Processes
Who
Needs
Ongoing Who
Records
Process Receives
Measures Records
Clinical
Practices
Out of Scope
01 Define phase 21
Metrics
How will we observe/quantify improvement?
Example …
Good: Not Good:
Reduce late (after commit date) Orders have to be submitted
custom footwear deliveries to <__ both electronically and via fax.
%. Implement improvement by Eliminate causes of late orders
end Q3. Confirm project results as soon as we can.
end Q4.
01 Define phase 22
Estimated Benefits: Hard or Soft
Initial Projections based on Problem & Objective
Soft impacts generally drive hard ones, but not in a proven, 1:1 ratio...
• Constraints – Limitations on
resources and/or capabilities that
may influence the Design project and/or solutions
E.g. Space, money, people, talent, time ...
01 Define phase 23
Team Members and Time
Who are Key Contributors?
01 Define phase 24
Milestones & Preliminary Plan
Some Key Questions:
Milestones
Target date for each
DMAIC phase Kick-off: 15 July 2016
Define: 2 Aug 2016
Measure: 18 Aug 2016
Analyze: 3 Sep 2016
Improve: 15 Oct 2016
Control: 15 Dec 2016
Targeted end date: 31 Dec 2016
Preliminary Plan
• Key tasks/actions, with assigned
responsibilities and dates
MEASURE
Chris B. March 9 March 12
• Prepare data collection plan: Output & Process
measures Mary W. & Pat H. April 1 April 2
• Complete initial data collection
Kumi T. April 9 April13
• Present refined Charter to SLT
ANALYZE
Chris B. April 12
• Confirm Process Improvement strategy (vs. April 13
Redesign)
Mary W. April 20
April 21
• Develop initial cause hypotheses
• Complete cause verification José R. May 3
01 Define phase 25
Customer
Project Charter Focus
End of Module
Contact
01 Define phase Confidential business and proprietary information of Givaudan, may not be copied or distributed to anyone without the express written permission of Givaudan
26
Minitab Introduction
and Basic Statistics
Learning Objective
qMinitab Introduction, data entry, data
Management
qPerformance Analysis Tools
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
27
MINITAB
INTRODUCTION,
DATA ENTRY &
DATA MANAGEMENT
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Module Objectives
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
28
Minitab Windows: Types of windows
l Data window
l Graph window
MINITAB’s Default
windows
l Session Window (top)
Minimized window
l Project Manager window
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
29
Data Windows Details
Data Formats
When you enter data into MINTAB, the program will read it as one of three formats:
hNumeric
hText (alphanumeric)
hDate/Time
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
30
Minitab Projects and Worksheets
} Differences between projects and worksheets
qProjects:
} Can be opened using the toolbar
} Cannot be opened into other projects
} Cannot be opened using Open Worksheet
q Worksheet files (.MTW) contain:
²Data window (1)
q Worksheets cannot:
²Retain the last settings used in the dialog box
²Retain any windows except the data window
²Be opened using the toolbar
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Edit Last
Command
Click this button
to return to the
last dialog box in Project Manager Brings the
which you made Toolbar, opens Brings the Project
any changes. the Project Session Brings the Manager to
(Ctrl+E) Manager Window window to current the front,
and the identified the front, and worksheet to and makes
folder makes it the the front, and it the active
active makes it the window.
window. active window. (Ctrl+I)
(Ctrl+M). (Ctrl+D)
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
31
Minitab Windows
Session window
Project Manager
Data window
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Data Entry
Common Method
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
32
Data Entry
How to do?
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Data Menu
q Data Menu:
m Data ØStack ØStack Blocks of Columns
m Data ØCode ØNumeric to Text
m Data ØChange Data Type ØText to Date/Time
m Data ØStack ØStack Rows
m Data ØExtract from Date/Time ØTo Text
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
33
Reorganizing Data
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
34
Stack Blocks of Columns
3. Click
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
3. Click
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
35
Change Data Type
Minitab Output
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
MODULE TOPICS
§ Descriptive Statistics
§ Store Descriptive Statistics
§ Graphical Summary
§ Bar Chart
§ Pie Chart
§ Time Series Plot
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
36
Bar Chart
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Bar Chart
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
37
Pie Chart
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
38
Understanding Statistics
What is statistics?
26
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
What is statistics?
q Six Sigma Belt- Common Tasks
mTo predict behaviour of the process
mTo compare the behaviour with the target
mTo identify the causes and link with the process
behaviour etc…
Population vs Sample
q Population :
Set of all possible outcomes of the process
q Samples:
Subsets of populations
q Data is obtained using samples because
m we seldom know the entire population
m Population statistics are desired, but often not
available
m Samples approximate the population
q Descriptive statistics apply to samples
02 Intro to Minitab 29 40
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Statistics- Practically
30
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
m Minimum
m Maximum
m Count
m Range
m Mean
m Median
m Sigma etc…
02 Intro to Minitab 31 41
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
MEAN: Centre of Gravity
The Mean of ‘n’ values is the total of the
values divided by ’n’
x1 + x2 + x3.....xn
x=
n
In Standard Mathematical Notation it is
i -n i -N
xi xi
x= n estimates
N m=
i -1 i-1
The symbol S” means ‘sum of’
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Median
Organize the data from low values to high when determining the median
02 Intro to Minitab 33 42
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Median
• The 1/4th data point is called
‘1st Quartile’ (Q1) or ‘25th Percentile’
• The 3/4th data point is called
‘3rd Quartile’ (Q3) or ’75th Percentile’
• Difference of Q3 and Q1 is called ‘Inter Quartile Range’
Examples
Odd data set: 1, 2, 3, 4, 5, 6, 7
The median is “4”
Q1 is 2 and Q3 is 6
34
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Mode
Examples
Data Set: 45 47 49 51
46 47 49 52
47 47 50 53
47 48 51 54
Mode is: 47
02 Intro to Minitab 35 43
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Standard Deviation
s =
å i
( x - x )2
estimates s = å i
( x - x ) 2
n -1 N
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Variance
s 2
=
å (x - x) i
2
n -1
•Square of Standard deviation is called VARIANCE
•Numerator is called ‘ Sum of Squares’
•Denominator is called ‘Degrees of Freedom’
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
44
Descriptive Statistics
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Descriptive Statistics
List
Box
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
45
Descriptive Statistics
Step 5: Click OK
Session Window Output
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. surendra@qgspl.com
Unpublished proprietary work available only under license. All rights reserved.
46
Graphical Summary
1. Choose Stat
ØBasic Statistics
ØGraphical
Summary
2. In Variables:
enter Sales.
3. . Click OK
surendra@qgspl.com
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
HISTOGRAM
Histograms give a graphical view of the
distribution of the values
D
5
I
C
FREQUENCY 3
64 65 66 67 68 69 70 71 72
02 Intro to Minitab
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
47
Important Population Attribute
For a normally distributed process:
68.26%
D
M
95.44%
A
I
C
99.73%
-4 -3 -2 -1 0 1 2 3 4
44
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Summary
q Minitab Windows
Five windows which allow you to:
m Store data (Data window)
m Generate analyses (Session window)
m Generate graphs (Graph window)
m Summarize the Data window (Columns folder)
m Store commands used to generate output (History folder)
q Data Entry
m Column names can be up to 31 characters
m Can be typed, imported, copied/pasted or brought in from a
stored Minitab file
m Preview and Options are available to customize data
imported from non-Minitab files
02 Intro to Minitab 45 48
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Certified Lean Six Sigma Green Belt
MEASURE PHASE
Measure Phase 49
Lean!Six!Sigma!– GB
Process!Mapping
1-1 1
Process Analysis
Process!step!analysis
Process!output!analysis!(CTC)
Process!input!analysis!(CTP)
– Process!Flow!diagram!+!CTC-CTP!map!(I-O!sheet)
– Value!stream!map!(Lead!time)
03 Process1 -Mapping
3 3 50
Process!Mapping!–
CTC-CTP!Analysis
Pivotal!Resources,!Inc
Module Objectives
Show!where!process!mapping!fits!into!the!strategy!for!
success
Show!the!connection!between!process!mapping!and!C&E!
matrices,!FMEAs,!and!control!plans
Define:
– Actual!process!vs.!Documented!process
– Value-added!!vs.!Non-value-added!(VA!vs.!NVA)
– Process!inputs!and!outputs!(X’s!&!Y’s)
Demonstrate!
– Macro-map
– Process!flow!diagram!(PFD)
– CTC-CTP!Detailed!map
Work!through!a!process!mapping!exercise
03 Process1 -Mapping
5 5 51
Types of Process Maps
SIPOC/Macro-map.
– A!macro-map!is!a!high!level!process!flow!chart.
– It!usually!shows!several!steps!on!both!sides!of!the!project!scope.
Process!flow!diagram!(PFD).
– A!process!flow!diagram!is!a!detailed!map!of!every!step!in!the!
process,!including!the!hidden!factory.
– The!process!flow!diagram!uses!standard!symbols!for!readability!
and!a!common!nomenclature.
Input/Output Worksheet.
– A!detailed!table!that!contains!all!of!the!process!details:!set!
points,!input!and!output!specifications,!etc.
– The!worksheet!also!contains!estimates!of!dpu,!RTY,!COPQ,!cycle!
time,!etc.
1-7 7
S I ® P O ® C
X
X “Requirements”
(Key Process
X Indicators)
X
Macro-map SIPOC
Suppliers Inputs Process Outputs Customers Requirements
•CNC lines •Cut glass •Tempered Glass •Final customer •Size as per final
•VQ lines •Consumables gauge
•Glass supplier •BCP ink • Appearance as per
-AIS Roorkee limit sample
D
-AMG • Curvature & Cross
-AIS Taloja Curvature as per
M A
•Ink supplier standard
-Ferro
-Johnson
•Print screen
maker
I
C
Start
End
Cut Glass Logo Glass Glass Heating Quenching Unloading
Inspection Packing
Receiving Marking Loading & Bending
• Background:
• An engineer for the Inspection Yields
Bearing Assembly
process in a Bearing 100
Manufacturing plant 98
noticed that the process 96
yield has gone down.
Yield
94
• A Six Sigma expert was 92
assigned to identify the 90
causes of these 88
differences and to 86
Week Week Week Week
reduce the defects.
1 2 3 4
Shift 1 Shift2
Target Process
Fail
Readjust Fail
Inspect Inspect
Inner race and Balls
Pass
Pass
Ship
• Cross-functional or
“swim lane” map
Flow Chart
Flow Chart
Different icons/symbols to indicate the different
types of actions in the process:
§Activities : §Connector :
§Movement :
§Storage :
How to create?
1. Observe for a few repetitions the process to be
charted.
2. Write down all the steps taken and decisions made in
the process.
3. Mark the path of the Flowchart from the beginning to
the end
4. Return to the beginning of the path and repeat Step
3 for any paths that branch off from the main path
(at the decision points)
5. Record the last step at the bottom of the page, draw
an ellipse around it, and connect the primary path
and any branching paths to the last step
6. Review for accuracy
Unpublished proprietary work
available only under license. 25
All rights reserved.
Start
Check for
Inject Grease
Clean Outer Shell Inspect Balls Insert Balls Any Ship
in Shell
Contamination
Stop
Fail for contamination Fail
Remove Balls &
Clean again
Start
SOP 954-21 SOP 954-32 SOP 954-32 SOP 954-44
D D
SOP 954-21 Check for
Inject Grease
Clean Outer Shell Inspect Balls Insert Balls Any Ship
in Shell
Contamination
Check for
Inject Grease
Clean Outer Shell Inspect Balls Insert Balls Any Ship
in Shell
Contamination
Check for
Inject Grease
Clean Outer Shell Inspect Balls Insert Balls Any Ship
in Shell
Contamination
Receive PO <999kg
Planner PO Size
Send RM
FG ready in
from
Warehouse warehouse
pallets
Small Batch
Compounding
Compounder
Large Batch
Compounding
>1000kg
Small Batch
Packing
Packer
Large Batch
Packing
03 Process1 -Mapping
32 32 61
Construct the Detailed Map
Identify!the!X’s!and!Y’s,!specs.,!C!or!N,!capability!and!equipment.
VA or I/O C or
Process Step NVA SOP # Inputs & Outputs Type Specifications N Capability Equipment
Clean Outer Shell NVA 954-21 Cleaned Outer shell Output No visible contam. IPY=92% Ultra-Wash
Solvent type Input IPA C
Solvent Contam. Conc. Input < .05% N
Soak Time Input 300 s C
Bath Temp. Input 110 F C
Scrub Brush Type Input Accord # 45 C
Scrub Brush Age Input < 5 weeks N
1 - 33 33
Glossary
Process!map: Control!variable:
– A!graphical!representation!of!an!actual – A!variable!or!an!input!which!can!be!readily!
process. changed!to!measure!the!effect!on!an!output!
(Y),!knob!settings,!materials,!etc.
Input/Output!Worksheet
– A!table!with!information!for!process!inputs! Noise!variable:
and!outputs!that!is!beneficial!to!improving! – A!variable!or!input!that!is!difficult!to!control,!
the!process. ambient!temperature!or!humidity,!machine!
differences,!customer!state!differences,!etc.
Value-added!step:
– An!operation!that!transforms!the!product!in! Documented!process:
a!way!that!is!meaningful!to!the!customer. – The!process!as!defined!by!the!controlling!
documents!(process!specifications,!work!
Customer: instructions,!training!manuals,!etc.).
– Defined!by!the!scope!of!the!project.!!The!
customer!may!be!the!end!user,!another! Actual!process!and!the!hidden!
internal!organization!or!a!later!process!step. “factory”:
Output!(Y): – The!process!as!it!is!in!the!“factory”,!
– Output!à any!item!or!feature!on!a!product! including!rework,!equipment!preferences,!
that!is!defined!as!critical!by!the!customer. undocumented!inspection!criteria,!
“experience”,!work-arounds,!etc.
Input!(X):
– Input!à item!that!has!an!impact!on!an!
Ideal!process:
output. – The!process!as!you!would!like!it!to!be,!
simple,!efficient,!defect-free,!and!without!
any!non-value-added!steps.
03 Process1 -Mapping
36 36 62
Process Capability
04 Process Capability 2 63
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Measure Types
Continuous/Discrete
Continuous Discrete/Attribute
• Number of orders filled
• Time to delivery
• Number of days lead
• Cost or price time
• Percent of on-time • Type of product
orders*
• Number of ingredients
3-3
04 Process Capability 4 64
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Batch 1: Batch 2:
08:00, 09:00, 10;00 08:50, 09:10, 09;30
Batch 3:
09:15, 09:14, 09:16
Average: 09:15
5
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
A Simple Definition
q Capability metrics report how well process output features meet
customer specifications
q EXAMPLE:
m PPM (Parts Per Million) reports the number of defective parts expected in a
group of 1,000,000, and thus the defective and compliance rate.
² Example: 100,000 PPM reflects 90% compliance to specifications
m Cp (sē-pē) reports the ratio of the specification range to the natural range of
process output
² Example:
v Temperature Spec range: 68 to 72 0F
v Process Temperature Range: 69 to 71 0F
v Cp = (72-68)/(71-69) = 4/2 = 2.0
04 Process Capability 6 65
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Capability Calculation : Cp/ Cpk
q Complaints are occasionally received by customers about the
temperature level in a waiting room.The environmental control system
for the room is designed to maintain temperature between two user-
defined values. Since the system was installed, temperature levels
were recorded hourly for each of ten hours per day.One of the
business partners who was trained in Six Sigma collected a year’s
worth of data from the records for analysis.
7
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
04 Process Capability 8 66
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Why Measure Capability?
q Enables allocation of resources based on DATA!
(This is not common!)
q Quantifies the defect rate
q Identifies opportunities for improvement
q Assessing capability allows the organization to predict its
true quality levels for all its goods and services
q Identifies the nature of the process problem – centering or
spread
9
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
04 Process Capability 10 67
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Step 4: Compute Cp
Process Potential Capability(Cp):
Product Tolerance USL - LSL
Cp = =
Process Tolerance 6s st
Product Tolerance
q Example:
m A process mean is 325,
standard deviation is 15,
an upper spec. limit is 380 and
lower spec. limit is 270
-5 -4 -3 -2 -1 0 1 2 3 4 5 q What is Cp?
q What is Cp if the mean is 355 but
-3s +3s the standard deviation does not
Process Tolerance change?
11
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
A Problem With Cp
qWhich is the better process?
qWhat is the difference in Cp between the two
processes?
qWhat can be done to make Cp more effective as
a process capability statistic?
-5 -4 -3 -2 -1 0 1 2 3 4 5 -5 -4 -3 -2 -1 0 1 2 3 4 5
04 Process Capability 12 68
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Step 4: Meet Cpk – Process Performance
-5 -4 -3 -2 -1 0 1 2 3 4 5
13
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
é X - LSL USL - X ù
Cpk = min ê , ú
ëê 3 ´ s st 3 ´ s ú
st û
= 0.74
123 124 125 126 127
04 Process Capability 14 69
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Defects vs. Defectives
q What is a defective?
m A unit that contains at least one defect.
q What is a defect?
m An output of a process that does not meet a defined
specification:
² concentration, time, accuracy, satisfaction index, etc.
15
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
40%
20%
m dpu = 1-IPY
q For multiple defect opportunities per unit:
m IPY = e-dpu
m dpu = -Ln(IPY)
04 Process Capability 18 71
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
DPMO: “Normalizing” for Complexity with Opportunities
defects/Unit
DPMO = ´ 1,000,000
Opportunities/Unit
19
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Primary Generate/Send
Physician Request for
Visit X-Ray
04 Process Capability 20 72
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
X-Ray Calculations Example
IPY = ____ IPY = ____ IPY = ____ IPY = ____ FTY = ____
dpu = ____ dpu = ____ dpu = ____ dpu = ____ dpu = ____
RTY = ____ RTY = ____ RTY = ____ RTY = ____ RTY = ____
Process Totals:
dpu = ________
Exercise: Calculate the missing quantities and RTY = ________
select the first area of focus for the Green DPPM = ________
Belt project. (Hint: Assume all of the DPMO = ________
opportunities counts are PASSIVE in this SigmaLevel = ________
example.)
21
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Process Totals:
dpu = å dpu k = 0.035 + 0.203 + 0.094 + 0.011 = 0.343
RTY = Õ IPY = (. 966 ) ´ (. 816 ) ´ (. 911 ) ´ (. 989 ) = . 71
k Where should the
DPPM =
defectives
´ 1,000,000 =
50
´ 1,000,000 = 111,111 Green Belt focus
# produced 450
his energy?
dpu .34282
DPMO = ´ 1,000,000 = = 42,852
Opp. / unit (3 + 2 + 2 + 1)
04 Process Capability 22 73
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Certified Lean Six Sigma Green Belt
ANALYZE PHASE
Analyze Phase 74
Tools for Determining Potential Cause & Effect
Module Objectives
q Introduce Fishbone Diagrams
q Introduce Cause and Effect Matrix Diagrams
05 C & E Tools
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 2 75
Common & Special Causes
Type of Cause will Often Appear in Data
COMMON
• Variation/effects created by normal factors in the process
• Random, normal, chronic, more frequent
Response: Look closer at process data, procedures, normal
environment
SPECIAL
• Variation/effects created by factors outside normal operation
• Non-random, unusual, sporadic, less frequent
Response: Avoid over-reacting, follow process and data to
pinpoint unusual conditions/causes
05 C & E Tools
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 4 76
Processing? Transactional?
Policies Place
4 P’s
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 5
Finance Finance
Account Charge Personnel Distributes Departme
Documentation Enter Departmental nt Heads
Sent to Finance Charge on Expense Review
System Reports Report
Departmental
Heads Request
Change
05 C & E Tools
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 6 77
Fishbone Diagrams in Minitab Fishboneexample.mtw
q Open Minitab.
q Select STAT > Quality Tools > Cause-and-Effect…
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 7
05 C & E Tools
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 8 78
Correcting Expense Account Posting
q Create a simple fishbone from the Expense macro-process map.
Cause-and-Effect Diagram
Measurements Material Personnel
O utdated guides
Error key punching
D ifferent dept
definitions
C omputer glitch
O utdated procedures
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 9
05 C & E Tools
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 12 80
C-E MATRIX DIAGRAMS
q Benefits
m The team agrees on the cause priorities.
m The C&E matrix priorities are customer driven.
?
05 C & E Tools
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 15 81
C_E Matrix
High
A. Cause-1
B. Cause-2
C. Cause-3
D.Cause-4
Frequency
E. Cause-5
F. Cause-6
G. Cause-7
H.Cause-8
I. Cause-9
J. Cause-10
K. Cause-11
L. Cause-12
Low M.Cause-13
Low High
Impact
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
05 C & E Tools
© Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. Pg 17 82
Graphical!Methods
Quality!Growth!Services
Module Objectives
u Introduce!a!few!methods!of!graphical!analysis
– Box!plots
– Main!effects!Plot
– Interval!Plot!and!
– Interaction!plots
– Scatter!Plot
06 Graphical tools
1 - 2© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 2 83
Process Variation
Intentional Unwanted
y = f Process ( x1 , x2 ,..., xk ) + f Noise (n1 , n2 ,..., nk )
u Noise!variation!from!discrete!inputs
– Different!employees,!systems
– Different!days,!shifts
– Different!batches,!inputs
u Noise!variation!from!continuous!inputs
– Time!of!day
1 - 4© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 4
06 Graphical tools
1 - 5© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 5 84
Graphical Analysis – Example
u A!company!is!facing!long!cycle!time!of!testing!of!
product.
u The!company!expects!its!Experts!to!test!120!items!
every!2!hrs.!
u The!following!data!is!collected:
– Number!of!items!tested!per!2!hr period
– Month
– Day!of!Week
– Name!of!the!Expert
– Time!of!Day
1 - 6© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 6
u What!is!the!capability?
u Are!these!good!or!bad!values?
06 Graphical tools
1 - 7© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 7 85
Capability Analysis
u Run!Stat!>!Quality!Tools!>!Capability!Analysis>Normal…
LSL
P rocess D ata Within
LS L 120.00000 Overall
Target *
USL * P otential (Within) C apability
S ample M ean 152.61944 Cp *
S ample N 1440 C PL 0.33
S tDev (Within) 32.89894 C PU *
S tDev (O v erall) 33.51791 C pk 0.33
C C pk 0.33
O v erall C apability
Pp *
PPL 0.32
PPU *
P pk 0.32
C pm *
1 - 8© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 8
u Construct!a!Box!plot!for!Lines!by!Loan!Officer
– In!Minitab!select!Graph!>!Box!plot…
06 Graphical tools
1 - 9© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 9 86
Box Plot Output
u Notice!that!box!plots!do!NOT!require!“numeric”!x!
variables.!!This!plot!will!accept!“categorical”!text!
variables.
Boxplot of Apps_Rev vs LO
250
200
Apps_Rev
150
100
1 - 10
© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 10
First Quartile
u Construct!a!Box!plot!for!Lines!by!Loan!Officer!and!Time!
of!Day
– In!Minitab!select!Graph!>!Box!plot…
1 - 12
© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 12
u Plots!averages!of!the!data!at!the!factor!levels
– Stat!>!ANOVA!>!Main!Effects!Plot…
190
Mean of Apps_Rev
180
170
160
150
140
130
Fred Joe Mary Sam EarlyAM LateAM EarlyPM LatePM
06 Graphical tools
1 - 13
© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 13 88
Interaction Plots
u An!interactions!plot!is!a!plot!of!
means!for!each!level!of!a!factor!
with!the!level!of!a!second!factor!
held!constant.!!
u Parallel!lines!indicate!no!
interaction!is!present.
Interaction Plot (data means) for Apps_Rev
EarlyAM LateAM EarlyPM LatePM
LO
200
qStat > ANOVA > Interactions Plot… Fred
Joe
180
Mary
Sam
LO 160
140
120
Time of Day
200
EarlyAM
LateAM
180
EarlyPM
LatePM
160
Time of Day
140
120
1 - 14
© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 14
The (w)Rap!
u List!your!key!takeaways! u How!can!I!use!this!info!in!
from!this!module. my!project?
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
06 Graphical tools
1 - 18
© Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 18 89
Test of Hypothesis
Introduction
Module Objectives
07 Test of Hypothesis 4 90
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
The Purpose of Inferential Statistics
q The purpose:
m To draw conclusions (inferences) about populations from
measurements of samples
5
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Hypotheses
q Hypotheses:
m Hypotheses are statements about population parameters, not
statements about samples.
q The Null Hypothesis:
m Abbreviated as H0
m Usually H0 is a statement of no effect or no difference.
m We reject or fail to reject H0 based on statistical evidence.
q The Alternative Hypothesis:
m Abbreviated as Ha
m A statement about a population parameter that is suspected of
being true, if H0 rejected
07 Test of Hypothesis 6 91
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Order in the Court
q Hypothesis testing is like
the Indian legal system
where a defendant is True State of Nature
assumed innocent until Did Not Commited
proven guilty. Commit Crime A Crime
mH0: Defendant is
Commit Crime
Did Not
Guilty
innocent (Assumed) Correct
Goes
Verdict
mHa: Defendant is guilty Free
Verdict
(Must be proved)
Commited
A Crime
Innocent
Correct
Person
Verdict
Convicted
?
7
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Correct Type II
Ho
Error Decision
07 Test of Hypothesis 8 92
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Means Tests
Module Objectives
q To demonstrate four hypothesis tests for means
m One sample z/t test
m Two sample t test
m Multisample Test- One way ANOVA
07 Test of Hypothesis 11 93
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Testing Means
mtarget
q Testing for one mean Ho: m = mtarget
m Z-test for large samples or Ha: m < mtarget
when s is known m > mtarget
m t-test for smaller samples or m ¹ mtarget
when s is unknown m
12
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Population 1 Population 2
-6 -4 -2 0 2 4 6 8
14
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
07 Test of Hypothesis 15 95
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
1-sample t-test – Minitab Results
Test of mu = 85.5 vs mu < 85.5
Variable N Mean StDev SE Mean
Trained 15 80.79 8.93 2.31
Variable 95.0% Upper Bound T P
Trained 84.86 -2.04 0.030
N Mean StDev
New System 10 44.30 4.64
Std System 10 46.50 5.40
07 Test of Hypothesis 17 96
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Analysis of Variance
ANOVA One-way –
Hypothesis testing for multiple means
m1 m2 m3 m4
QGS-Business Excellence Division
Introduction to ANOVA
07 Test of Hypothesis 20 97
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Introduction to ANOVA
One way ANOVA:
If in the experiments if the effect of various
levels of a single factor is to be compared,
it is called one way ANOVA.
How is it done?
07 Test of Hypothesis 22 98
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
One Way ANOVA
q One way ANOVA form two mean sum of squares,
m one for treatments (MST) and
m the second for error (MSE).
qThese are typically displayed in the tabular form, known
as an ANOVA Table.
SOURCE ANOVA TABLE
df SS MS F
Treatment l-1 SS(T) SS(T)/(l-1) MS(T)/ MSE)
23
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
ANOVA Introduction
q Example:
m A rubber manufacturing company wants to compare
effects on rapid modulus with four new formulations
against the existing formulation.
m Ten different batches were sampled for each formulations
and the average RM was computed.
q Problem:
m Were any of the formulations different from each other and
the existing formulation?
q Case 1
Trial TmtA TmtB TmtC TmtD TmtCntrl1
1 6.6 7.0 6.1 7.5 5.5 q Case 2
2 6.4 7.0 5.9 7.5 5.7
Trial Tmt1 Tmt2 Tmt3 Tmt4 TmtCntrl2
3 6.5 7.0 6.0 7.6 5.5
1 6.0 5.5 6.4 6.4 7.6
4 6.6 6.8 6.1 7.4 5.4
2 4.3 8.3 5.3 7.1 4.9
5 6.5 7.1 5.9 7.4 5.5
3 4.7 4.8 6.7 8.8 6.3
6 6.6 7.0 5.8 7.4 5.5
4 6.7 6.8 6.9 8.8 5.3
7 6.5 6.9 6.0 7.6 5.4
5 5.0 8.2 6.8 6.0 2.2
8 6.6 6.9 5.9 7.7 5.5
6 4.9 8.2 5.0 7.4 6.5
9 6.7 6.9 6.1 7.5 5.5
7 7.6 6.3 7.2 9.1 5.7
10 6.5 6.9 6.1 7.6 5.7
8 7.9 7.8 6.1 5.5 6.0
Mean 6.6 7.0 6.0 7.5 5.5
9 7.2 6.6 6.4 9.4 7.4
10 8.2 10.1 5.6 8.6 3.0
Mean 6.3 7.3 6.2 7.7 5.5
H 0 : m1 = m 2 = ... = m j
H a : @ least one m j is different
29
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
The (w)Rap!
q Key Takeaways q How can I use this in my
m ANOVA is a powerful tool for project?
determining the differences m ______________________
between multiple means m ______________________
m ANOVA compares the
between-group variability to the
m ______________________
within-group variability m ______________________
m The assumptions of ANOVA m ______________________
are subgroup normality and m ______________________
equal variances
m ______________________
m Others:
m ______________________ m ______________________
m ______________________ m ______________________
m ______________________ m ______________________
Quality!Growth!Services
Module Objectives
u To!introduce!simple!linear!modeling
u To!demonstrates!methods!in!Minitab!for!
constructing!models
u To!practice!regression!techniques
u To!explain!the!difference!between!correlation!&!
regression
u Used!for!quantitative!variables!(X’s!and!Y’s)!
u For!review:!!What!is!the!focus!of!Six!Sigma?
Y = f (x )
Q. What!does!this!equation!represent?
A. A!mathematical!model!of!a!process
u Purpose!of!Regression:!!to!predict!Y!from!a!setting!of!x
u Examples:
– Revenue!=!f(sales!rep!experience,!#!of!client!visits,!length!of!client!visit)
– Loan!Default!Rate!=!f(credit!rating,!monthly!income,!age)
– Time!=!f(#!tasks,!complexity,!queue!length!)
1 -©3Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 3
y = b 0 + b1 x
24 Dependent
22 Variable
20
18
Intercept Slope
16
is the value of = Rise/Run
14 y(x=0) 6 =6/3=2
12
10
8 3
y = 4 + 2x
6
4
2
0
0 1 2 3 4 5 6 7 8 9 10
1 -©5Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 5
u An!engineer!is!trying!to!find!the!
causes!of!Low!breaking!strength.!
One!of!the!suspect!causes!are!the!
number!of!twists!in!the!linking!rod! Twists x Strength y
Data!are!collected!related!to! 1 500
twists!and!stength 3 1750
u Construct!a!simple!linear!model! 6 1800
for!Strength!as!a!function!of!the! 7 2000
2 600
Number!of!twists.
4 1400
12 4200
15 5400
5 1700
9 3000
1 -©7Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 7
6000
6000
5500
5500
5000
5000
4500
4500
Strength
4000
4000
($)
3500
3500
Expense
3000
3000
2500
2500
2000
2000
1500
1500
1000
1000
500
500
00
0 1 2 3 4 5 6 7 8 99 10 11
11 12
12 13
13 14
14 15
15 16
16
DaysTwists
in Confinement
6000
5500
5000
4500
Y
4000
3500
3000
2500
2000
1500
1000
500
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
X
1 -©9Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 9
6000
5000
e10
4000 e9
3000
e8
2000 e3 e5 e7
e4 e6
1000 e1
e2
0
0 2 4 6 8 10 12 14 16
08 Scatter,1 Correlation
-©10
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 10 107
Residuals versus Fits
u Regression!Model!and!Residual
YActual = b 0 + b1 X + e
1 -©11
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 11
6000
5000 X
4000 X
3000 X
X
2000 X
X
X
1000 X
X
X
0
0 2 4 6 8 10 12 14 16
08 Scatter,1 Correlation
-©12
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 12 108
The Method of Least Squares
u The!true!“best!fit”!model!minimizes!the!total!
deviation!of!the!observations!from!the!model.
åe i
2
= Minimum
u The!equations,!derived!with!calculus,!for!calculating!
the!coefficients!using!the!method!of!least!squares!are:
b1 » slope =
n å x y - (å x )(å y )
i i i i
nå x - (å x ) 2 2
i i
b 0 » y - intercept = y - b1 x
1 -©13
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 13
Regression in Minitab
u Open!a!blank!Minitab!worksheet!and!enter!the!food!
service!data.
u From!the!main!menu!select:
Stat!>!Regression!>!Regression…
08 Scatter,1 Correlation
-©14
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 14 109
Minitab Session Output
S!=!324.3!!!!!!!R-Sq =!96.1%!!!!!R-Sq(adj)!=!95.6%
1 -©15
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 15
08 Scatter,1 Correlation
-©16
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 16 110
Minitab Output – Fitted Line Plot
4000
Revenue
3000
2000
1000
0
0 2 4 6 8 10 12 14 16
Takeoffs
1 -©17
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 17
Minitab Storage
u Store!“Residuals”!and!“Fits”
08 Scatter,1 Correlation
-©18
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 18 111
Exercise – Linear Models in Excel
u In!a!random!sample!of!nine!invoice!processing!
employees,!their!years!of!experience!(x)!and!
the!number!of!transactions!processed!per!hour!
x y
(y)!were!recorded.!!
4.2 2.8
u Use!the!data!to!obtain!a!least!squares!estimate
3.8 2.5
for!the!model!!y!=!b0 +!b1x.!!Check!the!adequacy
4.8 3.1
of!the!model!through!residuals!analysis. 3.4 2.1
4.5 2.9
u Answer: 4.6 2.8
4.3 2.6
y predicted = 0.086 + 0.616 x 3.7 2.4
3.9 2.5
1 -©19
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 19
Correlation
uCorrelation:
– Measures!the!strength!of!the!linear association!between!
two!variables
– Will!not!measure!strong!non-linear!relationships
– DOES!NOT measure!cause!&!effect
– Can!be!used!as!a!first!step!to!regression
– Should!be!used!in!conjunction!with!graphical!techniques
uThe!measure!of!the!strength!of!the!linear!
association!in!a!correlation!analysis!is!“R” –
the!correlation!coefficient.
08 Scatter,1 Correlation
-©20
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 20 112
Properties of R
u -1<R<1
– R!>!0!indicates!a!positive!linear!relationship
– R!<!0!indicates!a!negative!linear!relationship
u R2 gives!the!proportion!of!total!variability!of!the!y!
values!that!can!be!accounted!for!by!the!independent!
variable!x.
u Correlation!coefficients!are!not!additive.
– If!y!=!f(x1)!with!R1 and!y!=!f(x2)!with!R2,
Then!for!y!=!f(x1,!x2),!the!total!variability!¹ R12 +!R22
– x1!&!x2! may!be!highly!correlated!and!contribute!the!same!information!!
for!correlation!with!y.
1 -©21
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 21
08 Scatter,1 Correlation
-©22
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 22 113
Correlation Calculation
u Calculating!R: u In!the!same!notation:
SS xy SS xy
R= b=
SS xx SS yy SS xx
n
SS xx
SS yy = å (y - y ) = å y -
2 (å y ) 2 i
2
R= b
i
n
i
SS yy
( )( )
SS xy = å xi - x yi - y = å xi yi -
(å x )(å y )
i i
1 -©23
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 23
u Minitab!Calculations:
– Stat!>!Regression!>!Regression… – Session!window!output!
reports!R2! (See!Slide!23)
– Stat!>!Regression!>!Fitted!Line!Plot… – R2 is!reported!on!the!
graph!under!the!model!(See!Slide!25)
– Stat!>!Basic!Statistics!>!Correlation… – Session!Window!
Output!(R)
Correlations!(Pearson)
Correlation of C1 and C2 = 0.980, P-Value = 0.000
u Excel!Calculations:
– On!the!Add!Trendline >!Options menu!select!“Display!R-
squared!value!on!chart”!!(See!Slide!21)
08 Scatter,1 Correlation
-©24
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 24 114
Other Models
u Both!Minitab!and!Excel!have!the!capability!to!fit!higher-order!
models,!that!is,!models!with!squared!and/or!cubed!terms.
u The!Minitab!Fitted!Line!plot!has!choices!for!Linear,!Quadratic!
or!Cubic!fits
u Excel!can!also!fit!these!other!models:
– Polynomial!(to!sixth!order):!!
y!=!a +!b1x6 +!b2x5 +!b3x4 +!b4x3 +!b5x2 +!b6x!
– Logarithmic:!! y!=!a +!b*ln(x)
– Exponential:!! y!=!a ebx
– Power: y!=!a xb
u Both!Minitab!&!Excel!calculate!the!coefficients!and!provide!
estimates!of!R2 for!each!model
1 -©25
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 25
The (w)Rap!
u List!your!key!takeaways! u How!can!I!use!this!info!in!
from!this!module. my!project?
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
– _____________________ – ______________________
08 Scatter,1 Correlation
-©27
& Regression
Quality Growth Services Pvt. Ltd. Unpublished proprietary work available only under license. All rights reserved. 27 115
Certified Lean Six Sigma Green Belt
IMPROVE PHASE
Module Objectives
q To introduce the concept of statistically designed
experiments
q To discuss terminology for statistically designed
experiments
q To provide a tool for planning designed experiments
Douglas C. Montgomery
Y=f(x)
8
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Noise Inputs
(Continuous)
q Determine
m Which variables are most influential on the output
m Where to set the inputs to produce the output at the most
desirable level
m Where to set the influential inputs to reduce the variability in
the output
m Where to set the controllable inputs so that the effects of the
uncontrollable inputs are minimized
10
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
q Drawbacks
m If the first guess is wrong, more trials are necessary –
inefficient and time consuming
m If first guess is acceptable, experimentation stops and “best”
solution may never be found
12
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
q Often used when there is little q For the factors beers & walking
prior knowledge or riding:
q Procedure
m Select a combination of
Main Effects Plot (data means) for Score_2
factor levels for a Trans Beers
89
baseline
m Change the levels of one 88
87
out the range of interest
of the factor 86
designed
Mean
88
experiments 87
86
85
84
83
0 4
Beers
14
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
q Factorial experimentation
m Factors are varied together instead of one at a time
m Experiments are a complete set of experimental runs and
are analyzed when all of the runs are complete
Beers
4
Beers
0
w r
0 Carts
w r
Carts
q Other objectives:
m Increase the length of my drive by 30 yards
m Increase the number of times my drive stays in the fairway
m Increase the average number of greens reached in
regulation
23
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Factor Level
Clubs Ping, Callaway
Balls Top Flite, Titleist
Transportation Walking, Riding
Beers 0, 4
25
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
31
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
Module Objectives
q To demonstrate two-level factorial designs
q To review DOE terminology and notation
q To practice DOE design & analysis
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 2 130
Unpublished proprietary work available only under license. All rights reserved.
The Top Five Reasons for Using 2k designs
Y = f(x)
2. They are easy to understand and analyze (Minitab has many
routines for 2k designs).
3. They form the basis of fractional factorial designs, the subject of
the advanced techniques.
4. They can be augmented to form composite designs when more
detailed information is necessary.
5. They require relatively few runs per factor studied.
5
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 6 131
Unpublished proprietary work available only under license. All rights reserved.
Adding to the DOE Method DOE Recipe
7
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 8 132
Unpublished proprietary work available only under license. All rights reserved.
DOE Method Demonstration
q A Black Belt wants to determine the impact of different
levels of advertising on product recognition.
9
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
A Few Basics
q In a 2k experiment:
m One factor level is designated “low” and coded as a –1
m The other level is designated as “high” and coded as a +1
q Standard order: Table of contrasts
TV Radio Press
-1 -1 -1 PRACTICAL
INTERPRETATION
1 -1 -1
-1 1 -1
1 1 -1
-1 -1 1
1 -1 1
-1 1 1
1 1 1
• Exercise
• Create a 24 Factorial Design Matrix
• How many runs are needed?
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 10 133
Unpublished proprietary work available only under license. All rights reserved.
Main Effects
q In a 2k DOE:
m The Main Effect for a factor is the
average of all the data when the factor
is at the high level minus the average
of all the data when the factor is at the
TV Radio Press Recognition
low level
-1 -1 -1 43
m Or: = Response
MainEffect - Response 1 -1 -1 45
High Low
-1 1 -1 45
1 1 -1 49
-1 -1 1 43
q For our experiment the Main Effect 1 -1 1 46
for TV is: -1 1 1 45
1 1 1 49
45 ++49 ++ 46 + 49 43 ++45 ++43 ++45
TV == --
4 4
== 47.25 - 44 = 3.25
11
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
47
Response
46
+3.25
45
44
43 Response TV = Low = 44
42
Low (-1) High (+1)
Level (TV)
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 12 134
Unpublished proprietary work available only under license. All rights reserved.
Calculating Main Effects from Contrasts
q Multiply the response by the contrast for that factor,
sum the values, divide by n (the number of values for
any level)
TV Radio Press Recognition Rec. x TV Rec. x RadioRec. x Press
2x/wk daily weekend 43 -43 -43 -43
3x/wk daily weekend 45 45 -45 -45
2x/wk 3x/wk weekend 45 -45 45 -45
3x/wk 3x/wk weekend 49 49 49 -49
2x/wk daily daily 43 -43 -43 43
3x/wk daily daily 46 46 -46 46
2x/wk 3x/wk daily 45 -45 45 45
3x/wk 3x/wk daily 49 49 49 49
Sum 13 11 1
n 4 4 4
Sum/N 3.25 2.75 0.25
13
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
15
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 16 136
Unpublished proprietary work available only under license. All rights reserved.
Step 4: Design the Experiment (cont.)
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 18 137
Unpublished proprietary work available only under license. All rights reserved.
Step 4: Minitab Output
q Click “OK” on the main DOE Menu
19
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 20 138
Unpublished proprietary work available only under license. All rights reserved.
Minitab Output – Full Model
Estimated Effects and Coefficients
for Recognition (coded units)
AB
Term
AC
ABC
BC
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 24 140
Unpublished proprietary work available only under license. All rights reserved.
Minitab Results – Reduced Model
50 0.00
-0.25
10
-0.50
1
-0.50 -0.25 0.00 0.25 0.50 44 46 48
Residual Fitted Value
4.5 0.25
Frequency
Residual
3.0 0.00
1.5 -0.25
-0.50
0.0
-0.4 -0.2 0.0 0.2 0.4 1 2 3 4 5 6 7 8
Residual Observation Order
TV
48
2x/Week
3x/Week
46
TV
44
Radio
48 daily
3x/week
46
Radio
44
Press
48
weekend
daily
46
Press
44
Mean of Recognition 46
45
44
2x/Week 3x/Week daily 3x/week
Press
47
46
45
44
weekend daily
q What is the value for Recognition if all of the levels were set high?
31
© Quality Growth Services Pvt Ltd. Unpublished proprietary work available only under license. All rights reserved.
10 2K factorial© Quality
DOE Growth Services Pvt Ltd. 33 144
Unpublished proprietary work available only under license. All rights reserved.
Optimizer Output
Optimal TV Radio
Hi 3x/Week 3x/week
D Cur 3x/Week 3x/week
Optimal Response 0.63333 Lo 2x/Week daily
Recognit
Maximum
y = 49.0
d = 0.63333
% to Target
CONTROL PHASE
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 1
Module Objectives
2
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
11 Intro to Control phase & Control Plan 147
Six Sigma Summary
?
3
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
4
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
11 Intro to Control phase & Control Plan 148
Keys to Control
Metrology
6
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
Responsible Metrology
Process
Decision
Process Process
Rules
Control
Documentation
Flow
7
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
11 Intro to Control phase & Control Plan 149
Remember Control Plans?
Process What's Input or Spec. Limits / Meas. Control Sample Who/What Where Decision Rule/
Freq. SOP's
Step Controlled? Output? Requirements Method Method Size Measures? Recorded Corrective Action
Bake Plate Omega T/C PM Maintenance
Input 112 ± .3 °C 100% Weekly ASI Adjust to spec. 9785-123
Temp. Meter Calib. Technician
PMGI 3 sites/ 1 wafer/
Coat/ Output 195 ± 45 nm SP SPC Operator Oracle None 9785-369
Thickness wafer lot
Bake
PMGI 36 sites-
PMGI Output 14 ± xx mm KLA SPC 100% Operator Oracle Rework wafer 9785-411
Width all wafers
36 sites-
Undercut Output 2.0 ± 0.5 mm KLA SPC 100% Operator Oracle 9785-615
all wafers
8
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
9
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
11 Intro to Control phase & Control Plan 150
Developing a Process Control Plan
Process What's Input or Spec. Limits / Meas. Control Sample Who/What Where Decision Rule/
Freq. SOP's
Step Controlled? Output? Requirements Method Method Size Measures? Recorded Corrective Action
10
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
11
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
11 Intro to Control phase & Control Plan 151
Control Plan Buy-in
q The most important part of the control plan:
The approval block.
Approved by:
Approved by:
Approved by:
12
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 1
Module Objectives
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 2
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 3
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 4
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 7
Control Charts
• Transformation of a normal distribution curve in the form of 3 parallel lines, where
• The middle line indicates mean and called central line (C.L.)
• The upper line indicates Mean +3 Sigma and called upper control limit (UCL)
• The lower line indicates Mean -3 Sigma and called lower control limit (LCL)
U.C.L.
+3s
C.L.
Mean
-3s
L.C.L.
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 8
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 11
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
1 5.10 5.27 5.34 5.34 4.10 4.92 5.05 5.37 4.94 4.66 4.16 4.43 4.95 4.73 5.09 4.99 5.90 4.45
2 4.77 5.40 5.17 5.32 4.78 4.69 4.57 4.22 4.88 5.55 4.63 5.69 4.70 4.74 5.20 4.47 5.49 5.95
3 4.32 5.34 5.47 5.81 4.73 5.81 4.58 6.14 5.29 5.58 4.51 4.64 5.11 5.08 4.46 5.09 5.57 5.90
4 4.43 5.71 4.78 5.11 5.12 4.26 4.60 5.20 5.06 5.23 5.34 4.19 4.87 5.01 4.45 5.10 5.55 4.21
5 4.80 5.60 5.36 5.60 5.62 4.95 5.03 5.50 5.23 5.22 4.83 5.16 5.28 4.82 4.70 4.75 5.60 5.73
X-Bar 4.69 5.46 5.22 5.44 4.87 4.92 4.76 5.28 5.08 5.25 4.69 4.82 4.98 4.88 4.78 4.88 5.62 5.25
R 0.78 0.44 0.69 0.70 1.52 1.55 0.47 1.92 0.41 0.92 1.19 1.50 0.58 0.36 0.74 0.63 0.41 1.74
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 12
1
3.0SL=5.583
5.5 5
Sample Mean
5.0 X=5.029
4.5 -3.0SL=4.475
Subgroup 0 5 10 15 20 25
2 3.0SL=2.030
Sample Range
1 R=0.9602
0 -3.0SL=0.00E+00
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 14
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 15
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 16
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 17
1
3.0SL=5.583
5.5 5
Sample Mean
5.0 X=5.029
4.5 -3.0SL=4.475
Subgroup 0 5 10 15 20 25
2 3.0SL=2.030
Sample Range
1 R=0.9602
0 -3.0SL=0.00E+00
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 18
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 19
R = ( R1 + R 2 + ........Rk ) / k
X 1 + X 2 + ... + X k
X= = 50.24
k
UCL R = D 4R n 2 3 4 5 6 7 8 9 10
LCL R = D 3R D4 3.27 2.57 2.28 2.11 2.00 1.92 1.86 1.82 1.78
UCL X = X + A2R D3 * * * * * 0.08 0.14 0.18 0.22
A2 1.88 1.02 0.73 0.58 0.48 0.42 0.37 0.34 0.31
LCL X = X - A2R
d2 1.13 1.69 2.06 2.33 2.53 2.7 2.85 2.97 3.08
s X = R/d2
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 20
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 21
q Identify the cause of the OOC and recalculate the control limits,
excluding the OOC point
q If a point is excluded for the range calculations, exclude it from the
X-bar calculations, also.
?
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 22
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 30
?
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 31
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 32
UCL MR = D 4R n 2 3 4 5 6 7 8 9 10
LCL MR = D 3R D4 3.27 2.57 2.28 2.11 2.00 1.92 1.86 1.82 1.78
UCL X = X + E 2R D3 * * * * * 0.08 0.14 0.18 0.22
E2 2.66 1.77 1.46 1.29 1.18 1.11 1.05 1.01 0.98
LCL X = X - E 2R
d2 1.13 1.69 2.06 2.33 2.53 2.7 2.85 2.97 3.08
R
s X =
d2
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 33
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 34
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 35
q Management summaries
¦ Many management summaries are attribute forms and could benefit from
control chart analysis.
¦ Examples: scrap rates, quality audits, first-run yields, etc.
¦ Calculate p – ^
the proportion non conforming:
^ r
p=
n
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 37
q Subgroup size
¦ The previous formulae assume n is constant. As an approximation, if the
subgroups are ± 25% of the mean then substitute n-bar, the mean subgroup size
for n in the control limit formulae.
¦ To minimize confusion it is best to have constant sample sizes whenever practical.
¦ Minitab calculates p-chart control limits for each subgroup based on that
subgroup’s size.
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 38
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 40
Interpreting p-Charts
q Attribute charts have fewer tests for special causes due to the
lower information content of attribute data.
q A point outside the 3s limit is assumed to be from special cause
variation.
q When np<5, the
likelihood of runs
below p-bar
increases.
Take care when
interpreting runs
in these cases.
Unless the subgroups are equal sizes, only the first test is used.
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 41
P Chart of Passed
1
0.40 1
UCL=0.3711
0.35
Proportion
0.30 _
P=0.2928
0.25
LCL=0.2145
0.20 1
4 8 12 16 20 24 28 32 36
Sample
Tests performed with unequal sample sizes
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 42
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 43
c 1 + c 2 + ... + ck
c=
k
UCLc = c + 3 c
LCLc = c - 3 c
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 44
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 46
u u
UCLu = u + 3 =u+3
n n
¦ Calculate the control limits:
u u
LCLu = u - 3 =u-3
n n
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 47
q Charting defects is one step closer to the Six Sigma ideal, but these
charts still focus on outputs, not inputs.
q c- and u-charts are also charts of attribute data and suffer the same
shortcomings as p- and np-charts.
q It is easier to calculate c-charts than u-charts due to the constant
sample size.
q When the sample sizes are large enough and the number of defects is
large, attribute data behaves like variable data from a normal
distribution.
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 48
Responsibility
Time
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved. 1
Module Objectives
• To discuss project transitions
– To introduce Project Transition Action Plans
• To discuss Realization
2
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
13 Project Report 170
Let’s Talk About Project Closure
• What is a closed project?
– What does it look like?
– Who is in charge?
– Who owns the results?
3
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
temptation to do it
all yourself.
• Complete a Project
Transition Action Plan.
Time
4
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
13 Project Report 171
Project Transition – Keys to Success
• Get buy-in early – no surprises!
– Project team must include all the important players.
– Regular reports to process management
• Establish systems for frequent measurements.
– Process control and product monitoring
• Update the procedures and documentation.
– Document any remaining parts of the hidden factory.
• Take the time to truly educate (not just train or retrain) everyone.
– New procedures and new measurements require education.
• Monitor the results.
– Frequent measures of key project metrics
• Complete a project transition action plan.
5
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
6
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
13 Project Report 172
Project Transition Action Plan Template
Project Title:
Transition Ow ner:
Plan Date:
Signatures:
Transition Owner
Process Owner
Black Belt
7
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
D à M à A à I à C à Realization
• When is Realization?
– After the project is closed!
• What is Realization?
– The twelve months after project closure
– When the profits are counted (realized)
8
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
13 Project Report 173
The Green Belt’s Swan Song
9
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
10
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
13 Project Report 174
Final Report – Major Elements
• Executive Summary
• Project Definition
• Project Management Details
• Summary of changes made to affect improvement
• Outstanding actions required to obtain improvement
• Phase Conclusions
• Phase Deliverables
• Control Methods Applied
• Control Plan
• Key Six Sigma Elements
• Appendices
11
© July 23, Quality Growth Services. Unpublished proprietary work available only under license. All rights reserved.
This proposal is solely meant for M/S Zetwerk Manufacturing Businesses Pvt Ltd. It must not be modified, edited,
distributed, published or claimed for other purposes without the written consent from Quality Growth Services.
Quality Growth Services | H-13, Kirti Nagar, New Delhi -110015 | www.qgspl.com