View
202
Download
4
Category
Tags:
Preview:
DESCRIPTION
Six Sigma
Citation preview
1
Six Sigma Black Belt ProgramDefine & Measure phase
2
Objectives of Black Belt Training
Why 6 Sigma in our organization?
Benefits of 6 Sigma
General Introduction
3
Objectives of Black Belt training
• To develop professionals in quantitative analytical skills, project management, group dynamics, team building and change management to address the organizational challenges.
• To build quality in to the systems to meet and exceed customer requirements.
4
Why 6 Sigma in our organization?
• Overall strategy is to:Accelerate improvements in all processes and services by:
- Identifying customer needs- Crafting a value proposition- Designing business models
• Reduce cost of poor quality by eliminating waste, reducing defects and variations
• To delight the Customers / Clients
• To Grow revenues, Sustain Margins, Improve revenue productivity, Grow human capital
5
When the principles and methodologies of Six Sigma are properly applied in a business process, they return positive top line & bottom-line results.
Some of the 6 Sigma Benefits are
* Improved overall customer satisfaction
* Increased productivity and added value
* Improved capacity and output
* Reduced total defects and cycle time
* Increased product and service reliability
* Improved process flow
* Improved ROI
Benefits of 6 Sigma
6
Introduction to Quality
Fundamental Principles
Quality Concepts
7
Aim So High… You’ll Never Be Bored…
The greatest waste of our natural resources is the number of people who never achieve their potential.
Get out of that slow lane. Shift into that fast lane
If you think you can’t, you won’t. If you think you can there’s a good chance you will.
Even making the effort will make you feel like a new person.
Reputations are made by searching for things that can’t be done and doing them.
Knowledge Skills
AttitudeHabit
8
Moving up the competency ladder
1. Unconsciously Incompetent
2. Consciously Incompetent
3. Consciously Competent
4. Unconsciously Competent
Unconsciously Incompetent
Consciously Incompetent
Consciously Competent
Unconsciously Competent
1
2
3
4
9
Quality
Simply stated, quality comes from meeting customer expectations. This occurs as a result of four activities:
• Understanding customer requirements
• Designing products and services that satisfy those requirements
• Developing processes that are capable of producing those goods and services
• Controlling and managing those processes so they consistently deliver to their capabilities.
Quality Concepts
10
Importance of Quality
Classical Business Model
PROFIT = PRICE - COST
Classical Quality Belief
Better Quality Means:
Modern Technology
Modern Machinery
High Skilled Resources
In short
BETTER QUALITY = HIGH COST
Six Sigma Approach
Better Quality Means:
Less Defects
Less Reworks
Less Buffer
Low Cycle Time
In short
BETTER QUALITY = LOW COST
11
Quality•Better products and
services•Improved processes
Product Quality•Reduced Scrap•Improved customer Response time
Process Quality•Reduced Rework•Elimination of in process•inspection
Customer Satisfaction
Productivity•Decreased•Cycle time
•Elimination of setup time
Market Share
Profit
CostOpportunity for
profit
Price
Quality’s contribution to Profitability
Organization
Internal
External
Compete with value
12
Typical Waste….
Wastes of Manufacturing Process
• Defects
• Waiting
• Processing
• Over production
• Motion
• Inventory
• Transportation
• Under-utilization
• Safety hazards
• Defects
--- rework
• Unsatisfied Customer
---Customer not satisfied, wrong input
• Under-utilization of Resources
--- Poor usage of infrastructure , manpower
• Over- Processing
--- Over-support to customer ,unwanted information
• Redundant Process steps
--- Wrong processes / methods
Wastes of Service Industry
Wastes Kill ….. Business and ProfitsWastes Kill ….. Business and Profits
13
Cost of Poor Quality
Lost Opportunity
Lost sales
Late delivery
Long cycle time
Expediting costs
Excess inventory
Additional Costs of Poor Quality
(intangible)
(tangible)
Lost Opportunity
Scrap
Rework
InspectionWarranty
Rejects
Traditional Quality Costs
(intangible)
(tangible)
25~35% of Sales
(Easily Identified)
(Difficult or impossible to measure)
Lost Customer Loyalty
4~ 6% of Sales
14
Cost of Poor Quality (COPQ)?
Costs incurred due to product or process quality not meeting the customer requirement all the time.
Costs that would not exist if there were no defects.
15
Cost of Quality Categories
Internal Failure Costs - costs that would disappear if no defects existed in the product
prior to shipment to the customer.
External Failure Costs - costs that would disappear if no defects were shipped to the
customer.
Appraisal Costs - costs incurred to discover the condition of the product (during “first
pass through”).
Prevention Costs - costs incurred to keep failure and appraisal costs to a minimum.
16
Internal Failure Costs
Examples
• Scrap (labor and material)
• Rework
• Retest / Recheck/ Re-inspection / Re-testing
• Productivity loss due to defects
• Excess inventories
• Failure analysis
• 100% sorting inspection
Before shipment to customer
17
External Failure Costs
Examples
• Lost business
• Warranty
• Dealing with complaints
• Returned product
• Price concessions due to lower
grade product
After shipment to customer
18
Appraisal Costs
Examples
• Maintaining test equipment
• Quality audits / Transaction Monitoring
• Materials consumed through destructive testing
• Incoming, In process & Final inspection & testing
• Quality system audits
• Inspection & testing of materials & services
19
Prevention Costs
Examples
• Maintaining production/operations equipment
• Process Control & Capability evaluation
• Process improvement: -Error proofing -FMEA -DOE
• Training
20
Evolution of Quality
1920s: Invention of Control Charts by Walter A. Shewhart, Bell Labs
Each Phase Built on the Structure and Gains From the Previous Phases
1940s: Statistical Process Control
1960s: Japanese Quality Movements
1980s: Six Sigma
21
DEMING’S Philosophy
22
Deming’s 14 points for Quality Management (Principles of transformation)
1. Create constancy of purpose for continual improvement of product and service. : Replace short-term reaction with long-term planning.
2. Adopt the new philosophy for economic stability.
3. Cease dependency on inspection to achieve quality. : If variation is reduced, there is no need to inspect manufactured items for defects, because there won't be any.
4. End the practice of awarding business on price tag alone.
5. Improve constantly and forever the system of production and service. : Constantly strive to reduce variation.
6. Institute training on the job. : If people are inadequately trained, they will not all work the same way, and this will introduce variation.
23
7. Adopt and institute modern methods of supervision and leadership.
8. Drive out fear. Deming saw management by fear as counter- productive in the long term,
because it prevents workers from acting in the organization's best interests.
9. Break down barriers between departments and individuals. : The concept of the 'internal customer', that each department serves not the management, but the other departments that use its outputs.
10. Eliminate the use of slogans & posters : Another central TQM idea “it's not people who make most mistakes - it's the process they are working within. Harassing the workforce without improving the processes they use is counter-productive”
Deming’s 14 points for Quality Management
24
11. Eliminate numerical quotas. : Deming saw production targets as encouraging the delivery of poor-quality goods.
12. Remove barriers that rob the hourly worker of the right to pride in workmanship. : Many of the other problems outlined reduce worker satisfaction.
13. Institute a vigorous program of education and retraining.
14. Define top management’s permanent commitment to ever-improving quality and productivity.
Deming’s 14 points for Quality Management
25
Introduction to Six Sigma
26
Introduction to 6 Sigma
Six Sigma History
Success Stories
Definitions and Drivers
6 Sigma : What Makes It Different?
6 Sigma : Meaning
6 Sigma : Benefits
6 Sigma : The Organization
27
In 1979 during a executive meeting, Motorola engineers stated “The real problem at Motorola is that our Quality stinks”
They had data confirming 10 - 20% of annual revenues was spent on correcting poor Quality, costing the company 800 - 900 million US $ PA
To get rid of this problem Motorola came up with Six Sigma Breakthrough strategy.
Six Sigma History
28
And the Results?
1986
4.2
1997
5.6 ~ 16 BillionProducts Manufactured
Motorola
In 1988 Bob Galvin (CEO Motorola) while accepting first “Malcom Baldrige National Quality Award ” for Motorola, described about something called as Six Sigma.
29
Six Sigma History
Six Sigma is a Structured, Project based approach to achieve BREAKTHROUGH results, leading to sustainable and significant FINANCIAL impact on our organization through intensive application of statistical tools and techniques by our people.
Since then, hundreds of companies around the world have adopted Six Sigma as a way of doing business.
30
Success stories - General Electric
General Electric’s, Jack Welch describes Six Sigma as the most important initiative GE has ever undertaken. GE had an operative income of 10% for decades and they were not able to improve this despite various efforts. After implementation of Six Sigma GE was able to improve its operative income from 10% in 1995 to 16.7% in 1998.
As Jack Welch explains it:The best Six Sigma projects begin not inside the business but outside it, focused on answering the question—how can we make the customer more competitive? What is critical to the customer’s success? . . . One thing we have discovered with certainty is that anything we do that makes the customer more successful inevitably results in a financial return for us.
31
Asea Brown Boveri
ABB, after application of Six Sigma has reduced measurement equipment error by 83%. The company also made drastic improvements in material handling resulting in an annual estimated cost savings of US $775000.
32
Allied Signal
Allied Signal, which was on the verge of bankruptcy was made profitable by CEO Larry Bossidy through Six Sigma. The company implemented Six Sigma program in 1994. The cumulative impact on the savings in the direct costs during this period was more than US$ 2 billion.
Allied’s leaders view Six Sigma as “more than just numbers—it’s a statement of our determination to pursue a standard of excellence using every tool at our disposal and never hesitating to reinvent the way we do things.”
33
A Six Sigma Journey • • 1987 to 2006
Over the past few decades six sigma has evolved from a focus on defects to cost reduction to value creation.
34
Companies Implemented 6 Sigma -Worldwide
SonySony
KodakKodak
MicrosoftMicrosoft
DupontDupontBlack and DeckerBlack and Decker
FordFord
Johnson and Johnson
Johnson and Johnson
Asea Brown BoveriAsea Brown Boveri
CitibankCitibank
American ExpressAmerican Express
Johnson ControlsJohnson ControlsCaterpillarCaterpillar
GEGE
ToshibaToshiba
Federal ExpressFederal Express
IBMIBM
Allied SignalAllied Signal
Motorola CorporationMotorola
Corporation
Texas InstrumentsTexas Instruments
35
Definition & Drivers
6 Sigma• A comprehensive and flexible system for achieving, sustaining
and maximizing business success
• Close understanding of customer needs
• Disciplined use of facts, data, and statistical analysis
• Diligent attention to managing, improving, and reinventing business processes
6 Sigma Drivers
Six Sigma - A concept for Quality improvement
The Goal of Six Sigma is not to achieve six sigma levels of quality. It is about improving profitability, though improved quality and efficiency are the immediate by-products of Six Sigma.
36
Quality is a state in which value entitlement is realized for the consumer and the provider in every aspect of business relationship.
• Entitlement for companies means that they have rightful expectation to produce quality products at the highest possible profits.
• Entitlement for customer means they have a rightful level of expectations to high-quality goods at the lowest possible cost.
Definition of Quality as per Six Sigma
37
• Six Sigma metric provides a standard for communicating process status and improvement goals.
• Project selection tied to organizational strategy / balanced scorecard Customer focused / proactive vs. customer driven / reactive.
• Project outcomes / benefits tied to financial reporting system.
• Recognition and reward system established to provide motivation.
• Executives and upper management drive the effort through:– Understanding Six Sigma.– Significant financial commitments.– Actively selecting projects tied to strategy.– Setting up formal review process.– Selecting Champions.– Determining strategic measures.
6 Sigma : What Makes It Different?
38
Six Sigma v/s TQM and ISO
Focus on Money
Leadership & Top-Down Support
Continuous Improvement
Deployment Strategy & Guidelines
Measurement Criteria of Quality Goals
Performance Targets
Application of Statistical Tools
Quality Career Path
Extension to Cost, Cycle Time & Other Business Issues
Integration of Business Goals with Quality
Functional Focus
Project Approach
ISO TQM Six Sigma
39
6 Sigma: Meaning
• Reduce Variation to the half of Tolerance Band
• Minimize defects to the level of 3.4 defects per million opportunities
6σσ Definition
σσ It is a Greek term which designates the spread or distribution about the mean of any process.
66 It is a metric that indicates how well the monitored business process performs. Higher the no., better the process.
Precise But not Accurate: Process is off target
LSL USL
Defect !
Not Accurate Not Precise : High variation in the process
LSL USL
Defect !
LSL USL Six sigma tool tries to reduce variation in the process and shifts process mean towards the target
40
Time taken to process order - Advisor A
3536373839404142434445
1 2 3 4 5 6 7 8 9 10
Order no.
Tim
e in
Min
ute
s
40
Process mean is 40Process mean is 40
USL
LSL
Example 1
41
Time taken to procees order - Advisor B
3536373839404142434445
1 2 3 4 5 6 7 8 9 10
Order no
Tim
e in
Min
ute
s
40
Process mean is 40Process mean is 40
USL
LSL
42
Time taken to process order - Advisor A & B
3536373839404142434445
1 2 3 4 5 6 7 8 9 10
Order no
Tim
e in
Min
ute
s
40
USL
LSL
Which Advisor will you prefer ?
B
A
43
The Goal
FromFixing products
and services to become acceptable
Quantum Leap
6 Sigma GoalTo
Optimized processes that produce defect
free products and services
Influence of 6 Sigma
0
10
20
30
0 10 20 30 40 50
Process before and after Six Sigma
Before After
44
USL
LSL
Target
Six Sigma is about on target performance
with reduced variability around the target
A Six Sigma Landing .
45
Consider 10,00,000 (1 Million) planes take off per year. The plane crash results…at each Sigma Level will be as follows
2 sigma – 3,08,537 Planes per year
3 sigma – 66,807 Planes per
year
4 sigma – 6,210 planes per
year
5 sigma – 233 Planes per
year
6 sigma - 3 Planes per
year
Sigma LevelUnderstanding
46
Sigma LevelDefects Per Million
Opportunities% Defects
1 691,462 69 %
2 308,507 31 %
3 66,807 7 %
4 6,210 0.6 %
5 233 0.02%
6 3.4 0.00034%
Sigma Level and Quality
471 122 3 4 5 6 7 8 9 10 11
1 1 1 1 11
3
Six Sigma Process Predictably twice as good as what the customer wants
LSL USL
6
48
Sigma –
A measure of variation from “target”.
One standard deviation around the mean is about 31% of the total “opportunities” included with in specification limit.
1 Std. Dev. / 1 Sigma
LSL USL
49
If we can fit six standard deviations on both side of the mean in between our target and the specification limits . . .
….. then
99.99966% of our “opportunities” are included!
1 2 3 4 5 6123456
LSL USL
6 Std. Dev. / 6 Sigma
50
• Generates sustained success
• Sets a performance goal for everyone
• Enhances value to customers
• Accelerates the rate of improvement
• Promotes learning
• Executes strategic change
6 Sigma : Benefits
• Cost reduction
• Productivity improvement
• Market-share growth
• Customer retention
• Cycle-time reduction
• Defect reduction
• Culture change
• Product/service development
• And many more.
51
Y = f(X)
Which one should we focus on the Y or X?
Y
■ Dependent Function
■ Output
■ Effect
■ Symptom
■ Monitor
X1,.…,Xn
■ Independent Variable
■ Input
■ Cause
■ Problem
■ Control object
6σ Methodology
6σ Application assures that problem is solved by focusing on the factors that cause the problem.
Focus of the 6σσ Approach of Problem solving
52
If we don’t know the relationship between the Output (Y) and the Inputs (X’s) all we can do is to monitor and sort the good from the bad Y’s.
And Pray Hard that Y will turn out good !
Implement Six Sigma ...
It discovers relationships between Y & Xs
OrOr
Lets Avoid This Situation !
53
Y = f (X)
Everything has a cause. Cause influences the effect
Focus on the cause (x) to change the response
54
Pr = f (V)
Problem is any deviation from the defined standards of a distinguishing feature.
Focus on source of variation to eliminate the problem
55
V = E - O
Variation is the gap; A deviation from the expectation
Measure the variation. We can’t improve what is not measured.
56
act do
docheck
Assurance
6 / ISO / COPC/ e-SCM Performance Management System
Time
PerformanceContinual Improvement
act plan
docheck
Continual Improvement
57
Six Sigma
Organization, Roles and Responsibilities
Six Sigma Organization
Six Sigma focuses on reducing the variation in every process , makes intensive use of the statistical tools, enables decisions based on facts & data rather than gut feelings and puts customer in first place .
Every improvement we are doing in house must be ultimately linked with the customer satisfaction whether it is internal or external . There must be strong and clear linkages between the internal processes, which ensures the end customer satisfaction .
To transform our organization into Six Sigma organization , we must ensure that six sigma is used as a philosophy / strategy and not a tool .
In an essence , It's a new way of managing the enterprise .
59
6 Sigma : The Organization
Leadership
Commitment at Executive Level
Governance
Review,Enable,Monitor,Institutionalise
e.g. Quality leaders, Master Black Belts etc.
Implementation
Scope, Apply Six Sigma Tools And Enhance Business Processes
e.g. Project sponsors, BU Managers, BB, GB, YB – who apply Six Sigma on-the-job
60
Deplo
ym
ent
Cham
pio
n /
Top
Managem
ent
Link Six Sigma to
overall business strategy
Creat Customer Feed back
process
Determine strategic goals and objectives
Allocate appropriate resources
Develop incentive plan & reward system
Establish accountibility in annual performance reviews
Creat a core six sigma
leadership team with defined
responsibilities
Extend the initiative in other group companies
MB
B /
Six
Sig
ma
core
team Design and
implement six sigma deployment
Creat an overall training plan
Design training Material & Periodic updates
Define a project selection process & criteria
Establish project review & project trac mechanism
Define a project validation & project closeout process .
Evaluate cultural obstacles & raise red alerts to management
Design a common database of closed projects & key learnings
Compile the lessons learned & share best practices .
Pro
ject
Cham
pio
n /
BU
Managers
Identify opportunities for breakthrough improvement
Identify appropriate project leaders & arrive at cross functional team
Ensure resource availability for project execution
Review the team progress, remove barriers and resolve issues
Appreciate and recognize good efforrts. Motivate project team
Ensure controls are in place and project gains are sustained forever .
Bla
ck B
elt /
Gre
en
Belt /
Pro
ject
Team DEFINE
the project.Pboblem statement, Objective, Scope, Team, Timeline
MEASURE the response variable Y.Baseline, Target, MSA
ANALYZE the negative effect. Identify root causes(X's) & verify statisticaly
IMPROVE peroformance ( Y's) by implementing the counter solutions for X's
CONTROL the KPIV's to sustain improved Y. Control plan, Control chart etc..
Ispat Industries Ltd. Six Sigma Deployment Process Six Sigma Deployment Model
61
Organization / Infrastructure
Sponsor
Senior executive who sponsors the overall Six Sigma Initiative and who is responsible for implementing Six Sigma within the business. Select meaningful business impact projects. Responsible for achieving Six Sigma project results. Identify, prioritize, select & scope projects. Review, track, and report Six Sigma project progress and results. Eliminate project barriers, assure proper project resources. Reward, recognize Six Sigma project team
Achievements.
Master Black Belt
Mentor black belts on their projects. Ensures effective application of DMAIC.
Coach on appropriate, effective use of Six Sigma Tools & effective project
management to achieve on-time results. Helps sponsor in improvement
opportunity identification. Apply Six Sigma skills and expertise to their own
Projects. Train BBs, Sponsors and Managers in Six Sigma.
Black Belt
Highly experienced person with four weeks of classroom training, has managed several projects and is an expert in Six Sigma methods / tools. Organize, plan and lead Six Sigma projects. Escalates project barriers to sponsors. Project tracking, reporting. Apply Six Sigma skills and expertise toproject execution. Responsible for coaching / mentoring / training Green andYellow Belts, team members and for helping the Sponsor keep the initiativeon track
62
Organization / Infrastructure
Project leader
Understand DMAIC Process and Sustain Improvements. Provide domain /
Process expertise. Accountable for timely completion of projects. Support
improvements. Identifies project team and ensures their availability.
Green BeltProfessional who leads small scope Six Sigma projects. Typically has one week of classroom training in methods, statistical tools, and (sometimes) team skills, participates in Black Belt project team or leads smaller projects.
Yellow Belt
Typically has two days of classroom training in methods and basic statistical tools, participates on a Green Belt project team or leads smaller
improvement projects.
Team Member
Professional who has general awareness of Six Sigma. (through no formal training) and who brings relevant experience or expertise to a particular project.
Apply Six Sigma tools with help of Black / Green / Yellow Belts Contribute ideas during meetings and carry out action items. Collect and analyze data
Lead small activities such as process capability studies, measurement system studies, verifying causes and solutions. Implement and sustain solutions.
63
Green Belt • Participates in Black Belt project team.• Leads smaller projects
Yellow Belt• Leads smaller improvement projects
Master Black Belt- Mentor black belts on their
projects
- Ensures effective application
of DMAIC- Helps sponsor in improvement opportunity identification
- Apply Six Sigma skills & expertise to their own projects
- Train BBs, Sponsors & Managers in Six Sigma
Sponsor
- Sponsors the overall Six Sigma initiative
- Select meaningful business impact projects
- Eliminate project barriers, assure proper project resources.
- Reward, recognize Sigma project team achievements
Six sigma Roles
Black Belt• Organize, plan & lead Six Sigma projects• Project barrier escalation to sponsors• Project tracking & reporting coaching / mentoring / training Green & Yellow Belts, team members
Team Members• Bring expertise to projects• Contribute ideas during meetings and carry out action items• Collect and analyze data• Lead small activities • Implement & sustain solutions
Project Leader- Understand DMAIC Process
and Sustain improvements
- Accountable for timely
completion of projects
- Provide domain / Process expertise
- Support improvements. Identifies project team and ensures their availability
64
Champion or Sponsor
• Representative of Top Management
• Has Authority over the scope of project
• Statistical Knowledge is preferable
• Experience in carrying out Quality improvement Projects is preferable
At Start: * Identify the Goals of the Project* Select the Project Leader or Black Belt* Ensure the Project Scope is under his control* Identify Milestones* Prepare Project Schedule
On Going: • Provide Resources• Conduct Periodic Review & add value• Ensure the Project is on Right Track• Control Budget
At End: • Handle Implementation Issues• Quantify the Project Results • Verify whether the Goals are achieved• Prepare Future Action Plan, if required• Conclude Project
Requirements Responsibilities
65
Master Black Belt
• Sound knowledge of Statistical Tools are required
• Good Communication & Teaching Skills are essential
• Should be a Good Consultant
Requirements Responsibilities
At Start:• Create Six Sigma awareness among top
management• Provide guidance & training for gathering Voice of
Customer’s & Stakeholders • Help to identify projects• Provide guidance to prepare Project Charter
• On Going:• Provide training on various Statistical Tools useful at
different phases of the project life cycle• Give Statistical Consultancy at different phases of
the project life cycle
66
Black Belt
• Statistical Knowledge essential
• Experience in Data Collection, Analysis & Interpretation required
• Should posses Leadership Qualities
• Should posses Good Communication Skills
Requirements Responsibilities
At Start:• Select Team Members• Prepare Project Charter• Identify Ys
On Going:• Analyze Ys & compute Baseline Sigma Value• Identify Xs• Establish relationships between Ys & Xs• Optimize Xs• Devise Control Mechanism to ensure that Xs
are at optimum always.
At End:• Be equivalent to Master Black Belt.• Conducts programs on Statistical tools• Facilitates Business Unit Head for project
selection
67
Green Belts \ Team Members
On Going:Collect data on YsHelp BB to analyze Ys data.Be part of identifying XsCollect data on XsHelp BB to establish relationship between Xs &YsHelp BB to optimize & control XsHelp design the new process Drive the project to completion
At End:* Green Belt will be equivalent to Black Belt
Responsibilities
68
Team Members
• Guidelines for team members– Manageable team size up to 5 people
– People who are part of the process
– People who are benefited by removal of pain area
– People who have domain knowledge
– People from same location
– Guest members as required
69
Project Teams
Champion
Black Belt
Green Belt
Team Members
B
G
G G
70
Team Dynamics
Mutual Accountability
Small group of people
Individual Accountability
Specific Goals
Common Approach
Meaningful Purpose
Problem Solving
Technical/Functional
Interpersonal
AccountabilitySkills
Commitment
Per
form
ance
R
esu
lts
Collective
Work ProductKnowledge
Enhancement
71
Deployment Strategy
Elements of 6 Sigma Deployment
Role Matrix
Training
Certification Criteria
Project Classification
Project Benefits Evaluation
72
TopManagementCommitment
1.Project Identification -
To contribute to the bottom-line of the organization, client
satisfaction, etc
2. Project Classification
•Black Belt Project•Green Belt Project•Yellow Belt Projects
3.Enhanced Training and certification
•Awareness Program•Yellow Belt Training•Green Belt Training•Black Belt Training
4. Review Mechanisms
•Program Reviews•Project Reviews
5. Project Benefits Evaluation
Project Benefits (QNI) = Project Savings – Project Expenditure
6. Fostering Quality Culture
•Branding• Yearly Awards •dedicated facilities and staff
Six Sigma is being implemented in the organization as a catalyst for change in culture and achieve competitive advantage.
DMAIC methodology is being used to develop and fine tune both core and enabling processes.
Six Sigma Deployment plan is properly documented as PACE guidelines.
Elements of 6 Sigma Deployment
“commitment for Six Sigma” and “Champions/ Sponsors” are the most powerful success factors
“commitment for Six Sigma” and “Champions/ Sponsors” are the most powerful success factors
73
Team Structure - Role Matrix
Role DescriptionQuality Team
Operations / Support Team
HR TeamSS Executive
Committee (EC)
Training Curriculum preparation and implementing improvements as per recommendations
√
Preparing the Training Calendar √
Identifying the Trainers √
Certifying the Trainers √
Defining the Selection Criteria and improving the same on an ongoing basis
√
Identifying the Trainees √ √
Selection of Projects based on ongoing experience and Business Goals
√ √
Sponsor for the Project √ √
Project Management / Coordination √ √
Project Execution Respective BB / GB / YB
Project Certification √
Review of training material √ √
74
Types of Training Awareness Program – The program will be of a 4 Hrs duration.
This is engagement specific and will be conducted internally by the dedicated certified trainer.
Yellow Belt Training – Two continuous days as stipulated in the program calendar
Green Belt Training – Five Days program spread over 3 months (2days + 2days + 1day) as stipulated in the program calendar
Black Belt Training – 20 Days program spread over 4 months, 5 continuous days every month as stipulated in the program calendar.
For YB/GB/BB trainings all trainees are expected to be fully available during the training period
75
Certification Criteria
All trainees would need to attend a 2 day YB session/ 5 day GB session / 20 days BB session & pass the respective certification examination.
Qualifying mark would be 70% for YB, 75% for GB & 80% for BB
Project Certification will be granted to the BB, GB and YB identified and dedicated for the particular project
Certification will be done internally by the Certification Board. The criteria for certification will be
Complete training
Pass the examination
Meet any of the options as mentioned in the next page.
76
Certification Criteria
Black Belt Certification
Green Belt Certification
Yellow Belt Certification
OPTIONS
1 3 4 52
Black Belt Project Leader
Green Belt Project Leader
Yellow Belt Project Leader
Yellow Belt/ Green Belt/ Black Belt Project Member
77
Project Classification
Black Belt Projects• Projects that have cross functional scope and are of high impact and
complexity with respect to NET revenue earnings / NET savings. (Expected savings $ 50,000/ annum min).
• Should complete within 4 to 6 months• Base line, target and savings should be validated by Financial
representative
Black Belt Projects• Projects that have cross functional scope and are of high impact and
complexity with respect to NET revenue earnings / NET savings. (Expected savings $ 50,000/ annum min).
• Should complete within 4 to 6 months• Base line, target and savings should be validated by Financial
representative
Yellow Belt Projects• YB projects are Cell wise projects, where cross functional team is not
required and very limited statistical knowledge required to carry out the project
• This projects are taken as value addition to GB projects. YB projects by themselves do not generate / save revenue.
Yellow Belt Projects• YB projects are Cell wise projects, where cross functional team is not
required and very limited statistical knowledge required to carry out the project
• This projects are taken as value addition to GB projects. YB projects by themselves do not generate / save revenue.
Green Belt Projects• GB projects are less complex than BB projects. (Expected savings $10,000-$
50,000/ annum) • Should complete within 2 to 4 months• Base line, target and savings should be validated by Financial
representative
Green Belt Projects• GB projects are less complex than BB projects. (Expected savings $10,000-$
50,000/ annum) • Should complete within 2 to 4 months• Base line, target and savings should be validated by Financial
representative
IMP
AC
T Green Belt Black Belt
Yellow Belt X
Complexity
78
Project Benefits Evaluation
Soft (Intangible)
Benefits
Hard (Tangible) Benefits
Definition: Any measurable Improvement
which cannot be quantified and converted
into Dollar Savings.
Examples:• Increase Customer Satisfaction
Scores• Improve VOC Scores• Improving Federal / State regulatory
Compliance Scores
Definition: Any measurable Improvement which can be Quantified and converted into Dollar Savings.
Examples: • Reduction of 3 FTE’s • 1% market share Increase• Eliminate 2 Temporary Positions• Rework down 20%• Lower vendor cost per transaction• Save 1 temp. worker 2 hrs of work
per week• Eliminate 30% call volume• Decrease AHT of the process by 60
secs• Reduction in past due receivables
by $10MMProject Benefits (QNI) = Project Savings – Project Expenditure1. Project Charter will include Sign-off from Finance Analyst2. Project validation / authentication is done by Sponsor, Head Quality and Finance Dept. The
return on the project is verified.3. Only upon successful completion of the project, the respective BB will be eligible for
certification.
Project Benefits (QNI) = Project Savings – Project Expenditure1. Project Charter will include Sign-off from Finance Analyst2. Project validation / authentication is done by Sponsor, Head Quality and Finance Dept. The
return on the project is verified.3. Only upon successful completion of the project, the respective BB will be eligible for
certification.
79
Six Sigma
Project Identification and Selection
80
Project Identification & Selection
Project Selection approach
Project Sources and Selection Criteria
Kano Model
CTQ Tree
Quality Function Deployment
81
Project Selection Approach
Steps
What is the process ?Who are the potential customers ( Internal, External, Business
owners)What are the process deliverables ?What is the Unit of Measurement ?Is Measurement System in place ?What is the Current Performance (Baseline) ?What is a expected / targeted performance ?Does gap exists ?Does gap carries significant impact to customer / business ?Identify the improvement approach Continuous Improvement OR Re
designed ?
82
Step I Step III Step IV Step V Step VI Step VII Step II
One of the Business / Functional unit (e.g. Operations / Purchase
Study the critical process and the pain areas identified during customer survey
Identify External & Internal (Business Owner & other functions) for the critical process
Categorize the customer expectations (VOC / Pain areas) from critical process obtained form from surveys in to
Deliverables of Quoting process
For each deliverable of critical process identify If the sub process or product / service feature exists
Business UnitBusiness Unit
Business ProcessBusiness Process
Identify Critical ProcessIdentify Critical Process
Study ProcessStudy Process
Identify CustomersIdentify Customers
List out Customer Expectations/Deliverables
List out Customer Expectations/Deliverables
Does ProcessExists?
AYes
No B
Approach Steps Tools Approach Description
Business Process Framework
Process Flow Chart
Customer Surveys
Brain storming
Prioritization
High level process mapping
SIPOC
SIPOC
Sub Process Flow Chart
Process flow steps
SS Project Selection Approach – Identifying Critical processes
83
Ensure that the critical process metrics are linked with Business objectives, Strategy & VOC
Collect data for all critical process metrics to define baseline process performance
Conduct a gap analysis between the criticalprocess baseline performance and targeted /
customer expected performance level
If the sub process exists for a metric / feature apply DMAIC for long term approach OR Problem solving approach for Quick wins for early
realization of improvements where solutions are known
For each deliverable of critical process identify unit of measure & measurement system
Does Gap Exists?
If the sub process does not exist for a metric / feature apply DMADV or
Process Re Engineering Approach.
Are Metrics
in place?
Define Metrics & implement
data collection plan
Define Metrics & implement
data collection plan
A
Validate the MetricsValidate the Metrics
Process Base liningProcess Base lining
Does Process exists?
No
NoExplore need of
performance excellence
Explore need of performance
excellence
Apply DMADV / Process
Re Engineering
Apply DMADV / Process
Re Engineering
Long Term Approach – Apply DMAIC
Long Term Approach – Apply DMAIC
Quick wins – Apply REIS
Quick wins – Apply REIS
Yes
Yes B
Approach Steps Tools Approach Description
SIPOC
Data collection checklists
Dashboard
Process Capability
Control Charts
Enterprise Performance
Management Matrix
Process
Performance
Enhancement
implementation
1. DMAIC
2. DMADV
3. Re Design / Re Engineer
4. Problem Solving Approach
Six Sigma Project Selection Approach (Contd.)
84
SS Project Selection Approach – Identifying Undesirable Conditions
1. Identify the undesirable business process conditions. (Customer complaints, VOC, Quality reports, Monthly performance indicators)
2. Quantify the undesirable conditions in to Defect (%, PPM, DPMO, Sigma level) or Cost .
3. Identify the potential causes for the undesirable conditions.(High level process map, C & E analysis, Pareto analysis)
4. Arrive at the vital few potential causes and prioritize .(NVA's, Multivoting, 80:20 principle )
5. Identify the business process to which each prioritized suspected cause belongs to .
6. Select the core or enabling process for improvement, quantifying the gap between current performance and what customer wants (Process Deliverable).
Apply...
Apply...
Six Sigma Project 1
Define
Measure
Analyze
Improve
Control
Six Sigma Project 2
Define
Measure
Analyze
Improve
Control
Six Sigma Project N
Define
Measure
Analyze
Improve
ControlTarget : Reliable (Stable & controlled) processes, Satisfied Customers, Happy Employees
85
Improvement Methodology Selection
Product / Process Development : Where a sub process
does not exist to take care of the customer requirement
Product / Process Development : Where a sub process
does not exist to take care of the customer requirement
Product / Process Performance Enhancement : Where the sub
process exists but unable to meet the customer requirement with
current process performance level
Product / Process Performance Enhancement : Where the sub
process exists but unable to meet the customer requirement with
current process performance level
Process ImprovementProcess Improvement
DMADV(Define, measure, analyze,
design and validate)
DMADV(Define, measure, analyze,
design and validate)
DMAIC(Define, measure, analyze,
improve and control)
DMAIC(Define, measure, analyze,
improve and control)
Design processes that do not exist for e.g. knowledge
management
Design processes that do not exist for e.g. knowledge
management
Enhance process metrics e.g. turnaround time, customer problem
resolution effectiveness etc
Enhance process metrics e.g. turnaround time, customer problem
resolution effectiveness etc
Product / Process Metric monitoring :
Where the tentative solutions are known and systematic execution is required to ensure the Improved process performance &
sustenance
Product / Process Metric monitoring :
Where the tentative solutions are known and systematic execution is required to ensure the Improved process performance &
sustenance
REIS(Recognize, evaluate, implement
and sustain)
REIS(Recognize, evaluate, implement
and sustain)
Monitor and reduce errors in the process e.g. ineffective call close
etc
Monitor and reduce errors in the process e.g. ineffective call close
etc
86
Six Sigma DMAIC Project Flow Define
Tollgate
DefineDefine
Step 1: Identify customer & their care abouts.
Covert their needs in to Critical to Satisfaction
(CTS) i.e. CTQ – Critical to quality, CTC – Critical to
Cost & CTD Critical to Delivery.
Step 2: Develop Project Charter
MeasureTollgate
MeasureMeasure
Step 3: Take the snapshot of the process,
how the process performing currently & fix
the baseline.
Step 4 : Validate the measurement system from which we collect the
data.
AnalyzeTollgate
AnalyzeAnalyze
Step 5: Identify the key process input variables that affects the outputs
most.
Step 6 : Verify the identified causes to see whether those are real or
not.
ImproveImprove
ImproveTollgate
Step 7: Determine the solutions to optimize the
output & eliminate / reduce defects &
variations.
Step 8 : Implement the solutions partly & statistically verify their impacts on output.
ControlControl
ControlTollgate
Step9 : Put the control in place to sustain the gains
made by the process improvement.
Step 10 : Integrate in daily work by Process Owner &
team
Phase Deliverables
Required• List of Project CTQs• QFD/CTQ Tree• Project Charter• SIPOC
Tools Box• Project Risk Assessment• Stakeholder Analysis• High Level Project Plan• In Scope/Out of Scope• Customer Survey
Methods (focus groups, interviews, etc.)
Required• Process Baseline
capability• Operational definition,
Specification limits, target, defect definition for Project Y(s)
• Measurement System Analysis
Tools Box• Benchmarking• Data Collection Plan• Gage R&R• Process Map• FMEA• Pareto Analysis
• Required• Data Normality Test • List of Statistically
Significant Xs .• List of vital few Xs &
their verification
• Tools Box • RCA / Fishbone
Diagram• Hypothesis Testing• Correlation &
Regression Analysis
Required• Optimization of Xs• Improvement
verification• Tolerances on Vital Few
Xs
Tools Box
• Design of Experiments• Improved Process Maps• FMEA on new process
Required• Post Improvement
Capability• Improvement tracking • Process Control Plan• Process Owner Signoff
Tools Box• Control Charts• Control Plan• Hypothesis Testing• Error Proofing
D M A I C
87
DefineTollgate
DefineDefine
Step 1: Identify new product, Process, Service
Step 2: Develop Project Charter
MeasureTollgate
MeasureMeasure
Step 3: Identify caustomer & their care abouts (CTQ)
Step 4 : Identify , deploy data collection plan & process base lining
AnalyzeTollgate
AnalyzeAnalyze
Step 5: Develop design alternatives
Step 6 : Develop & evaluate high level design capability
DesignDesign
DesignTollgate
Step 7: Optimize the micro level design parameters
Step 8 : Evaluate and verify the micro level design.
ValidateValidate
ValidateTollgate
Step9 : Validate results on full scale & Put the controls in place to sustain the gain.
Step 10 : Integrate in daily work by Process Owner & team
Phase Deliverables
Required• Customer surveys,
Interviews• Project Charter
Tools Box
• High Level Project Plan• In Scope/Out of Scope• Customer Survey
Methods (focus groups, interviews, etc.)
Required• Operational definition, • Specification limits, • Current Performance• Performance target, • defect definition
Tools Box• Benchmarking• Data Collection Plan• CTQ Tree• Affinity Diagram• Process capability• MSA
Required• Design alternatives • Selection of best
alternative• Detail design
requirements• High level design
capability predicated
Tools Box
• Benchmarking• Design scorecard• FMEA• Layout diagrams• Process map / model• Prototyping
Required• Design alternatives • Selection of best
alternative• Detail design
requirements• High level design
capability predicated
Tools Box
• Benchmarking• Design scorecard• FMEA• Layout diagrams• Process map / model• Prototyping
Required• Post Design Capability• Scale up decisions• Full scale process
implementation• Improvement tracking • Process Owner Signoff
Tools Box• Control Charts• Control Plan• Process capability• Design scorecard• Process management
chart• Standards &
Procedures
Required• Post Design Capability• Scale up decisions• Full scale process
implementation• Improvement tracking • Process Owner Signoff
Tools Box• Control Charts• Control Plan• Process capability• Design scorecard• Process management
chart• Standards &
Procedures
D M A D V
Required• Optimization • Of design parameters• Prediction model• Design Verification• Updated design
scorecard
Tools Box
• Design of Experiments• Improved Process Maps• Updated FMEA • Software simulation• Pilot / Test plan• Design score card• Tolerance analysis• Process management
chart
Six Sigma DMADV Project Flow
88
Step 1: Identify Improvement area/ Problem
Step 2: Define Problem statement & Goal statement
RecognizeRecognizeRecognizeRecognize
Step 3: Evaluate the causes for problem
Step 4 : List the vital causes
EvaluateEvaluateEvaluateEvaluate
Step 5: Identify the countermeasures / solutions
Step 6: Implement the countermeasures / solutions
ImplementImplementImplementImplement
Step 7: Validate results on full scale & Put the controls in place to sustain the gain.
Step 8 : Integrate in daily work by Process Owner & team
SustainSustainSustainSustain
Phase Deliverables
Required• Customer surveys,
Interviews• Problem & Goal
statement• Business dashboard
Tools Box
• In Scope/Out of Scope• Customer Survey
Methods (focus groups, interviews, etc.)
• Problem solving charter
Required• Vital causes
Tools Box
• Cause & Effect Analysis
Required• List of
countermeasures/ solutions
• Countermeasure / solution implementation plan
Tools Box
• Countermeasure Matrix• Implementation plan
RequiredImprovement tracking Process Control PlanProcess Owner Signoff
Tools Box
• Control Charts• Control Plan• Visual Management
Phase ExitReview
Phase ExitReview
Phase ExitReview
Phase ExitReview
Phase ExitReview
Phase ExitReview
Phase ExitReview
Phase ExitReview
R E I S
Problem Solving (Quick Win) Project Flow
89
Sources of Project Generation
C-Sat / V-Sat
Short Term / Long Term Organizational Goals (Business and operations)
Service Level
Transaction Quality
Resource Utilization
Engagement Targets client complaints
etc
Six Sigma Project Selection criteria
• Business Filters• contribute to bottom-line
of the organization• Drastic, long term, risk free
Improvement• Solution Not known• No other teams working on
project
• Project Filters• Availability of Data• Measurable• Time bound• Realistic and Attainable
Project Sources and Selection Criteria
90
• Project identification begin not inside the business but outside it• Focus on answering the following questions:
* How can we make the customer more competitive?* What is critical to the customer’s success?
Anything we do that makes the customer more successful inevitablyresults in a financial return for us.
• Kano Model is one of best technique to collect and understand Voice of Customer (VOC)
• Business and operational goals can be analyzed by CTQ Drill down or Quality Function Deployment
Customers Perception
91
Who is a customer?One who is paying for your service or product a person who buys goods or services
Types of customers •External or internal•Lost customers•Prospective customers
Customers
92
What do customer wants?
What do shareholder wants?
Utility:They buy product or service for a needTime : They want it when they need itValue : They pay for it only if they perceive a value
The business must successfully servethe customers “ Wants “ and
Still provide good profits.
What do customer & shareholders want?
customer voice
I need right product,
at right time,
at good price.
Shareholder
voice
I need happy customers
and good profit
93
Performance
Customer Satisfaction
CUSTOMER’S WANTS
•Must Be
•Satisfiers
•Delighters
Must BeThe better I do, the less dissatisfied the customer is.(e.g., airlines get no credit for getting bags to you on-time)
DelightersNo penalty for not doing them However, if you do them, you get bonus points
SatisfiersThe better we do, the happier the customer is (Plane gets to the destination on time)
Kano Model
94
Kano Model : Exercise
Exercise 1:
Classify the following as Must Be, Satisfiers & Delighters
1. Air Conditioner free with every purchase of four wheeler
2. Availability of hot water in 5 star hotel bath room
3. More mileage per liter of fuel
Exercise 2:
1. Identify a Must Be, Satisfier & Delighter from your own process
95
Gather voice of the customer
* Review existing voc data* Decide what to collect* Select tool to collect*Collect data
Voice of Customer : Ways to Capture
• SURVEYS• FOCUS GROUPS• INTERVIEWS• WORD OF MOUTH• COMPLAINTS
96
Surveys :A method of gathering information From a sample representing the population these are comprehensive data driven information vehicles that are useful in capturing customer requirements as well as
measuring performance against those requirements.
Focus groups: in this group you group together similar customers and ask for their opinion on the requirements as well as performance against those requirements
Customer interviews: could be informal or Structured. Informal interviews give good insight into the customer perspective of the product and services and depends on probing open ended questions
Voice of Customer : Ways to Capture
97
Word of mouth: the customer feedback comes through direct and different channels. Internal: employee feedback External: reports from known sources
Be a customer yourself : feel the quality of Service yourself
Voice of Customer : Ways to Capture
98
Voice Of The Customer
What do we do with VOC input?What do we do with VOC input?
Sample Comments/Data
“I’m Tired Of Having To Call up for this Lousy Product Every Ten Days”
“I simply don’t understand what the Customer Support Professional Talks about ”
“Why Don’t You Guys Get Your Act Together?!”
“The Phone Must Have Rung Ten Times Before I Got An Answer”
“I’m Not Very Happy With Your Service”
99
Translating Customer Needs To Requirements
Voice Of The Customer Key Issue(s) Requirement
I Am Always On Hold Or Transferred To The Wrong Person.
Add Additional Menu Items To Voice System (BAD)
Customer Gets To The Correct Person The First Time (GOOD)
Want To Talk To The Right Person Quickly
I’m Getting My BillAt Different Times Of The Month.
Customer Wants Timely Bill (BAD)
Customer Bill Received Same Day Of Month (GOOD)
Consistent Monthly Bill
Take Too Long To Process The Application.
Customer Wants Fast Loan (BAD)
Customer Receives Approval On Customer Request Date (GOOD)
Speed Up Loan
100
1. Refer to your workbook.
2. You have given a “Voice Of Customer (VOC)”
3. Using these VOC
- Identify key issues customer is facing - Identify the specific requirements which will tackle these issues
Translating Customer Needs (VOCs) To Requirements.
Exercise 1.1
101
Collection of Voice of Customer (VOC)
Step 1: Get the Voice of Customer (VOC) and the importance rating through survey, feedback, market research, etc.
Example: VOC of a BPO company
VOC Rating
Utilization of billable resources should be maximum
Very Important
Abandoned calls should be minimum Important
Right & Complete Resolution Very Important
Talk Time should be reasonable Important
Good Customer Service Skills Important
102
Step 2: Quantify the Customer rating numerically.
VOC Rating
Utilization of billable resources should be maximum 5
Abandoned calls should be minimum 3
Right & Complete Resolution 5
Talk Time should be reasonable 3
Good Customer Service Skills 3
Very Important: 5
Important: 3
Reasonable: 1
Step 3: Get Voice of Stakeholder.
Voice of Stakeholder Rating
Buffer should be minimum Very Important
Reduce loss of login hours Important
Reduce Rework Important
Increase CSat Score Very Important
Reduce Customer Complaints Very Important
103
Step 4: Quantify Stake holder's voice.
Voice of Stakeholder4 Rating
Buffer should be minimum 5
Reduce loss of login hours 3
Reduce Rework 3
Increase CSat Score 5
Reduce Customer Complaints 5
Very Important: 5
Important: 3
Reasonable: 1
104
Step 5: Map the Customer Requirements & Stake Holder Requirements
to Business and Operational Goals.
Requirements Rating QualityCycle Time
TrainingMulti Tasking
Utilization of billable resources should be maximum
5 High High Medium
Abandoned calls should be minimum 3 Medium High Medium
Right & Complete Resolution 5 High Medium
Talk Time should be reasonable 3 High Medium
Good Customer Service Skills 3 High
Buffer should be minimum 5 Medium High
Reduce loss of login hours 3 Medium high
Reduce Rework 3 High Medium
Increase CSat Score 5 High Medium
Reduce Customer Complaints 5 High Medium
105
Step 6: Calculate importance ranking to Business and Operational Goals.
Requirements Rating QualityCycle Time
TrainingMulti Tasking
Utilization of billable resources should be maximum
5 High High Medium
Abandoned calls should be minimum 3 Medium High Medium
Right & Complete Resolution 5 High Medium
Talk Time should be reasonable 3 High Medium
Good Customer Service Skills 3 High
Buffer should be minimum 5 Medium High
Reduce loss of login hours 3 Medium High
Reduce Rework 3 High Medium
Increase CSat Score 5 High Medium
Reduce Customer Complaints 5 High Medium
Rank 124 118 42 55
106
Step 7: Identify Key Process Output Variables or CTQ’s (Ys).
KPOV’s (Ys) Ranking
Quality 124
Cycle Time 118
Identify Six sigma projects to improve these KPOV’s
107
1. Refer to your workbook.
2. Identify Six Sigma projects for the areas identified through VOC
3. Allowable time : ----- Minutes
Six Sigma project identification through VOC
Exercise 1.2 (10 minutes)
108
CTQ (Critical to Quality) & CTQ Tree
Business CTQBusiness CTQ Customer CTQCustomer CTQ
Internal CTQInternal CTQ
Project CTQProject CTQ
The basic reason any process exists for is to satisfy the
requirements of the customer / Stakeholders.
The critical customer satisfaction parameters can be broadly
categorized under Cost, Quality, Delivery, Service, productivity etc which are called as CTQs
CTQ is a CTQ is a Product , Process or Service Product , Process or Service
characteristic characteristic that satisfies a that satisfies a
Customer Requirement Customer Requirement (External & Internal – Business (External & Internal – Business
Owners, all functions )Owners, all functions )
BusinessY
BusinessY
Process Y
Process Y
Project Y
Project Y
109
Business Objective
Voice of customer /
Organizational Goals/ Pain Areas
Identification of Improvement Areas
Customer
Satisfaction
Score
Service Level
Agreement
Transactional
Quality
Development
Cost
Resource
Utilization
Six Sigma Black Belt & Green Belt Projects
Leading to significant Top line improvement with Customer Satisfaction &
Bottom line impact through Revenue generation and cost saving
Business YBusiness Y
Process YProcess Y
Project YProject Y
110
CTQ TREE
A methodology to break the CTQ’s in the over all level to the CTQ’s at sub process level
Example 1:
CTQ Tree for the KPOV: Quality
Quality
Voice Quality
Data Quality
Voice & Accent
Empathy
Culture
Example 2 :
CTQ Tree for the KPOV: Cycle Time
Cycle Time
Handling Time
Waiting Time
111
Examples of Customer CTQs
A Car Purchaser
• Mileage
• Spacious
• Low price/affordable
• High technology
• Loan Facility
A Car Purchaser
• Mileage
• Spacious
• Low price/affordable
• High technology
• Loan Facility
A Prospective Employee
• Good Salary• Location Preference
• Flexible Working Hours
• ESOPs
• Good working place
A Prospective Employee
• Good Salary• Location Preference
• Flexible Working Hours
• ESOPs
• Good working place
• Customer Sat (Ext)
• Customer Sat (Int)
• Service CSat (Ext)
• Service CSat (Int)
• Customer Sat (Ext)
• Customer Sat (Int)
• Service CSat (Ext)
• Service CSat (Int)
Order Management Client
Service Quality
• Yield
• Productivity
• 1st Touch Cycle Time
• Case Res Cycle Time
Service Quality
• Yield
• Productivity
• 1st Touch Cycle Time
• Case Res Cycle Time
112
•Define what is critical (CTQ) to Business Champion, BB•Define your Customer Champion, BB•Explore Customer CTQ Champion, BB•Define Internal CTQ / Critical Business Process Champion, BB
CTQ Drill Down responsibility Matrix
113
1. Refer to your workbook.
2. Draw a CTQ tree for your any one of the project you have selected
3. Allowable time : ----- Minutes
Drawing a CTQ Tree
Exercise 1.3 (15 minutes)
114
What is QFD?
• QFD begins with Customer. It is also called House of Quality
• QFD links the needs of the customer with design, development, engineering, manufacturing and service functions. It helps organizations seek out both spoken and unspoken needs, translate these into actions and designs, and focus various business functions toward achieving this common goal. QFD empowers organizations to exceed normal expectations and provide a level of unanticipated excitement that generates value.
• The basic idea of QFD is to translate the voice of customer, throughout the marketing, R&D, engineering and manufacturing stages of product development.
Quality Function Deployment (QFD)
115
• It is a structured approach that facilitates the translation of the customers voice into specific requirements.
• These specific requirements are mapped to the design process production process and delivery processes to determine the process and design requirements.
• Quality function deployment allows customers to prioritize their requirements
• Benchmark with competition, help optimize and to attain competitive advantage
Quality Function Deployment (QFD)
116
QFD: When to use ?
• To develop new product or service capability, specially for complex ones.
• For products and services where clarification and prioritization of efforts on key customer wants is needed.
• For developing or refining existing internal systems in order to build or product delivery capability.
• For development of products or services that do not have a clear mapping of the customer requirements and the design attributes.
Quality Function Deployment (QFD)
117
Benefits of QFD
• QFD brings in involvement of all the departments thus improving the communication among them.
• Provides excellent frame work for cross functional deployment of quality cost & delivery.
• Since QFD is a documentation process it helps in reducing mistakes.
• Brings robustness in the product.
Quality Function Deployment (QFD)
118
How to build one?1. The first portion of QFD matrix is the Customer Requirements. All the VOC that have been
translated into tangible requirements form the Customer Requirements part.
2. The second part, on the right hand side of QFD, is the Planning Matrix. It is used to quantify Customers’ requirement priorities and their perception of the performances of existing products. This is done through Importance Weighting of the Customer Requirements.
3. The third part of QFD is the Technical Requirement section. This describes the product or service in the terms of the company and it includes all the measurable characteristics of the product / service that might be related to meet the customers’ requirements. Often an additional row is included to illustrate the direction in change of variables which is supposed to result in improvement in product / Service performance.
4. The fourth part of QFD is Interrelationships section. This is used to translate the customer requirements into the technical characteristics of the product / service. Inter-relationships between each of the Customer requirements and technical characteristics are analyzed. Generally, the level of inter-relationship is shown with the help of symbols denoting a 3 point scale (High, medium, low ). Each level of inter-relationship is assigned a score. Generally a score of 5-3-1 is used to denote High-medium-low-none.
Quality Function Deployment (QFD)
119
5. The fifth part of QFD is the Roof. This is used to denote the areas where the technical requirements characterizing the product or service support or inhibit each other. Where there is a deterioration because of interaction, “-” sign is used. Where there is an improvement in one characteristic because of the other, “+” sign is used.
6. The final component of QFD is the Targets. It summarizes the conclusions drawn formthe matrix and team’s discussion. It consists of three parts: a) Technical Priority b) competitive Benchmarks and c) Targets.
a) Technical Priority signifies the relative importance of each of the technical requirement in meeting the Customers’ specified needs. It is calculated by summing up the products of interrelationships weightings with the overall weighting in the planning matrix.
b) Competitive Benchmarking: Each of the technical requirements identified as important characteristic of the product or service are compared vis-à-vis the existing product and the competitors’ product.
c) Targets: The final outcomes of QFD are the targets. These are a set of target engineering values to be met by the new product or service.
Quality Function Deployment (QFD)
120
Structure
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
A I CD M
Techincal Priorities
1
2
3
5
4
6
7
Quality Function Deployment (QFD)
121
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
Customer Requirement : Basically focus on desired deliverable/outcomes from the
process
Customer Importance : Rating of Customer requirement on a scale of 1-5
Complete Resolution 5first Time Resolution 5good Customer Service 4Ontime delivery 5Aware of support boundaries 3Quick Response 5Customer Education 3Technical Priorities% of total
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
gCustomer Requirement
Technical Requirement
STEP - 1
122
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
Technical Requirement : This describes the product or service in the terms of the company and it includes all the measurable characteristics of the product / service that might be related to meet the customers’ requirements ; Basically focus on input parameters which will have direct/indirect impact on deliverable/outcomes of the process;
Complete Resolution 5first Time Resolution 5good Customer Service 4Ontime delivery 5Aware of support boundaries 3Quick Response 5Customer Education 3Technical Priorities% of total
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
g
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
Customer Requirement
Technical Requirement
STEP - 2
123
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
Roof : It shows the inter-relationship between two or more inputs Parameters ` + ’ means positive relationship` - ’ means negative relationship
Complete Resolution 5first Time Resolution 5good Customer Service 4Ontime delivery 5Aware of support boundaries 3Quick Response 5Customer Education 3Technical Priorities% of total
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
gCustomer Requirement
Technical Requirement
++
STEP - 3
124
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
Inter-Relationship Block : It shows the inter-relationship between inputs parameters and output deliverable` 5 ’ means Strong relationship` 3 ’ means Medium relationship` 1 ’ means Weak relationship` 0 ’ means No relationship
Complete Resolution 5 5 5 1 5 5 1 3first Time Resolution 5 5 5 1 5 5 1 2good Customer Service 4 3 5 1 1 3 5 3Ontime delivery 5 3 3 5 5 1 1 3Aware of support boundaries 3 1 3 1 3 3 1 3Quick Response 5 3 3 5 3 3 3 3Customer Education 3 5 5 1 1 5 3 2Technical Priorities% of total
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
g
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
Customer Requirement
Technical Requirement
++
STEP - 4
125
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
Planning Matrix :
Our Product : Rating of existing product/service on scale of 1-5 against customer requirements
Competitor A/B : Rating of Competitor product/services against our customer requirements
Competitor Rating shall be benchmark rating
Complete Resolution 5 5 5 1 5 5 1 3 4 4first Time Resolution 5 5 5 1 5 5 1 2 4 3good Customer Service 4 3 5 1 1 3 5 3 4 4Ontime delivery 5 3 3 5 5 1 1 3 5 4Aware of support boundaries 3 1 3 1 3 3 1 3 4 4Quick Response 5 3 3 5 3 3 3 3 4 4Customer Education 3 5 5 1 1 5 3 2 4 3Technical Priorities % of total
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
g
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Total
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
Customer Requirement
Technical Requirement
++
STEP - 5
126
Planned Rating : Target rating for customer requirement based on competitor rating
Improvement Factor : Ratio of Planned rating and Our product Rating
Sales Point : Rating of Sales team (sales perception)on ability to sell the product/services based on how well each customer need is met, on scale of 1.1, 1.3 and 1.5
1.1 : Low inter-relationship
1.3 : Medium inter-relationship
1.5 : Strong inter-relationship
Complete Resolution 5 5 5 1 5 5 1 3 4 4 4 1.3 1.3first Time Resolution 5 5 5 1 5 5 1 2 4 3 4 2.0 1.3good Customer Service 4 3 5 1 1 3 5 3 4 4 4 1.3 1.1Ontime delivery 5 3 3 5 5 1 1 3 5 4 5 1.7 1.3Aware of support boundaries 3 1 3 1 3 3 1 3 4 4 4 1.3 1.1Quick Response 5 3 3 5 3 3 3 3 4 4 4 1.3 1.1Customer Education 3 5 5 1 1 5 3 2 4 3 4 2.0 1.3Technical Priorities % of total
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Total
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
gCustomer Requirement
Technical Requirement
++
STEP - 5
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
127
Complete Resolution 5 5 5 1 5 5 1 3 4 4 4 1.3 1.3 8.7 14.97%first Time Resolution 5 5 5 1 5 5 1 2 4 3 4 2.0 1.3 13.0 22.45%good Customer Service 4 3 5 1 1 3 5 3 4 4 4 1.3 1.1 5.9 10.13%Ontime delivery 5 3 3 5 5 1 1 3 5 4 5 1.7 1.3 10.8 18.71%Aware of support boundaries 3 1 3 1 3 3 1 3 4 4 4 1.3 1.1 4.4 7.60%Quick Response 5 3 3 5 3 3 3 3 4 4 4 1.3 1.1 7.3 12.67%Customer Education 3 5 5 1 1 5 3 2 4 3 4 2.0 1.3 7.8 13.47%Technical Priorities 224 244 131 211 211 112 1133 57.9% of total 20% 22% 12% 19% 19% 10%
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
g
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Total
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
Customer Requirement
Technical Requirement
++
Overall Weighting : Product of Sales Point, Improvement factor and Customer Importance
Technical Priorities : signifies the relative importance of each of the technical requirement in meeting the Customers’ specified needs. It is calculated by summing up the products of interrelationships weightings with the overall weighting in the
planning matrix.
STEP - 6
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
5 8.7 43.5
5 13 65
3 5.9 17.7
3 10.8 32.4
1 4.4 4.4
3 7.3 21.9
5 7.8 39
224
128
CustomerRequirements
INTER- RELATIONSHIPS
Targets
Technical Requirements
Roof
PlanningMatrix
Techincal Priorities
1
2
3
5
4
6
7
Targets : The final outcomes of QFD are the targets. These are a set of target engineering values to be met by the new product or service.
STEP - 7
Complete Resolution 5 5 5 1 5 5 1 3 4 4 4 1.3 1.3 8.7 14.97%first Time Resolution 5 5 5 1 5 5 1 2 4 3 4 2.0 1.3 13.0 22.45%good Customer Service 4 3 5 1 1 3 5 3 4 4 4 1.3 1.1 5.9 10.13%Ontime delivery 5 3 3 5 5 1 1 3 5 4 5 1.7 1.3 10.8 18.71%Aware of support boundaries 3 1 3 1 3 3 1 3 4 4 4 1.3 1.1 4.4 7.60%Quick Response 5 3 3 5 3 3 3 3 4 4 4 1.3 1.1 7.3 12.67%Customer Education 3 5 5 1 1 5 3 2 4 3 4 2.0 1.3 7.8 13.47%Technical Priorities 224 244 131 211 211 112 1133 57.9% of total 20% 22% 12% 19% 19% 10%
Benchmarking all productsCompetitor product ACompetitor product BDesign Targets
Cus
tom
er im
port
ance
Exp
erie
nce
Tra
inin
g
Tec
hnol
ogy
Pro
cess
too
ls
Qua
lity
mon
itorin
g
Cul
ture
% o
f T
otal
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
rat
ing
Total
Impr
ovem
ent
fact
or
Sal
es p
oint
Ove
rall
wei
ghta
ge
Customer Requirement
Technical Requirement
++
129
4 House QFD
CustomerMeasurements
(HOW’s)
Houseof
Quality # 1
CustomerCustomerHouseHouse
CustomerCustomerHouseHouse
Cu
sto
me
r C
TQ
’s(W
HA
T’s
)
InternalActions(HOW’s)
Houseof
Quality # 2
FunctionFunctionHouseHouse
FunctionFunctionHouseHouse
Cu
sto
me
r M
ea
sure
me
nt
s (W
HA
T’s
)Process
Requirements(HOW’s)
Houseof
Quality # 3
ProcessProcessHouseHouse
ProcessProcessHouseHouse
Inte
rna
lA
ctio
ns
(WH
AT
’s)
ProcessVariables(HOW’s)
Houseof
Quality # 4
ControlControlHouseHouse
ControlControlHouseHouse
Pro
cess
Re
qu
irem
en
ts(W
HA
T’s
)
Define Improve Control
130
Points to Remember in QFD
Do’s
Focus on the end-user
Charts as the ends & not the
means
Find reasons to succeed, not
excuses for failure
If there are no “tough spots” the
first time, it probably is not being
done right !
Don’ts
Don’t apply QFD on everything
Avoid too much chart focus
Don’t hurry up & get done
131
A Restaurant Example – Manila Pizza CTQs from VOC (WHAT’s)
– Type of Menu 3– Economic aspect 4– Presentation of staff 3– Type of service 3– Quality of service 4– Quality of Food 5
Measurements on CTQs (HOW’s)
– Varieties available in terms of Chinese, Indian, Continental food– Price – Availability of parking space– Food served at right temperature– Waiting time for food/order– Options to pay ( Cash/Debit card/Credit card)– Cleanliness of cutlery / attendant’s attire– Staff courtesy– Taste of food– Ambience
Importance Rating
132
Type of Menu 3 5 3 3 1 5 1 3 4 4 4 1.3 1.3 5.2 13.10%
Economic aspect 4 3 5 1 5 5 3 2 4 3 4 2.0 1.3 10.4 26.20%
Presentation of staff 3 1 3 1 1 1 5 3 4 4 4 1.3 1.1 4.4 11.08%
Type of service 3 3 3 5 5 1 1 3 5 4 5 1.7 1.3 6.5 16.37%
Quality of service 4 1 3 5 5 3 1 3 4 4 4 1.3 1.1 5.9 14.78%
Quality of Food 5 3 3 1 1 5 1 3 4 4 4 1.3 1.1 7.3 18.47%Technical Priorities 109 140 100 131 143 78 700 39.7% of total 16% 20% 14% 19% 20% 11%
Benchmarking all products 3 5 10 mins3 opts 5 5Competitor product A 3 4 USD15mins2 opts 4 5Competitor product B 4 6USD10mins3 opts 5 3
Design Targets 3 V
arit
ies
5 U
SD
pe
r pe
rso
n
10m
ins
All
thre
e o
ptio
ns
Ab
ove
4
Ab
ove
4
Total
Impr
ove
men
t fa
cto
r
Sa
les
poi
nt
Ove
rall
we
igh
tag
e
% o
f T
ota
l
Our
pro
du
ct
Co
mp
etit
or
A
Co
mp
etit
or
B
Pla
nne
d ra
ting
Wai
ting
tim
e
Opt
ion
s to
pa
y
Ta
ste
of
food
Am
bie
nce
Cu
sto
me
r im
port
ance
Va
rietie
s a
vaila
ble
Pri
ce
Customer Requirement
Technical Requirement
+++
+
Capturing Customer
Requirement along with importance
rating
Define
133
Taste as per expectation 5 5 5 5 1 5 5 3 5 4 5 1.7 1.3 10.8 20.79%
Temperature of Food 4 5 1 3 5 1 1 3 4 3 4 1.3 1.3 6.9 13.31%
Add-on over ordered food 1 3 3 1 1 5 5 3 4 5 5 1.7 1.1 1.8 3.52%
Effect on Health 5 3 3 1 1 1 5 1 5 4 5 5.0 1.3 32.5 62.38%Technical Priorities 192 164 109 80 103 233 881 52.1% of total 22% 19% 12% 9% 12% 26%
Benchmarking all products min 4yrsPG 5 5 5 5Competitor product A 5yrs PG 5 3 4 3Competitor product B 8yrs PG 5 5 5 5
Design Targets 5yrs
PG
10m
ins
5 Abo
ve 4
Abo
ve 4
Cus
tom
er im
port
ance
Exp
erie
nce
of C
hef
Qua
lific
aito
n o
f Che
f
Com
mun
icat
ion
syst
em
Del
iver
y S
yste
m
Raw
Mat
eria
l ava
ilabi
lity
Raw
Mat
eria
l Qu
ality
% o
f Tot
al
Our
pro
duct
Com
petit
or A
Com
petit
or B
Pla
nned
ra
ting
Total
Impr
ovem
ent
fact
or
Sal
es p
oin
t
Ove
rall
wei
ghta
ge
Customer Requirement
Technical Requirement
++ +
Improve
134
Freshness of Vegetable used5 5 5 1 1 3 4 4 4 4 1.0 1.3 6.5 28.77%
Quality of Food grain used 4 1 5 3 3 1 5 4 3 4 0.8 1.3 4.2 18.41%
Storage System 1 5 1 1 5 5 4 4 4 4 1.0 1.1 1.1 4.87%
FIFO 5 5 1 1 1 5 3 5 4 5 1.7 1.3 10.8 47.95%Technical Priorities 96 65 31 35 83 311 22.6% of total 31% 21% 10% 11% 27%
Benchmarking all products SOP PG Cert. 5 5Competitor product A SOP PG Cert. 5 5Competitor product B SOP PG Cert. 5 5
Design Targets SO
P
PG
5 5 5
Total
Impr
ovem
ent f
acto
r
Sal
es p
oint
Ove
rall
wei
ghta
ge
% o
f Tot
al
Our
pro
duct
Com
petit
or
A
Com
petit
or
B
Pla
nne
d ra
ting
Inw
ard
Qua
lity
Insp
ect
ors
Sup
plie
r
Tra
inin
g to
Mat
eria
l Han
dle
rs
Inve
ntor
y M
anag
emen
t
Cus
tom
er im
port
ance
Tem
pera
ture
of S
tora
ge
syst
em
Customer Requirement
Technical Requirement
+
+Control
135
1. Refer to your workbook.
2. Prepare the first two houses of the QFD for a given example
Preparing the first two houses of the QFD
Exercise 1.4 (30 minutes)
136
Brainstorming & Multi-voting Example
• Below table illustrates the number of group members & total ideas generated
• Now each member gives votes to ideas (maximum one vote to each idea) & below is the vote distribution for ideas
• Top ideas whose vote count adds upto 32 are as below (30 is not possible), there are 7 such ideas
137
Brainstorming & Multi-voting Example
• Take these 7 ideas for further round of multi-voting
• Give each member 4 votes (round off 50% of 7 to next higher integer) & ask them to distribute these 4 votes among these 7 ideas. Below could be the distribution in this fresh round of voting
• Take top ideas whose vote count adds upto 9
• This list is manageable
138
Points to Remember in Brainstorming
• All ideas are important, don’t out rightly reject any idea
• Participation should be ensured from all team members
• To ensure this, project teams could use the round-robin method of idea generation
• It’s advised to use the Black Belt as the facilitator here
139
1. Refer to your workbook.
2. Do a multi-voting exercise for a given example
Multi-voting
Exercise 1.5 (15 minutes)
140
A Right Project Selection is key to success
Right Project Selection• Selecting the right project
can have a tremendous effect on your business. If done properly,
• Processes will function more efficiently in 3 to 6 months, employees will feel satisfied and
• Appreciated for making business improvements and ultimately stakeholders will see the benefit
Wrong Project Selection• If project selection is
done improperly, a project may be selected that doesn't have the full business buy-in, project roadblocks may not be removed due to other business priorities, the team may feel Ineffective and the end result may be less than ideal. No one wins in this situation
141
Once the potential projects are identified next step is to verify, whether these are six sigma project or not ?
There are two type of filters to qualify the potential projects as six sigma projects. They are 1) Company filters and 2) Six Sigma Filters
Six Sigma Project Selection Filters
1. Company Filters and 2. Six Sigma Filters
142
1. Company Filters
Why do it? Drop
Why do it? Drop
Aligned withcompany Strategy?
Aligned withcompany Strategy?
No
Does project lead USD--- min. per annum?
Does project lead USD--- min. per annum?
No
Scope is too large. Consider making Multiple projects.
Scope is too large. Consider making Multiple projects.
Can it be completed in4 to 6 months?
Can it be completed in4 to 6 months?
Yes
Yes
1) Is it a severe pain in the process 2) Is documental proof available to prove pain 3) Is the pain sensed by process experts
1) Is it a severe pain in the process 2) Is documental proof available to prove pain 3) Is the pain sensed by process experts
No
Move to Six Sigma Filters
Yes (all 3)Yes
No
143
Implement the solution
Implement the solution
Is the solutionalready known?
Is the solutionalready known?
yes
No
yes
no
Implement a datacollection plan
Implement a datacollection plan
Is needed data available to quantify the
problem?
Is needed data available to quantify the
problem?
No
Derive and implement the solution
Derive and implement the solution
Is the rootcause known?
Is the rootcause known?
Yes
Implement the other team’s solution
Implement the other team’s solution
Is someone elseworking on the problem?
Is someone elseworking on the problem?
Yes
Yes
No
Six Sigma Potential Project
No
2. Six Sigma Filters
144
Types of Projects
1. Projects cutting across processes focusing on CTQ.
2. Projects cutting across CTQ focused on process.
3. Projects focusing on a specific CTQ for a process.
SS Project Execution Process MapVoice Of Customer & Pain Areas
MBB
BB
GB & Team
Client & Top Management
Recognition of Improvement opportunity
Initial Financial Validation
Champion
Project Evaluation &Methodology
selection (DMAIC / DMADV)
Problem Definition
Project Agreement& Target setting
Finance Controller
Project Kick Off
(MBB & Champion are secondary resp.)
Project Execution ( DMAIC / DMADV )(MBB & Champion
are sponsor )
Project Validation,
Closure and sign-
off by champion,MBB and Financial controller
Project Charter
(Champion / sponsor)
Project Review at Pre-defined Frequency
(MBB & Champion)
Integration & Deliver
Final Financial Validation
146
Improvement Project Execution
Business Process
Improvement
Business Process
Improvement
REIS Project
DMAIC Project
DMADV Project
1- MBB 1- BB, 1- GB4 - PA
1- BB, 1- GB4 - PA
1- GB4 - PA
Project 1
Project 2
Project 3
Project 1
Project 2
Project 3
Project 1
Project 2
Project 3
147
Six Sigma
DMAIC Methodology
148
Overview of DMAIC Methodology
DMAIC Vs DMADV
DMAIC Methodology– Define– Measure– Analyze– Improve and – Control
149
DMAIC Vs DMADV
DMAIC
DMADV
Process Management
IMPROVEMENT PROCESS
(DMAIC)
Define, Measure, Analyse, Improve and Control
Improve processes, products, services, and organisation to 6 Sigma quality
CREATION PROCESS
(DMADV)
Define, Measure, Analyse, Design and Validate
Create new processes, products, services, and plants to 6 Sigma quality.
It is also Called DFSS
PROCESS MANAGEMENT
Leverage and sustain the gains achieved by improvement and creation with BPMS, QMS, COPC etc
150
Step 5: CONTROL
Step 2: MEASURE
Step 1: DEFINE
Step 4: IMPROVE
Step 3:ANALYZE
DMAIC
DMAIC
Define Define project goals & customer (internal & external) deliverables
Measure Measure the process to determine current performance
Analyze Analyze and determine the root cause of the defects
Improve Improve the process by eliminating defects
Control Control future performance
DMAIC Methodology
151
DefineObjective: is to define the problem in a clear manner and in a way that is related to an internal or external customer.
• Fully trained team, committed to work on improvement project.• Customers identified and defined (CTQ’s)• Project charter and Process map
Deliverables:
Check points:• Trained team
Customers (and CTQ’s)• Project Charter• Business process mapping (SIPOC)• Process Map
152
Objectiveto measure what you care about most, making certain that your measurementapproach is sound and not based on questionable formulas or data.
Deliverables:• Key measures identification, data collection plan, data on process variation
performance baseline, sigma level calculation.
Check points:• Identification of Key measures - Defining high impact defects• Data Collection Plan - Data collection• Measurement system analysis• Process Variation • Long term and short term variability accounted for.• Performance Baseline/Sigma Calculation• Measure baseline process performance (capability, yield, sigma level).
Measure
153
Deliverables:• Data and process analysis, root cause analysis, quantifying the
gap/opportunity.
Check points:• Data and Process Analysis – Identification of gaps between current• performance and the goal performance• Root Cause Analysis- Verify and quantify the root causes of variation• Quantifying the Gap/Opportunity - Determine the performance gap.
Analyze
Objectiveis to look for the critical root causes of the variability by applying statistical tools to determine what factors are contributing to the problem.
154
Objectiveis to determine and confirm the optimal solution.
Deliverables:• Generate (and test) possible solutions, select the best
solutions, design implementation plan
Check points:• Generating (and Testing) Possible Solutions• Selecting the best Solution (s) • Designing Implementation Plan
Improve
155
Objectiveis to be sure the quality improvements remain in effect and the problem
does not recur.
Deliverables:• Documentation and implementation of monitoring plan, standardized
process, documented procedures, response plan established and deployed, transfer of ownership (project closure).
Check points:• Monitoring Plan• Process Standardization • Documented Procedures • Transfer of Ownership (Project Closure)
Control
156
Define Phase
157
1.0 Define
Opportunities
Define
Main ActivitiesStep 1: • Identify customer & their care about. • Convert their needs in to Critical to Satisfaction (CTS) i.e.
CTQ – Critical to quality, CTC – Critical to Cost & CTD Critical to Delivery.
Step 2: • Develop Project Charter reporting improvement
opportunity and effective project team
Main Tools used Define Phase Outcomes
• List of Project CTQs• QFD/CTQ Tree• Signed off Project
charter • Team Charter• CCRs• SIPOC• Macro Level Process
Map
VOC
Project Timeline
Project team
Project Scope
Goal Statement
Business Case
Problem Statement
Project Charter
Baseline
Target
Gap Analysis
Process Map
S I O CProcess
SIPOC
Define
158
Define Phase Topics
Project Charter
Team Charter
SIPOC
Process Mapping– Top Down Model– Cross Functional Process Mapping &– ICOM model
159
Project Charter
160
What Is A Project Charter?
• A project charter is a document that provides purpose and goals for an improvement team
Six Major Elements of a Project Charter1. Business Case
Explanation Of Why To Do This Project
2. Problem And Goal StatementsDescription Of The Problem/Opportunity And objective In Clear, Concise And Measurable Terms
3. Project ScopeDefined project beginning and end points
4. MilestonesKey Steps And Dates To Achieve Goal
5. RolesPeople, Expectations, Responsibilities
6. Financial Impact Savings, impact on bottom-line
Project Charter
161
Business Case
Business Case Development• The business case describes the benefit for undertaking a
project. The business case addresses the following questions:– Does this project align with other business initiatives?– What is the focus for the project team?
• What impacts will this project have on other business units and employees?
• What benefits will be derived from this project?
• Has the value of the benefits been quantified?
162
Problem Statement
Description Of The “Pain”
What Is Wrong Or Not Meeting Our Customer’s Needs?• Who is the customer of the process ?• What is the process ?
When And Where Do The Problems Occur?
How Big Is The Problem?
What’s The Impact Of The Problem?• If I let it be what will happen ?• If I reduce it what will happen ?• If I increase it what will happen ?
163
Problem Statement Example
Poor Example:Weak Problem statementOur Fatal accuracy score is at 80% against the client target of 95% and we Need to improve it
Improved Example:Fatal accuracy score (what) this quarter (when) has been observed to be at 80% for the last 4 months (extent) against the client target 95% which has significant impact on customer satisfaction. (impact)
164
The Problem Statement
Key Considerations/Potential Pitfalls
• Is The Problem Based On Observation (Fact) or Assumption (Guess)?
• Does The Problem Statement Prejudge A Root Cause?
• Can Data Be Collected By The Team To Verify and Analyze The Problem?
• Is The Problem Statement Too Narrowly or Broadly Defined?
• Is A Solution Included In The Statement?
• Would Customers Be Happy If They Knew We Were Working On This?
165
The Goal Statement
Project Objective
• Definition of The Improvement The Team Is Seeking To Accomplish
• Starts With …. Reduce, Eliminate, Control, Increase.
• Tends To Start Broadly – Eventually Should Include Measurable Target And Completion Date
• Must Not Assign Blame, Presume Cause, Or Prescribe Solution!
166
SMART Problem And Goal Statements
Specific
Measurable
Attainable
Relevant
Time Bound
167
Goal Statement Example
Poor Example:Put in place a Transaction monitoring evaluation system to increase the fatal accuracy.
Improved Example:Increase the fatal accuracy from the existing 80% to 95% by the end of first quarter 2005.
168
Project Scope
• What Process Will The Team Focus On?
• What Are The Boundaries Of The Process We Are To Improve? Start Point? Stop Point?
• What Resources Are Available To The Team?
• What (If Anything) Is Out Of Bounds For The Team?
• What (If Any) Constraints Must The Team Work Under?
• What Is The Time Commitment Expected Of Team Members? What Will Happen To Our “Regular Jobs” While We Are Doing The Project?
169
1. Identify the customer • Who receives the process output?• May be an internal or external customer
2. Define customer’s expectations and needs• Ask the customer• Think like the customer• Rank or prioritize the expectations
3. Clearly specify your deliverables tied to those expectations• What are the process outputs?• Tangible and intangible deliverables• Rank or prioritize the deliverables• Rank your confidence in meeting each deliverable
4. Identify CTQ’s for those deliverables• What are the specific, measurable attributes that are most critical in
the deliverables• Select those that have the greatest impact on customer satisfaction
8 Steps to Scope a Project
170
5. Map your process• The process of producing the deliverables• The process as it is working prior to the project• If you are delivering something, there is a process, even if it has not
been formalized6. Determine where in the process the CTQ’s can be most seriously
affected• Use a detailed flowchart• Estimate which steps contain the most variability
7. Evaluate which CTQ’s have the greatest opportunity for improvement• Consider available resources• Compare variation in the processes with the various CTQ’s• Emphasize process steps which are under the control of the team
conducting the project8. Define the project to improve the CTQ’s you have selected
• Define the defect to be attacked
8 Steps to Scope a Project
171
Importance of Scoping
• Poor/improper scoping may result in following:
– Team loses interest in the project
– Project becomes difficult to implement
– Even after implementation, the desired/significant benefits are not
seen
– Team focuses on trivial pain areas, and missing out the vital ones
– Process selected is too broad to handle or too small to realise
breakthrough improvements.
172
Milestones
• A preliminary High Level Project Plan with dates
• Tied to phases of DMAIC process
• Should be aggressive (don’t miss “ window of opportunity “)
• Should be realistic ( don’t force yourself into corrective rather than preventative solutions)
Week : 1 2 3 4
Review Charter with Champion X
Collect VOC X X
Complete Map X X
Validate Map X
Collect Data X
1-May 7-May 16-Jun 23-Jun 7-Jul
Define
Measure
Analyze
Improve
Control
173
Take Aways –Project Charter
• Key elements of a charter include: Business Case, Problem and Goal • Statements, Project Scope, Milestones, and Roles.
• The team charter is a vital part of the project’s overall success. It communicates the project direction to all members of the team.
• A Problem Statement describes what is wrong while a Goal Statement defines the improvement objective.
• A charter clarifies what is expected of the project team, keeps the team focused, keeps the team aligned with organizational priorities, and transfers the project from the champion to the improvement team
174
Project Charter
Per
iod
Goal statement:Goal statement:
Problem Statement :Problem Statement :
Responses
Black Belt
Project Leader
Project Sponsor
Name Sign
ME
MB
ER
S
Business caseBusiness case
Revenue Enhancement Expense Reduction
Loss Avoidance
Costs
Start End
Basic ScheduleBasic Schedule
In-scope :In-scope : Measurable Goals :Measurable Goals :
Unit: Percentage
Current Target Current Target
Estimated Saving Expected, ( 1 Fin. Year)Estimated Saving Expected, ( 1 Fin. Year)
50,000 USD
Control 06.05.05
Improve 21.04.05
Analyze 07.04.05
Measure 25.03.05
Define 16.03.05
Out-scope :Out-scope :
Reduction of rework from 38% to 5%
xxxxxx
Xxxx (YB)
xxxxxxxxx
The rework percentage in this quarter (Apr’05 to June’05) has been observed to be at 38% based on the system reports which is leading to over stretching of PE to meet the production targets.
All exiting sub – processes in the process 38 5
jkafkj
jkjklajdxcv
1 Mar 05 30 May 05
Rework is one of the main concern area in our xxx process. Due to rework, most of our PE are over stretching their working hours. Our quality percentage is being maintained below the SLA.
The client also concern about the issue and suggested to take action with in 3 months to improve the same.
Failure of the same may lead to employee attrition and loss of business.
Rework reduction
2.5 4.2Any new sub-process to be adding with in the project period
Example
IMPCHG/QLTY/TMPL/6111 Version 1.0, Copy if printed
MBB
175
Scope the Project GB, BB, ChampionDevelop Team Charter GB, BB
Project Scoping responsibility Matrix
176
1. Refer to your workbook.
2. Define in-scope & out-scope given project example
Project Scoping
Exercise 1.6 (15 minutes)
177
Team Charter
178
Team Charter explains following •How do you want the champion to work with the team?
• Is the team’s role to implement or recommend?
• When must the team go to the champion for approval? What authority does the team have to act independently?
• What and how do you want to inform the champion about the team’s progress?
• What is the role of the team leader and the team coach?
• Are the right members on the team? Functionally? Hierarchically?
Team Roles Refresh
Sponsor • Review the project progress once a month• Provide/modify direction/alignment with
business realities• Provide resources required from time to time • Remove Roadblocks
Black Belt/MBB • Provide content knowledge on Six Sigma
tools to the team
Project Leader • Keep the team focused. Arrange logistics
and team meetings and raise issues with Sponsor
Team Member (s) • Participate in meetings, collect data, do
analysis using Quality tools, provide subject matter expertise related to process
Team Charter
179
1. Refer to your workbook.
2. Practice writing problem and goal statements
Team Charter – Breakout Activity
Exercise 1.7 (45 minutes)
181
SIPOC
182
SIPOCSIPOC:• A tool to identify all relevant elements of a process
• Helps to understand a complex process better
•Graphic display of steps, events and operations that constitute a process
S - Suppliers
I - Inputs
P - Process
O - Outputs
C - Customers
183
Suppliers– People who provide input to the process
Inputs– Information, material etc., goes into the process from some other group of
people (supplier)
Process – Process is a series of activities that takes an input, adds value to it and
produces an output for a customer
Outputs
– Output of a process creating a product or service that meets a customer need
Customers
– Users of the output
SIPOC
184
S I P O C
Requirements Requirements
Suppliers Inputs Process Outputs Customers
The Key Quality and Delivery Requirements
Placed On Your Suppliers.
Measures That Are Internal To Your
Process. They IncludeQuality and Delivery
Measures Important ToYour Internal Customers
As Well as Waste andCycle Time Measures.
Output MeasuresAre Measures UsedTo Determine How
Well Customer NeedsAnd Requirements
Are Met.
Input Measures Process Measures Output Measures
FlowThinking
Information Flow & Measures
185
SIPOC: Uses
• To know who supplies input to the process
• To know what are the inputs to the process
• To know step by step flow of process
• To know the outputs of process
• To know the customer of a process
186
Steps to create SIPOC1. Attaining a full understanding of all the steps of a process. This is done by
looking at processes from customer’s point of view.
2. Clearly define the process start and end boundaries
3. Brainstorm list of all process steps. Go on the floor, walk through the process and interview people working on the process as needed.
4. Recorded process steps using a sticky-note method. In this method each step in the process is recorded on a sticky-note and built in front of the individual completing the work.
5. Discuss, review & modify process step sequence to agree on “As Is” process map.
6. Add suppliers, inputs, outputs, customers
7. Add Input, Process, & Output measures
187
SIPOC: Format
Supplier Inputs Process Output Customer
A
B
C
X1
X2
X3
Y1
Y2
Y3
Alpha
Beta
Gamma
188
ProcessInputSupplier
Contract Record
SIPOC LEVEL I Contract Management : Contract Publishing
CustomerOutput
ABC Bank
3.Assign Contract Number
Electronic Documents
Global Sourcing-ABC Retained
Job Aid: Four-Eye principle,
Commodity Classification, Template from Emptoris, A-signatory list
Scanning Team
Vendor folder in shared drive
Vendor Number from SAP, Contract
template from Emptoris
Contract Number.xls,
Contract Information Sheet
6. Review Contract record
1.Receive scanned images
2.Sanity Check
4. Upload Data
5.Create Vendor Folder in the shared drive
7.Publish Contract
8.Update Contract Tracking Sheet
P2P Arthur
SIPOC: Example 1: Contract Management : Contract Publishing
189
ProcessInputSupplier
Resolution Document
SIPOC LEVEL I Contract Management : Contract Enquiry
CustomerOutput
ABC Bank
1. Receive Query and review Query ReceivedGlobal Sourcing-
ABC Retained
Job Aid
P2P Contract Management
Onshore Team Contract Number.xls, Authorization
Sheet
3. Verify Authorization
4. Check Availability
6. Communicate to the user
P2P Arthur
Communication to the Enquirer
5. Get information/document
2. Update Query tracker sheet
7. Close query in the tracker
Contract Management : Contract Enquiry
190
ProcessInputSupplier
Review Record
SIPOC LEVEL I Contract Management : Contract Review
CustomerOutput
ABC Bank2. Check Contract Compliance
Electronic Documents
Global Sourcing-ABC Retained
Job Aid
P2P Contract Management
Onshore Team
3. Validate Contract Data
4. Check terms & conditions Vs ABC standard terms & conditions
Emptoris
Contract Number.xls
P2P Arthur
1. View & analyze electronic document
5. Create Contract Summary records
Contract Management : Contract Review
191
Escalate?
S I P O C
Agents Team
ManagersACD
IncomingCall
CompletedCall
PerformanceReports
Client
Caller
Input Measures Process Measures Output Measures
Call Opening
Confirmation
Verification
Assistance
Response
Closing
After Call Work
InterpreterSpanish?Y
Y
N
InternalCustomers
Auditors IVR
Number of agents, Managers, Auditors,
Certified / On training
Experience /Vintage
Process Knowledge
Listening / Accent, speaking & keyboarding Skills
App
Number of ACD calls, Pattern, Q
Spanish?
Nature of Query
App
Response
Uptime of ACD, Alltel
Noise?
No of lines per step, Time per step, Time holding for response, Call Accuracy, Quality, Cycle Time
Speed of answer
Calls abandoned & Time
Quality
Talk Time
Wrap Time
Hold time
Resolved ?
Satisfaction score
ASA
Agent productivity
Talk Time
Wrap Time
% Abandoned
Aband Time
AHT,ACW
SIPOC: Example 2:Call Handling
192
1. Refer to your workbook.
2. Draw the SIPOC for your process
SIPOC
Exercise 1.8 (20 minutes)
193
Process Mapping
194
Process
PROCESS
INPUT
OUTPUT
RESOURCES
MEASURES
SLAs
“Process is a series of activities that takes an input, adds value to it and produces an output for a customer ’’
195
Process Mapping Definition
Process mapping is a graphical display of
steps, events and operations that constitute a
process.
196
Why Process Mapping?
• Validates our understanding of the process with the client (The way work gets done)
• Identifies hidden process steps
• Helps to understand weak links in the process
• Eliminate the ambiguity & brings standardization
• Helps to identify data collection point during measure phase
• Imparts training to others
• To design the “ to be” process
197
Flowchart Vs. Process Map
Flowchart Process Map
May only shows the connected steps in a process
Goes further showing who is doing what, with whom, when, for how long and with what documents.
It shows how operational decisions are made
198
Process Map Symbols
Symbol MeaningStart or End of Process
Activity or Process Step
Decision or Inspection Point
Delay
Connector
Document
Data
Direction of Flow
199
Basic Types of Process Maps
• Linear Process Maps
It can be used when
– The process is not very complicated – Micro level (detailed) map is “dropped down”
from the macro (high level) process steps.
• Cross Functional Process Maps (also known as “Swim Lane process map”)
It can be used when– The process complicated consists of several
activities between different departments or groups.
Call Opening
Confirmation
Verification
Assistance
Response
Closing
Dept 1
Step 4
Step 1
Step 3
Dept 2
Dept 3
Dept 4
Step 1
200
“As Is” Process Map in define phase
• Objective of “As Is” process map in define phase is to identify hidden process steps in process.
• It helps to understand weak links in the process
• It also helps to identify non value adding process steps in the process
201
Steps to create “As Is” Linear Process map
1. Attaining a full understanding of all the steps of a process. This is done by looking at processes from customer’s point of view.
2. Clearly define the process start and end boundaries
3. Brainstorm list of all process steps. Go on the floor, walk through the process and interview people working on the process as needed.
4. Record individual process steps on the sticky-note / post it.
5. Discuss, review & modify process step sequence to agree on “As Is” process map.
202
Analyze “As Is” Linear Process map
Next Step : Analysis of As Is process map
1. Analyze process map to identify NVA’s.(identify unnecessary approvals, isolating rework, removing duplicate forms and investigating decisions leading to no results)
2. Identify data collection & decision making points.
3. Compare with “To Be” process map to identify the gaps.
203
“As Is” Linear Process map : Example 1
InterpreterSpanish?
Call Opening
Confirmation
Verification
Assistance
Response
Assistance
Escalate?
Call closing
After call work
A
A
Yes
Yes
No
No
Collect data on % Spanish calls
Collect data on % escalated calls
Collect data on % accuracy
On holdNVA
NVA = Non value added activity
204
“As Is” Linear Process map : Example 2
Business UnitBusiness Unit
Business ProcessBusiness Process
Identify Critical ProcessIdentify Critical Process
Study ProcessStudy Process
Identify CustomersIdentify Customers
List out Customer Expectations/Deliverables
List out Customer Expectations/Deliverables
A
Yes
No
Does
ProcessExists?
B
Does Gap Exists?
Are Metrics
in place?Define Metrics &
implement data collection plan
Define Metrics & implement
data collection plan
A
Validate the MetricsValidate the Metrics
Process Base liningProcess Base lining
Does
Process exists?
No
NoExplore need of
performance excellence
Explore need of performance
excellence
Apply DMADV / Process
Re Engineering
Apply DMADV / Process
Re Engineering
Long Term Approach – Apply DMAIC
Long Term Approach – Apply DMAIC
Quick wins –Fix it
Quick wins –Fix it
Yes
Yes
B
205
Steps to create “As Is” cross functional map1. Attaining a full understanding of all the steps of a process. This is done by
looking at processes from customer’s point of view.
2. List all departments or groups involved in the process. (Record in left column of swim lanes)
3. Clearly define the process start and end boundaries
4. Brainstorm list of all process steps. Go on the floor, walk through the process and interview people working on the process as needed.
5. Record individual process steps on the sticky-note / post it.
6. Discuss, review & modify process step sequence to agree on “As Is” process map.
206
Analyze “As Is” Cross functional Process map
Next Step : Analysis of As Is cross functional process map
1. Analyze process map to identify NVA’s.(identify unnecessary approvals, isolating rework, removing duplicate forms and investigating decisions leading to no results)
2. Identify data collection & decision making points.
3. Compare with “To Be” process map to identify the gaps.
4. Identify the cross functional complexities involved in the process.
207
Cross Functional Process Mapping
Function 1
Function 2
Function 3
Function 4
Step 2
Step 3
Step4
Step 5
Step 6 Step 7
Step 8 Step 9Step 1
208
Example 1Voice Of Customer & Pain Areas
MBB
BB
GB & Team
Client & Top Management
Recognition of Improvement opportunity
Initial Financial Validation
Project Sponsor
Project Evaluation &Methodology
selection (DMAIC / DMADV)
Problem Definition
Project Agreement& Target setting
Finance Controller
Project Kick Off
(MBB & Champion are secondary resp.)
Project Execution ( DMAIC / DMADV )(MBB & Champion
are sponsoror.)
Project Validation,
Closure and sign-
off by champion,MBB and Financial controller
Project Charter
(Champion sponsoror.)
Project Review at Pre-defined Frequency
(MBB & Champion)
Integration in Deliver
209
Take Photograph
Return home
Place film in Pre paid envelope Send to Processor
Send to Photographer
1 Day Delay
Process Negative
Produce Prints
1 hr
Inspect
OK ?
Package for postingY
NAwait Photograph
Inspect Photograph
FramePhotographto be framed
Y
NStore in album
3 days
3 days
Photographer Postal System Processor
The flowchart below depicts the activities involved from taking a photo to developing, framing and storing the result.
Example 2 : Photo Process
210
Imports Supplier Finance Forwarder / Clearing Agent
Customs
Planning
Plan & Schedule
Obtain price &Prepare P.O
Approve P.O
Fax / Mail copy to supplier
Prepare Order Ack &Send to Company
Check paymentterms
Request for LC
Open LC & send Copy to Imports
Verify LC
OK
Not OK
LC
Inform & send Copy to supplier
Verify correctnessOf LC & Inform
Follow-up material
A
Direct Payment or Site Draft
Get Options detailsfrom Logistics if reqd.
Example3: Import process
211
Imports Supplier Finance Forwarder / Clearing Agent
Customs
B
A
Arrange material Collect material
Arrange shipment &Details to company/Supplier
Send documents to companyAnd Supplier
Send documentsTo bank
Prepare checklist forInternal circulation
Arrange InsuranceAdvise finance for
Insurance
Send to clearing agentFor preparation of BE
Receive info from Bank
Send to Imports forcertification
Check & certify docs
Arrange for payment &Retire documents
Send documents to Imports
Send documents toClearing agent
Prepare Bill of Entry &Files with Customs
Example3: Import process
212
Imports Supplier Finance Forwarder / Clearing Agent
Customs
B
Inspect & ascertainCustoms Duty
Inform Company - Imports
Prepare RFC to finance customs duty
Prepare payment and Send to clearing agent
Deposit amount inCustoms & clear the goods
Inform Company-Imports for Collection of material
Inform W.H to collectThe material
W.H – Collect the material& prepare GR
Move toStores.
Example3: Import process
213
Other Types of Process Maps
• Top Down Model • ICOM
214
Macro to Micro (M2M) Process Map
Step 1 Step 2 Step 3 Step 4
T h e P r o c e s s
The Sub-Process
The Micro-Process
It can be used when - Team wants to pay attention to the important process steps in detail.
215
Example 1 : Quotation Sub-Processes
Research Create Proposal Refine Proposal Accept Quote
PreviousAgreement
Look up
PreviousPurchase
HistoryLook up
PreviousQuote HistoryLook up
Cross geo customer quotingLook up
Guided selling tool maps requirements
to offerings
Create initial proposal with all relevant options
Profile defined
Validate customer and proposal pre-credit check credit
rating lookup
Refine proposal(iterative)
Create final proposal
Escalate concession
request
Concession process
Quote Acceptance
Optional indirect
Optional
216
Other process mapping models (additional information)
ICOM
ProcessInput Output
Control
Mechanism
Input: The material / information which enters in to the system with specified process capability
Control : The systems, policies which control or governs the process.
Output : The value added material / information through the process and entitled for next process
Mechanism : The entity which may consume or act as resource for processing. Eg. Machine, computer, Agent
fsdafdf
Single Process
High Level Process Or Multiple processes
217
ICOM Model (additional information)
Level 1Basic processstructure of the business
Level 2
Level 3
1.1
1.2
1.3
1.3.1
1.3.2
1.3.3
Business/Enterprise
Sub-process
Activities withinSub-process
1
Level 0
IncreasingIncreasinglevel level of detailof detail
It can be used when - Team want to map processes hierarchically
218
ICOM Model – Example (Process : Call Handling )
Soft skillsAgent’s training
Server, SoftwareAgent’s training &Process Knowledge
Peripheral
Agent press Button to
attend call
Customer Call
Peripheral
SOPSLA
Agent Verifies
Caller & authorization
Connected Call
IdentifiedGenuine
caller
Rules of verificationData base
Server, SoftwareAgent training
Identification Of Type of call
Obtaining theRequirement
Of caller
IdentifiedType of call
Rules of verificationData base
Server, SoftwareAgent training
Searching and processing
of information
Agent provides information
to caller
Agent press to button to give
access to attend next call
Agent does after Call work (ACW)
Agent check for the Caller
satisfaction
Termination of call
Need to verify from supervisor?
Hold call and Obtain information
from supervisor
Caller’srequirement
Processedinformation
Information as requested
by caller
Answeredcall
Satisfiedcaller
Completedcall
CompletedACW
Agentready
to takenext call
Yes
No
Soft skills
Database
DatabaseSOP
Server, SoftwareAgent’s training &Process Knowledge
Soft skills
Database
Peripheral
SOPSLA
SOPSLA
SOPSLA
219
1. Refer to your workbook.
2. Draw the process map for your process
Process Mapping
Exercise 1.9 (20 minutes)
220
Measure Phase
221
Measure
Main ActivitiesStep 3: • Take the snapshot of the process, how the process
performing currently & fix the baseline. Step 4 : • Validate the measurement system from which we
collect the data.
Tools used
Measure Phase Outcomes• Operational
Definitions• Measurement System
Analysis• Data Collection
Formats and Plans• Process Baseline • capability• Specification limits, • target, defect
definition for Project Y(s)
Benchmarking
Gage R&RData Collection Plan
Process Map
Baseline and target setting
PreparationPreparation FMEA ProcessFMEA Process ImprovementImprovementPreparationPreparation FMEA ProcessFMEA Process ImprovementImprovement
1. Select Process Team
2. Develop Process Map & Identify Process Steps
3. List Key Process Outputs To Satisfy Internal And External Customer Requirements
4. List Key Process Inputs For Each Process Step
5. Define Matrix Relating Product Outputs To Process Variables
6. Rank Inputs According To Importance
7. List Ways Process Inputs Can Vary (Causes) and identify associated Failure Modes and Effects
8. List Other Causes (Sources of Variability) And Associated FM&Es
9. Assign Severity, Occurrence And Detection Rating To Each Cause
10. Calculate Risk Priority Number (RPN) For Each Potential Failure Mode Scenario
11. Determine Recommended Actions To Reduce RPNs
12. Establish Timeframes For Corrective Actions
13. Create “Waterfall” Graph To Forecast Risk
Reductions
14. Take Appropriate Actions
15. Re-calculate All RPNs
16. Put controls into place
Process or Product Name:
Responsible:
Process Step/Part Number Potential Failure Mode Potential Failure Effects
SEV Potential Causes
OCC Current Controls
DET
RPN
Actions Recommended Resp.
Failure Modes and Effects Analysis (FMEA)
FMEA
222
Data Collection Plan– Operational Definition– Develop Measurement Plan– Data Collection– Data Display and evaluation of Data
Fundamentals of Minitab
Basic Statistics– Measures of Central Tendency– Measures of Dispersion– Probability Distribution
Measure Phase Topics
223
Measure Phase Topics
Gage R&R– Gage R&R for Continuous Data– Gage R&R for Attribute Data
Process Capability
Process Sigma Level Calculations
224
If we can’t accurately measure something
We don’t know enough about it
We can’t control it
We are at the mercy of chance!!!
Why to Measure ?
225
When you measure what you are speaking about and express in numbers, you know something about it.
Scientific Explanation : Very little progress is possible in any field of investigation without the ability to measure. The progress of measurement is in fact the progress of science !
Non Scientific Explanation : If you can not measure, just forget it ! It will be a sheer waste of time.
Without data you are just a loud mouth with an opinion ..
Science of Six Sigma
226
Data Collection Plan
Foundation of six sigma is Data based decision making, Data drives decisions and actions !!!
227
Data are measurements or observations we record and use to understand, characterize, optimize or control something such as process.
What is data
228
Knowledge is Power
Knowledge is not based on opinion, rather it is derived from facts & data.
In order to efficiently collect the data & effectively analyze it,to extract the maximum knowledge available, one must rely
On statistical techniques.
229
Use of Statistics
Data
Statistics convert
to
Usable Information
230
Decide objective
Step 1Operational definition
Step 2Develop Measurement Plan
Step 3Data Collection
Step 4 Data display
Evaluation of Data
• What, How, by Whom the measurement will be done
• Define a Metric
• Stick to procedure/plan
• By Plotting (Graphing) the Data, the result can be easily understood.
Data Collection Plan
231
An operational definition is a precise description of the specific criteria used for the measures (the what), the methodology to collect the data (the how), the amount of data to collect (how much) and who has the responsibility to measure the data
When developing an operational definition, it is important for the team to fully understand and agree that the DEFINITION reflects exactly what information the team is attempting to gather on the process.
Clarity is more important when developing and selecting the measures that will be used to determine the SIGMA PERFORMANCE of the process.
Operational Definition
232
Example :
Operational definitions may determine whether, a team is required to count all the defects on an invoice (required to calculate defects per million opportunities)
or
the total number of defective invoices (any invoice with any defect) or
the type of defects encountered on an invoice (to eliminate the most common defects first).
Each of these cases may require a very different approach for gathering the data.
Operational Definition
233
Operational definition provides the foundation for the team to
1. Reach an agreement on what data to be collected.
2. Build consistency and reliability into data collection.
3. Fully agree on how a particular characteristic of a process is to be measured.
Operational Definition
234
Poor Operational Definition: Cycle time of a transaction
Good Operational Definition: Collect data for all transactions processed from 1-Aug-05 to 31-Aug-05.
The cycle time of each transition will be determined by the date and time of transaction download from client server by an agent/CSR to the date and time of the PROCESSED transaction was submitted in client server as per the client server system time.
Example of Operational Definition.
235
Write operational definitions for the following cases
1) Maximization the server availability
2) Reduction of the attrition rate in ABZ
3) Improving the quality percentage a process
4) Reduction of call handling time
5) Minimization of abandoned calls in a call center
Exercise: Operational Definition
236
Develop Measurement Plan
Measurement PlanDetermining current process performance usually requires the collection of data. When developing a measurement plan ensure that:
– The data collected is meaningful– The data collected is valid– All relevant data is collected concurrently
What is the Purpose of Collecting the Data? Will it serve the purpose ? How will you collect the data?
-what result will you measure?-what kind of cause will you analyze for the
ineffective process?What kind of tool will be required?-form, check sheet ?
All related Data collected?-Sample size, frequency, sampling method?Is the Data Collecting method is adequate?-who will collect the data?-where can we collect the data?-when will we collect the data?-what kind of assistance will be necessary?
237
Before data collections starts, classify the data into different types: continuous or discrete.
This is important because it will:
– Provide a choice of data display and analysis tools
– Dictate sample size calculation
– Provide performance or cause information
– Determine the appropriate control chart to use
– Determine the appropriate method for calculation of Sigma
Data Classification
238
Continuous Data Discrete Data
Types of Data
Time (in hours) to process anapplication
% of applications with or without errors.
Number of errors in an application.
Customer satisfaction rating ofcall center service.
Measured on a continuum or scale
Binary : Classified into one of two categories
Count : Counted discretely
Ordered categories : Rankings or ratings
Description
Example
239
Continuous Data
Data generated by– Physically measuring the characteristic– Generally using an instrument– Assigning an unique value to each item
Examples:
1. The time it takes to write a proposal.
2. The time it takes to conduct a feasibility study.
3. The time it takes to close the books each month.
4. Invoice amounts.5. Sales order amounts.6. Handling Time, Time to Certify
PEs, etc.
Continuous Data: Example (Call Waiting Time in Secs)
SL No. Waiting Time
SL No. Waiting Time
1 98 11 102
2 103 12 98
3 100 13 101
4 100 14 101
5 99 15 99
6 101 16 100
7 97 17 101
8 102 18 99
9 100 19 100
10 99 20 102
240
Discrete Data
Data generated by
• Classifying the items into different groups based on some criteria
• All the items classified into a group will have same value
Examples:• Gender, Shade Variation, etc.
• Escalations, Repeat Calls, Defective Transactions, Defects in Transactions etc.
GONOGO
Good
Bad
241
Defects versus DefectiveOut of these 09 Invoices… there are...
3 Defective Invoices
6 Defects
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
242
Example of Opportunities
Errors in the rest ofthe invoice are not critical
Hence, there are 3 opportunities per invoiceeven though the invoice contains more than 3 line items.
Quantity: AAAAAPrice: $BBBBBDate: YY/YY/YY
Only three line items on thisinvoice are
critical to the customer.( Quantity, Price, Date)
243
Discrete : Binary Data (Binomial)
• Classifying the items into only two groups based on some criteria
• Each item will fall in either of the two groups
• All the items classified into a group will have same value
• Expressed or summarized as proportion p or percentage
Examples:
• Gender, Escalations, Repeat Calls, Defective Transactions, etc• An invoice is either “complete” or “incomplete”.• A delivery is either “late” or “not late”.• A product is either “damaged” or “not damaged”.• A hotel room is either “dirty” or “clean”.• A sales pitch is either a “thumbs up” or “thumbs down”.
244
Binary Data : Example (Month wise Escalation of Transactions)
Month No. of Transactions Processed No. of Transactions Escalated
Jan 2000 20
Feb 2500 30
Mar 1500 14
Apr 3000 27
May 4000 40
Jun 3500 33
Proportion of Escalated Transactions
p = No. of Transactions Escalated / Total No. of Transactions Processed
= (20 + 30 + 14 + 27 + 40 + 33) / (2000 + 2500 + 1500 + 3000 + 4000 + 3500) = 164 / 16500 = 0.0099 0.01 = 1 %
On an average, 1 % of Transactions are escalated
245
Discrete : Count Data (Poisson)
Data generated by
• Counting the exact number of occurrences of the characteristic in a group of items.
• It takes integer values as 0,1,2,-,-,-,
• Expressed or summarized as average number of occurrences
Examples:
• Number of fatal defects in transactions processed• Number of accidents in the city during June 2005• Number of suicides in the city during 2004.• The number of errors on twenty invoices.• The number of computer system failures in a month.
246
Count Data: Example (Data on Defects found during Transaction Audit)
No. of Items Audited
100 50 76 82 172 150 89
No. of Defects
3 4 0 1 5 10 1
Average Number of Defects = Total No. of Defects / Total No. Audited= (3 + 4 + 0 + 1 + 5 + 10 + 1) / (100 + 50 + 76 + 82 + 172 + 150 + 89 )= 0.033
On an average 0.033 defects found per TransactionsOn an average 3.3 defects found per 100 Transactions
247
Type of Data
Exercise 1.10 (20 minutes)
1. Refer to your workbook.
2. Identify the type of data given in example
248
Data Measurement Plan Format
Performance measure
Operational Definition Data Source & location
Sample size
Who will collect the data
Data collection period
How will date be collected
Other data that should be collected at the same time
Time to process a transaction
Date and time of transaction was download from client server by an agent to the date and time of the PROCESSED transaction was submitted in client server
Client server system time.
256 Raju
Smita
1-Aug-05 to 31-Aug-05
Random selection
Type of transaction, Day of week,
Agent name
The data is being collected to measure the performance is called PERFORMANCE DATA.
On the other hand, CAUSE DATA, focus on why the process performs as it does. Cause data supports the problem solving by helping to isolate root causes of the problems.
Most of the times, however, we won’t know enough about potential causes until we have determined our processes current performance level. Be prepared to document current performance first, then brainstorm potential causes and collect additional data related to those causes at a later date.
Cause Data
249
Data Collection
While collecting data ensure that the data measurement plan is followed. Note any deviations from the plan.
Avoid bias and ensure consistency.
Use various tools like check sheets to record and grouping of the data.
Ensure that the sample selected is representative of the population. If there is any concern on this issue, record the things that may cause the data collected to not be representative of the population.
Poor Information
Data Rich
Ensure Effective and Efficient Data Collection
250
Sampling
251
Sampling Objectives
• Understand the purpose and advantages of sampling
• Understand the application of different sampling techniques to ensure accurate process representation
• Gain experience in asking appropriate questions to ensure a robust sampling plan is implemented effectively and efficiently
• Understand guidelines and formulas used to determine sample size
252
Basic Definitions and Symbols
Population (N): The entire set of objects or activities for a process
μ: the mean (arithmetic average) calculated for a population
σ: the standard deviation calculated for a population
Sample (n): a group that is a part or subset of a population
x: the mean (arithmetic average) of a sample
s: the standard deviation of a sample
253
Sampling Definition
Sampling is the process of:Collecting only a portion of the data that is available or could be available, and drawing conclusions about the total population (statistical inference)
x x xx x x x
x x x x xx x x xx x xx x
x x
x xx x
x x
Population
N = 5000 n = 100
Sample
Example:Estimating the average height of students in a college by measuring the heights of only 250 students (250 is a subset of entire students population).
254
Sample ….. When ?
When to …..• Collecting all the data is impractical or too costly
• Data collection can be a destructive process
• When measuring a high-volume process
When not to ……• A subset of data may not accurately depict the process, leading to a
wrong conclusion (every unit is unique-e.g., structured deals)
255
Kinds of Sampling
Random Sampling• This sampling ensures that the
characteristics of the population are collected with equal possibility.
Stratified Sampling • Make stratifying plan for
population characteristics.• Select the sample among each
stratified group
Group A Group B
256
Frequency of Sampling
• Recommended more often for unstable processes
(Systematic, Subgroup sampling)
• Recommended less than usual for stable processes.
• To make a useful business decision we have to decide the
precision of data and frequency of data.
257
Sampling: Methodology
• Select a sample of items from the population
• Measure the characteristics on each item in the sample
• Calculate the sample statistics
• Provide the sample statistics as an estimate of population statistics
258
Methodology: Example
Select a sample of items from the population, say 250 students
Measure the characteristics on each item in the sample
i.e. measure the height of all the 250 students in the sample
Calculate the sample statistic
i.e Calculate the average height of 250 students ( = 5.5 feet)
Provide the sample statistic as an estimate of population statistics
Estimate of average height of students in the college = 5.5 feet
To estimate the average height of students in a college
259
Methodology: Issues
The following are the Waiting Times (Seconds) values of 36 Calls:
Mean Waiting Time = 28.55
The following data is a sample of 10 from the above data:
Sample Mean = 25.3
Sample Statistics may not be exactly equal to Population Statistics
10 30 30 25 34 15 52 12 30 24 20 50
50 60 40 35 25 30 20 10 40 10 10 20
40 26 34 17 20 50 37 24 16 10 40 32
30 34 17 10 24 12 16 50 50 10
260
The following data is another sample of 10 from the parent data:
Sample Mean = 27.8
The estimate may vary from sample to sample
To overcome these issues Confidence Intervals are developed
26 40 34 30 20 16 10 50 20 32
Methodology: Issues
261
Confidence Interval: Methodology
• Select a sample of items from the population
• Measure the characteristics on each item in the sample
• Calculate the sample statistics
• Provide two limits: an upper bound & a lower bound to the population statistics such that the true value of population statistics will lie within these limits with a specified level of confidence
262
Continuous Data: CI for Population Mean
• Select a sample of n items from the population
• Measure the characteristics on each item in the sample
• Calculate the sample Mean & Standard Deviation (SD). Then
(1-) % Confidence interval : Sample Mean Constant (confidence level) x Standard Error (SD of Sample Mean)
263
Continuous Data: CI for Population Mean
(1-) % Confidence interval on Mean:
Sample Mean Z /2 x SD / n
Z /2 is the Standard Normal variate for an area of /2 as shown in figure
0
1
2
3
4
-3 -2 -1 0 1 2 3Z /2
/2
264
Continuous Data: CI for Population Mean
CI Z /20.05 95 % 1.960.01 99 % 2.570.10 90 % 1.64
Obtained from Z table
265
CI for Population Mean: Example
The following data is a sample of 10 from the above data:
30 34 17 10 24 12 16 50 50 10
Sample Mean = 25.3
Sample SD = 15.34
The following are the Waiting Times (Seconds) values of 36 Calls:
10 30 30 25 34 15 52 12 30 24 20 50
50 60 40 35 25 30 20 10 40 10 10 20
40 26 34 17 20 50 37 24 16 10 40 32
Mean Waiting Time = 28.55
266
CI for Population Mean: Example
95 % Confidence interval on Mean:
Sample Mean 1.96 x SD / n
= 25.3 1.96 x 15.34 / 10
= 15.79 to 34.80
Similarly, 2nd Sample:
Sample Mean = 27.8
Sample SD = 11.94
95 % Confidence Interval on Mean:
27.8 1.96 x 11.94 / 10 = 20.39 to 35.20
26 40 34 30 20 16 10 50 20 32
267
1. Refer to your workbook.
2. Calculate CI for a given example
CI for Population Mean
Exercise 1.11 (20 minutes)
268
Discrete Data: CI for Proportion
Collect a sample of size n from the population
Calculate sample proportion p
Calculate Standard error (SE): (p(1-p)/n)
Then
(1 - ) % CI interval for Population Proportion:
p Z /2 x (p(1-p)/n)
95 % CI interval for Population Proportion:
p 1.96 x (p(1-p)/n)
269
1. Refer to your workbook.
2. Calculate CI for a given example
CI for Discrete data
Exercise 1.12 (20 minutes)
270
Sample Size Calculation: Continuous Data
Using 95 % CITrue value of population Mean will lie between
Sample Mean 1.96 SD / nThen Population Mean - Sample Mean < 1.96 SD / n ( with 95 % Confidence)
Hence To estimate the population mean with an accuracy of say 5I.e Population Mean - Sample Mean < 5I.e 5 = 1.96 SD / n n = (1.96 SD / 5)2
Sample Size required to estimate population mean with an accuracy of 5: (1.96 SD / 5)2
271
Sample Size Calculation: Methodology Continuous Data
Collect a small sample
Calculate Sample Mean & Standard Deviation
Equate accuracy required to 1.96 SD / n
Solve for n
272
1. Refer to your workbook.
2. Calculate sample size for a given example
Sample Size Calculation for continuous data
Exercise 1.13 (20 minutes)
273
Sample Size Calculation: Discrete Data
Using 95 % CITrue value of population proportion will lie between
p 1.96 x p(1-p) / n where p is sample proportionThen Population Proportion - Sample Proportion < 1.96 x p(1-p) / n
Hence To estimate the population proportion with an accuracy of say 0.01I.e Population proportion - Sample proportion < 0.01I.e 0.01 = 1.96 x p(1-p) / n n =1.962 p (1-p) / 0.012
Sample Size required to estimate population proportion with an accuracy of 0.1: 1.962 p (1-p) / 0.012
274
Sample Size Calculation: Methodology Discrete Data
Collect a small sample
Calculate Sample proportion p
Calculate Standard Deviation (p(1-p) / n)
Equate accuracy required to 1.96 (p(1-p) / n)
Solve for n
275
1. Refer to your workbook.
2. Calculate sample size for a given example
Sample Size Calculation for discrete data
Exercise 1.14 (40 minutes)
276
Data Display and Evaluation
Once we collect the data, it is always preferable to evaluate the data for its accuracy and usage prior to calculation of the capability of the process
As a initial step, display the data using Patero charts, Scatter plot, control charts, Histogram or Normality etc to look for data errors, trends and outliers.
Be prepare to collect more data or different data based on the above observations.
Evaluate the data to confirm that the data is dependable, consistent, reliable and representative.
Also ensure that, we get similar results if we repeat the data collection.
Finally confirm that the data collected provide the information we need.
277
Fundamentals of Minitab (Statistical Package)
278
Once you start Minitab, Minitab opens with two main windows.
Session WindowIt displays the results of your analysis in text format.
Data WindowIt contains an open worksheet, which is similar in appearance to a spread sheet. We can open multiple work sheets.
Column
Row
Cell
279
Column NameTo be written by us
Row Number
Column with Text data
Column with Numeric data
Column with date/time data
All the columns are formatted by default to Numeric data. As per the requirement we can reformat the columns.Right click mouse>Format Column>numeric/text/date.
280
FileMost of the functions of the File Menu are similar to Excel sheet.
Open existing or new filesSave filePrint files, etc.
EditMost of the functions of the File Menu are similar to Excel sheet.
Cuts, Paste cells,Undo, redoClear cells etc.
281
DataIt is very useful function menu in Minitab. to immunize the duplicate data entry in the work sheets.
Using this menu, we can subset the worksheets, split and merge work sheets.Minitab automatically opens multiple data windows.
We can transpose and sort the columns.
Group of the points from the graphs can be selected and corresponding data subset can be stored in separate work sheet.
282
CalcCalculator: Data in various columns can be computed and stored in separate column
Column and Row statistics:Various statistics like mean, SD etc can be calculated for the data listed in the column / row
Random data: Can be generated for all the distributions
Probability Distribution: All the distribution statistics can be calculated.
283
StatAll sorts of statistical analysis can be done for the data stored in various columns in the work sheet.
GraphsVarious graphs can be plotted using this menu
EditorUsed for Formatting of the columns
284
Basic Statistics
285
1. Measures of Central Tendency
1. Mean 2. Median 3. Mode
2. Measures of Dispersion
1. Range 2. Variance 3. Standard Deviation
Describe Sets of Continuous Data
The following three characteristics can describe the continuous data set
3. Shape
1. Histogram
286
Continuous Data: Measures of Central Tendency
1. Mean
2. Median
3. Mode
287
Mean: • Numerical value indicating the central value of data• Sum of all observations / Total number of observations
Suppose x1, x2, - - - xn be the data, thenMean = (x1+ x2 + - - -+ xn ) / n = xi /n
Continuous Data : Example Call Waiting Time
Mean: Sum of all observations / Total number of observations
= (98 + 103 + 100 + 100 + 99 + 101 + 97 + 102 + 100 + 99 + 102 + 98 + 101 + 1.01 + 99 + 100 + 101 + 99 + 100 + 102) / 20
= 200.2 / 20 = 100.1 Minutes
288
Median: • Middle Value
• Value which divides observations arranged in ascending or descending order into two equal halves
Case 1: Total number of observations is odd
Median: Middle ValueCase 2: Total number of observations is even
Median: Average of two middle values
Median: Example Call Waiting Time in Minutes
97 98 98 99 99 99 99 100 100 100
100 100 101 101 101 101 102 102 102 103
Total Number of observations: 20 (even)
The middle Values : 100 & 100 (10th value and 11th value)
Median: Average of 2 middle values = (100 + 100) / 2 = 100
289
Mode:
• The observation which occurs maximum number of times in the data
Example Call Waiting Time in Minutes
97 98 98 99 99 99 99 100 100 100
100 100 101 101 101 101 102 102 102 103
Total Number of observations: 20 (even)
The observation with maximum number of occurrences : 100
Mode: 100
290
Continuous Data :Measures of Dispersion
1.Range
2.Variance
3.Standard Deviation
291
Range: Definition
Range: Maximum value – Minimum Value
Example:
5 4 7 3 2
15 9 8 5 2
Maximum Value = 15Minimum Value = 2Range = 15 – 2 = 13
292
0
2
4
6
8
10
12
14
16
1 2 3 4 5 6 7 8 9 10
Range
Better measure of Dispersion is “Standard Deviation”
Range: Issues
It depends only on extreme valuesHence affected by outliers
293
Example :
5 4 7 3 2
15 9 8 5 2
Step 1:
Calculate Mean = (5+4+7+3+2+15+9+8+5+2) / 10
Mean = 6
Standard Deviation: Definition
Square root of the average squared deviation from mean
Indicates On an average how much each value is away from the Mean
294
Step 2: Take deviations from Mean
0
2
4
6
8
10
12
14
16
1 2 3 4 5 6 7 8 9 10
-1 -2 1 -3 -4
9 3 2 -1 -4
Example: Standard Deviation
295
Step 3:
Since some values are positive & rest are negative, while taking sum they will cancel out.
So square the values & Sum
1 4 1 9 16
81 9 4 1 16
Sum of Squares = 142
Example: Standard Deviation
296
Example: Standard Deviation5 4 7 3 2
15 9 8 5 2
Step 2 : Take deviations from Mean
-1 -2 1 -3 -4
9 3 2 -1 -4
Step 1: Calculate Mean, Mean = 6
Step 3: Since some values are positive & rest are negative, while taking sum they will cancel out. So square the values & Sum
1 4 1 9 16
81 9 4 1 16
Sum of Squares = 142
Step 4: Standard Deviation = (Sum of Squares / (n -1))
= (142 / (10 -1))
= 15.77 = 3.972
Variance = (SD)2 = 15.77
297
Standard Deviation: Example 2 Call Waiting Time in Minutes
Data (xi – Mean) (xi – Mean)2
x1 98 -1.9 3.61
x2 103 3.1 9.61
x3 100 0.1 0.01
x4 100 0.1 0.01
x5 99 -0.9 0.81
x6 101 1.1 1.21
x7 97 -2.9 8.41
x8 102 2.1 4.41
x9 100 0.1 0.01
x10 99 -0.9 0.81
Sum 999 28.9
n 10
Mean 99.9
S D = (28.9) / (10 – 1)
= 1.7919
298
Continuous Data: Graphical Representation of Data: Histogram
97 98 98 99 99 99 99 100 100 100
100 100 101 101 101 101 102 102 102 103
Example: Call Waiting Time Data
Total Number of observations : 20
Minimum Value : 97
Maximum Value : 103
Number of Classes : n = 20 = 4.47 5
Class Interval : (Maximum – Minimum) / Number of Classes
: (103 – 97) / 5 = 1.2
299
Construction of Frequency Table
Lower Limit of a Class
• 1st Class Lower Limit : Minimum Value• Lower Limit of any class other than 1st class : Upper Limit of Previous Class
Upper Limit of a Class
• Lower Limit of the Class + Class Interval
Construction of Frequency TableSL No Lower Limit Upper Limit Tally Marks Frequency
1 97 98.2 lll 3
2 98.2 99.4 llll 4
3 99.4 100.6 llll 5
4 100.6 101.8 llll 4
5 101.8 103 llll 4
300
Graphical Representation of Data: Histogram
0
1
2
3
4
5
6
98.2 99.4 100.6 101.8 103
301
1. Refer to your workbook.
2. Calculate Mean, Median, Standard Deviation and construct histogram for a given example
Mean, Median, Standard Deviation and histogram
Exercise 1.15 (40 minutes)
302
Probability & Normal Distribution
303
Ratio of number of favorable outcomes to total number of outcomes
Probability Definition
ExampleNumber of tosses of a coin = 100Number of times Head occurred = 49Number of times Tail occurred = 51
Probability of getting Head in a toss of coin=Number of Times Head occurred / Total number of tosses= 49/100 = 0.49 = 0.5 (Approximately)
304
Example 1 :The day wise average waiting time in seconds (AWT) of calls for 10 days is given below:
50 58 56 48 62 61 55 55 54 51
a. Calculate the probability that average waiting time > 60 Seconds?
b. Calculate the probability that average waiting time < 50 seconds
a. Probability of AWT > 60
= Number of cases with AWT > 60 / Total number of cases = 2 / 10 = 0.2
20 % of the days AWT will be more than 1 Minute
b. Probability of AWT < 50 seconds = Number of cases with AWT < 50 / Total number of cases = 1 / 10 = 0.1
10 % of the days AWT will be less than 50 seconds
305
Example 2 :The number of transactions processed per day by 40 member team for 12 days during transition is given below:
750 780 760 690 725 743 820 810 750 740 775 796 765 735
Risk of not meeting SLA = Probability that productivity < 720 transactions
= Number of days with Productivity < 720 / Total number of days
= 1 / 14 = 0.071
Risk of not meeting SLA is 7 %
Suppose the SLA on productivity is minimum 720 transactions per day, calculate the risk of not meeting the SLA?
306
Issues
To estimate probability using this method,
huge amount of data is required
Solution
When data is less, identify the underline distribution & estimate probability from the distribution
307
Statistical Distributions
Continuous distribution
•Normal distribution
Discrete distribution
•Binomial distribution
•Poisson Distribution
308
Normal DistributionDefinition: Consider the following data on Average Handling Time (AHT) in minutes of 16 Days:
2.3 2.7 2.4 2.6 2.3 2.7 2.5 2.5 2.5 2.4 2.5 2.6 2.2 2.8 2.4 2.6
Plot of the Data:
0
1
2
3
4
5
2.1 2.3 2.5 2.7 2.9
309
Plot of the Data:
0
1
2
3
2.2 2.3 2.4 2.5 2.6 2.7 2.8
• Bell Shaped• Symmetric• Total Area under the curve is 1
Then : Normal Curve & Data follows Normal Distribution
310
Normality Test : Probability Plot using MinitabStep1: Copy the data to Minitab worksheet column
Step 2: Choose Stat > Basic Statistics > Normality Test
311
Step3: Enter the Column Title to the Variable Text Box and Click OK button
312
Step 4: Minitab Output
Interpretation:
If P-Value ≥ 0.05, then Data is Normal
313
Standard Normal Distribution
If
Data follows Normal Distribution
then
(Data - Mean) / SD will follow Standard Normal Distribution
For Standard Normal Distribution:
Mean = 0
SD = 1
314
Standard Normal Distribution: Example
2.3 2.7 2.4 2.6 2.52.5 2.4 2.5 2.6
Data:
Mean = 2.5
SD = 0.1225
Z : (Data - Mean ) / SD
-1.633 1.633 -0.8165 0.8165 0.000.00 -0.8165 0.00 0.8165
Mean = 0.00
SD = 1.0
315
Standard Normal Distribution: Properties
0
1
2
3
4
-3 -2 -1 0 1 2 3
68.26%
95.46%
99.73%
Between
Mean 1 SD : 68.26 % of Values will lie
Mean 2 SD : 95.46 % of Values will lie
Mean 3 SD : 99.73 % of Values will lie
316
0
1
2
3
2.3 2.4 2.5 2.6 2.7
If data follows normal distribution, then the probabilities can be estimated from Normal Curve
Example:
The probability that AHT will be more than 2.6 Minutes is the area above 2.6 Minutes in Normal Curve
317
0
1
2
3
2.3 2.4 2.5 2.6 2.7
Example:
The probability that AHT will be less than 2.35 Minutes is the area below 2.35 Minutes in Normal Curve
318
Normal Distribution: Examples
The Time to Certify PE’s is normally distributed with mean 40 days and standard deviation 8 days. If the client wants that all PEs shall be certified within 34 to 48 days, estimate the chance of meeting client requirement?
Mean = 40
SD = 8
Let x be the Time to Certify PE
Case 1: Probability of certifying PEs within 34 days
P(x < 34)
Transforming to Standard Normal
P[((x - Mean) / SD ) < ((34 - 40)/8)] = P ( z < -0.75)
319
Normal Distribution: Examples
Case 1: P ( z < -0.75)
0
1
2
3
4
-3 -2 -1 0 1 2 3
From Standard Normal Tables P(z < -0.75) = 1 – 0.7733 = 0.2266
320
Case 2: Probability of certifying PE in > 48 days
P ( x > 48) = P (z > (48 - 40) / 8) = P ( z > 1)
0
1
2
3
4
-3 -2 -1 0 1 2 3
321
Normal Distribution: ExamplesCase 2: From Standard Normal Tables
P (z > 1 ) = 0.1587
Chance of Meeting Client Requirement = 1 – 0.2266 - 0.1587 = 0.6147 = 61.47 %
0
1
2
3
4
-3 -2 -1 0 1 2 3
Chance of Meeting Client Requirement = 0.7733 - 0.1587 = 0.6147 = 61.47 %
OR
322
1. Refer to your workbook.
2. Calculate probability distribution for a given example
Probability distribution for continuous data
Exercise 1.16.1 (20 minutes)
323
Handing Non normal data
Handling Non Normal Data
324
Binomial Distribution:If the data is binary, then probabilities are estimated using Binomial Distribution
325
Binomial Distribution: Example
On an average, 2 % of the transactions processed in a process are defective. On a particular day, out of 400 transactions audited 21 turned out to be defectives. Is it an indication that the process performance deteriorated?
Let p = 2 % = 0.02
Number of Transactions Audited (n) = 400
Number of Defectives (d) = 21
The probability of getting 21 defectives out of 400 when p = 2 % = 0.02 is calculated using Binomial Distribution as shown below
P(getting x = 21 defectives out of 400 transactions) = nCxpx(1-p)n-x
= 400C210.0221(1-0.02)400-21
326
Binomial Distribution: Example
Let p = 2 % = 0.02
Number of Transactions Audited (n) = 400
Number of Defectives (d) = 21
The probability of getting ≤ 20 defectives out of 400 when p = 2 % = 0.02 is calculated using Binomial Distribution as shown below
P ( x ≤ 20) = P ( x = 0 ) + P ( x = 1 ) + P ( x = 2 ) + - - - + P ( x = 20)
P ( getting x = 20 defectives out of 400 transactions ) = nCxpx(1-p)n-x
= 400C200.0220(1-0.02)400-20
327
Calculation of Binomial Probabilities using Minitab
Step 1:
Copy the defective data to Minitab Worksheet as shown below:
328
Calculation of Binomial Probabilities using Minitab
Step 2:
Go to Calc Probability Distributions Binomial
329
Calculation of Binomial Probabilities using MinitabStep 3:
Select Cumulative probability, Enter Number of trials, Probability of success, Input Column, Option Storage and click ‘OK’ button
330
Calculation of Binomial Probabilities using MinitabStep 4:
Minitab will calculate Binomial Probabilities as display in Optional Storage Column as shown below
Note: % Chance = Probability x 100
331
Binomial Distribution: Examplep = 2 % = 0.02
From Binomial Distribution,
Number Audited (n) Defectives (d) Chance of getting d or less defects (%)
400 0 0.03
400 2 1. 31
400 4 9.73
400 6 31.09
400 8 59.26
400 10 81.79
400 12 93.81
400 14 98.38
400 16 99.66
400 18 99.94
400 20 99.99
332
Binomial Distribution: Example
Let p = 2 % = 0.02
Number of Transactions Audited = 400
From Binomial Distribution,
Probability of getting less than 20 defectives in 400 transactions = 0.9999
Hence Probability of getting 20 or more defectives = 1 – 0.9999= 0.0001 0
i.e. if the process is operating at 2 % defectives:
the chance of getting 21 defectives out of 400 is almost 0 ,
Process performance is deteriorated.
333
1. Refer to your workbook.
2. Calculate probability distribution for a given example
Probability distribution for Binomial Distribution
Exercise 1.16.2 (20 minutes)
334
Poisson Distribution:If the data is Count, then probabilities are estimated using Poisson Distribution
335
Poisson Distribution: Example
The average number of repeat calls per day in a voice process is 20. On a particular day , there were 25 repeat calls. Is there any problem with the process that day?
Let : Average number of Repeat Calls = 20
The probability of getting x ≤ 24 calls when average number of repeat calls is 20 is calculated using Poisson distribution as follows
P ( x ≤ 24) = P ( x = 0) + P ( x=1)+ P ( x = 2) + - - - + P ( x=24)
P ( x = 24 when = 20) = e-x / x!
= e-202024 / 24!
336
Calculation of Poisson Probabilities using MinitabStep 1:
Copy the different values of repeat calls to Minitab worksheet as shown below:
337
Calculation of Poisson Probabilities using MinitabStep 2:
Go to Calc Probability Distributions Poisson
338
Calculation of Poisson Probabilities using MinitabStep 3:
Choose Cumulative probability, Enter Mean, Input column & Optional storage as shown below and click “OK” button.
339
Calculation of Poisson Probabilities using MinitabStep 4:
Minitab will display the probabilities in the Optional storage column as shown below
Note: % Chance = Probability x 100
340
Poisson Distribution: Example
Average number of repeat calls per day = 20
Average Repeat Calls
Repeat Calls (d)
Chance of getting d defects or less (%)
20 0 0.00
20 5 0.01
20 10 1.08
20 15 15.65
20 20 55.91
20 24 84.32
341
Poisson Distribution: Example
Average number of repeat calls per day = 20
From Poisson Distribution,
Probability of getting less than 25 repeat calls = 0.84
Hence Probability of getting 25 or more repeat calls = 1 - 0.84= 0.16 = 16 %
i.e. if the process is operating at 20 repeat calls per day:
the chance of getting 25 repeat calls is 16 %
16 % is large enough to conclude that there is nothing wrong in the Process.
342
1. Refer to your workbook.
2. Calculate probability distribution for a given example
Probability distribution for Poisson Distribution
Exercise 1.16.3 (20 minutes)
343
Gauge R&R(Measurement System Analysis)
344
Introduction
IdentifyY(CTQ)
Identify theProject
Define
Gage R&Rfor Y
Data Collection Baseline
MeasureWhen ?
BaselineProcess Capability
• Gage R&R is pre-requisite for data collection / analysis
• Gage R&R study is a method to evaluate measurement system to determine the amount of variation it contributes to the total observed process variation.
• In Manufacturing industries, gages are evaluated for repeatability (of readings when a component is measured multiple times) and Operators / inspectors are evaluated for reproducibility (of same readings when the component is measured by different operations).
• In service industries, Appraisers will be assessed instead of Gage
345
Importance of Gauge R&R
• It is a method to determine how good the data is• A Simple method to aid in improving the measurement system• A simple method to evaluate new gage / agents repeatability• A simple method to quantify measurement reproducibility
Examples
• A black belt wants to reduce the variability in transaction time
• A black belt wants to improve the quotation process
• A black belt wants to assess the process knowledge of the agents
346
What is Measurement ?
To give the value to express specific function of a certain material.
What is Measurement System ?
A given value is called measurement data.
All equipment and tools to get the measured data are termed gage.
Gage, operator, software, measurement method and process are
termed as measurement system.
Measurement System
347
How might measurementvariation affect these decisions?
Verify product/process
conformity to specifications
Verify product/process
conformity to specifications
Assist incontinuous
improvement activities
Assist incontinuous
improvement activities
What if the amount of measurement variation
is unknown
?
Process
Measurement
Process
Measurement
Measurement variation can make our process capabilities appear worse than they are.
Why Worry about Measurement Variation?
Consider the reasons why we measure:
348
Accounting For Changes
While we can come up with many explanations, they would fit into three general categories
• Simple Day-to-day Random Variation
• An Event That Changed the Distribution of Calls Coming in to Agents
• A Difference in How Calls Are Classified Between You and the People Classifying Calls Yesterday
Expected Variation
A Change to the Process
A Change to the Measurement System
How Can We Determine the Cause?
349
Sources of variation
Product Variability(Actual variability)
MeasurementVariability
Total Variability(Observed variability)
350
Long Term Process variation
Short TermProcess variation
Within Sample Variation
ActualPart to Part
Variation
MeasurementVariation
Observed Process Variation
Variation due to operators
Variation due to Gage
σ2Total = σ2
Part-Part + σ2R&R
To study & reduce the process variation the measurement variation has to be identified and separated from process
Repeatability Issue
Reproducibility Issue
351
Accuracy (Bias)
The difference between the observed average of measurements and
the true average of the items measured.
Observed
AverageTrue
Average
Accuracy
352
Repeatability
The variation due to the Gauge.The variation observed when the same Appraiser monitors/evaluates the same transaction repeatedly using same facilities / aids.
Master Value
Master Value
Mean Mean
Good Repeatability
Poor Repeatability
353
Reproducibility
Appraiser to Appraiser VariationThe variation observed when different Agents process the same transaction using the same facilities / aids.
Master Value
Master Value
Good Reproducibility
Operator 1
Poor Reproducibility
Operator 2 Operator 3Operator 1 Operator 2 Operator 3
354
Stability
The variation in the average of at least two sets of measurements obtained with a gage as a result of time on the same pieces.
Time 1 Time 2
Stability
355
Methods of performing Gage R&R Studies.
Continuous Data
Xbar-R Method
ANOVA Method
General use.It does not evaluate the interaction effect.
It evaluates the interaction effect of the agents also.More effective when extreme values are present
Discrete Data
Attribute Agreement Analysis
356
Exercise: Gage R & R– Continuous Data
Given the data below for reading by 3 appraisers on 6 calls with 2 trails, determine whether the measurement system is acceptable
Call ID Appraiser A Appraiser B Appraiser C
1 2 1 2 1 2
1 65 60 55 55 50 55
2 100 100 100 95 100 100
3 85 80 80 75 80 80
4 85 95 80 75 80 80
5 55 45 40 40 45 50
6 100 100 100 100 100 100
357
Gage R & R– Continuous Data
Step 1:
Copy the data to Minitab worksheet as shown below
358
Gage R & R– Continuous Data
Step 2:
Choose Gage R&R Study (Crossed) from Stat Menu as shown below:
359
Gage R & R– Continuous DataStep 3:
Enter Part Numbers, Operators & Measurement Data.
Choose Xbar and R as shown below
Click “OK” button
360
Gage R & R– Continuous Data
Step 4:
Minitab will give the following Output
Source Var Comp % Contribution
Total Gage R & R 17.434 4.06
Repeatability 7.338 1.71
Reproducibility 10.096 2.35
Part-To-Part 411.568 95.94
Total 429.002 100
Source StdDev (SD) (6 * SD) (%SV)
Total Gage R & R 4.1754 25.052 20.16
Repeatability 2.7088 16.253 13.06
Reproducibility 3.1774 19.065 15.34
Part-To-Part 20.2871 121.723 97.95
Total Variation 20.7154 124.274 100
If < 20 %. Gage acceptableElse if > 30 %, Gage not acceptableElse some problem with
gage, use with caution
361
Gage R & R– Continuous DataStep 5: Graphical Output 1
Perc
ent
Part-to-PartReprodRepeatGage R&R
100
80
60
40
20
0
% Contribution
% Study Var
Gage name:Date of study:
Reported by:Tolerance:Misc:
Components of Variation
Gage R&R (Xbar/ R) for Data
Graphical Representation of the first table in the previous slide
362
Gage R & R– Continuous DataStep 5: Graphical Output 2
Interpretation:
All points in R chart should be within the control limits for all Appraisers
Xbar chart for all appraisers should have more or less same pattern and most of the points should fall outside control limits.
Sam
ple
Range
10
5
0
_R=3.06
UCL=9.98
LCL=0
A B C
Sam
ple
Mean
100
80
60
40
__X=77.36UCL=83.11
LCL=71.61
A B C
Gage name:Date of study:
Reported by:Tolerance:Misc:
R Chart by Appraiser
Xbar Chart by Appraiser
Gage R&R (Xbar/ R) for Data
363
Gage R & R– Continuous Data
Step 5: Graphical Output 3
Interpretation:
All readings for each call is shown with their means connected. Ideally the variation around mean for different calls should be equal and minimum.
Call Id654321
100
90
80
70
60
50
40
Gage name:Date of study:
Reported by:Tolerance:Misc:
Data by Call Id
Gage R&R (Xbar/ R) for Data
364
Gage R & R– Continuous Data
Step 5: Graphical Output 4
Interpretation:
All readings for each appraiser is shown with their means connected. Ideally the variation around mean for different appraisers should be equal and minimum.
AppraiserCBA
100
90
80
70
60
50
40
Gage name:Date of study:
Reported by:Tolerance:Misc:
Data by Appraiser
Gage R&R (Xbar/ R) for Data
365
Gage R & R– Continuous DataStep 5: Graphical Output 5
Interpretation:
Ideally the lines should overlap or at least parallel. Large deviations from parallelism indicates lack of agreement among appraisers with respect different calls.
Call Id
Avera
ge
654321
100
90
80
70
60
50
40
Appraiser
AB
C
Gage name:Date of study:
Reported by:Tolerance:Misc:
Appraiser * Call Id Interaction
Gage R&R (Xbar/ R) for Data
366
Example: Gage R & R for Transition Cycle Time
A Team Lead in a finance related data process is responsible to monitor and control the cycle time a sub-process. This sub-process is well established and consumes almost equal amount of time to process each transaction.
Over period of time volumes are increased and agents are also increased proportionately. But he observed that currently, the variation in cycle time is very high and not meeting the SLA some times.
He wondered how it can happen? He is in doubt about the agents capability and likes to measure and assess the same.
He chosen 2 agents and 10 transactions. He has conducted Gage R&R study by processing each transaction twice by each agent. The transactions are selected on random basis for the processing.
Gage R&R for Continuous Data
367
Agent 1 Agent 2 Transaction 1 2 1 2
1 21 20 20 20 2 24 23 24 24 3 20 21 19 21 4 27 27 28 26 5 19 18 19 18 6 23 21 24 21 7 22 21 22 24 8 19 17 18 20 9 24 23 25 23
10 25 23 26 25
Summary of Case Study (Manual Calculations)10 Transactions, 2 AgentsEach Agent processed each transaction twice
Data Collection
368
Gage R & R: Example
Operator 1 Operator 2Part 1 2 Range 1 2 Range
1 21 20 1 20 20 02 24 23 1 24 24 03 20 21 1 19 21 24 27 27 0 28 26 25 19 18 1 19 18 16 23 21 2 24 21 37 22 21 1 22 24 28 19 17 2 18 20 29 24 23 1 25 23 2
10 25 23 2 26 25 1Rbar 1.2 Rbar 1.5
Repeatability: Variation due to measurement instrument
Variation occurs when same operator measures the same part again and again
369
Gage R & R: Example
Repeatability: Variation due to measurement instrument
Variation occurs when same operator measures the same part again and again
Variation due instrument : Average of Rbars
= 1/2(1.2 + 1.5) = 1.35
Repeatability (EV) = K1 x Average Rbar = 1.19681
Trails K1
2 0.8862
3 0.5908
370
Gage R & R: Example
Agent 1 Agent 2 Part 1 2 Mean 1 2 Mean
1 21 20 20.5 20 20 20.0 2 24 23 23.5 24 24 24.0 3 20 21 20.5 19 21 20.0 4 27 27 27.0 28 26 27.0 5 19 18 18.5 19 18 18.5 6 23 21 22.0 24 21 22.5 7 22 21 21.5 22 24 23.0 8 19 17 18.0 18 20 19.0 9 24 23 23.5 25 23 24.0
10 25 23 24.0 26 25 25.5 Mean 1 21.9 Mean 2 22.35
Reproducibility: Variation caused by operators
Variation occurs when same part is measured by different operators
371
Gage R & R: Example
Reproducibility: Variation caused by operators
Variation occurs when same part is measured by different operators
Overall Variation between operators : Difference between xbars
= (22.35 - 21.9) = 0.45
Reproducibility AV) = ((xbar diff x K2)2 – (EV2 / n r))
= ((0.45 x 0.7071)2 – (1.196812 / 10 x 2))
= 0.1739
n: Number of Parts
r: Number of trails
Operators K2
2 0.7071
3 0.5231
372
Gage R & R: Example
Total Gage R & R : Repeatability2 + Reproducibility2
= ( 1.196812 + 0.17392) = 1.2094
373
Gage R & R: Example
Agent 1 Agent 2 Part 1 2 1 2
Mean
1 21 20 20 20 20.25 2 24 23 24 24 23.75 3 20 21 19 21 20.25 4 27 27 28 26 27.00 5 19 18 19 18 16.00 6 23 21 24 21 17.25 7 22 21 22 24 17.25 8 19 17 18 20 18.50 9 24 23 25 23 23.75
10 25 23 26 25 24.75
Rp: Mean max – Mean min = 27.00 – 16.00 = 11
Part Variation:
374
Gage R & R: Example
Rp: Mean max – Mean min = 27.00 – 16.00
= 11
Part Variation (PV): K3 x Rp = 0.3146 x 11
= 3.4606
Part Variation:
Parts K3
2 0.7071
3 0.5231
4 0.4467
5 0.4030
6 0.3742
7 0.3534
8 0.3375
9 0.3249
10 0.314
375
Gage R & R: Example
Total Variation: Gage R &R2 + Part Variation2
Total Variation: 1.20942 + 3.46062 = 3.6658
Total Variation:
376
Gage R & R: Example
Summary Table:
Source SD 5.15 x SD % Study Var
Repeatability 1.19681 6.1636 32.64
Reproducibility 0.1739 0.8956 04.74
Total Gage R &R 1.2094 6.2284 32.99
Part Variation 3.4606 17.8221 94.40
Total Variation 3.6658 18.8791 100
% Study variation is 32.99% > 30% hence variation in processing time is not acceptable.
Reasons shall be investigated and improvement plan shall put in place.
377
Agent 1 Agent 2 Transaction 1 2 1 2
1 21 20 20 20 2 24 23 24 24 3 20 21 19 21 4 27 27 28 26 5 19 18 19 18 6 23 21 24 21 7 22 21 22 24 8 19 17 18 20 9 24 23 25 23
10 25 23 26 25
Summary of Case Study (Using Minitab)10 Transactions, 2 AgentsEach Agent processed each transaction twice
Data Collection
378
Enter the data in Minitab as shown
Select Stat>Quality Tools> Gage Study> Gage R&R (Crossed)
Click on columns as shown
Choose either ANOVA or Xbar&R. It is preferable to chose ANOVA as it also analyses the interaction effect
379
% Study variation is 35.28% > 30% hence variation in processing time is not acceptable.
Reasons shall be investigated and improvement plan shall put in place.
380
Total Gage R&R. Focus only on Green Bars. These represents the % of total variation contributed from the data.The Gage R&R should be only 10% of total variation. Rest should be attributed to within transactions Variation. 35.28% is not acceptable
Within Agent (Repeatability)
Agent to Agent (Reproducibility)
Part-to-part variation (transaction to transaction) (estimate of process variation)
Represents the repeatability. Presence of of assignable causes (point out of control point) indicates stability problem. Excessive common cause variation to be addressed
Represents the reproducibility. The detectable shift in the pattern on X-bar chart and inconsistent pattern are unwarranted.
Remember : Most of the points should fall outside control limits.
381
This graph shows the data for eachAgent for all the Transactions.
Represents the Bias. Ideally, all the lines should overlap each other
This graph shows the data for the 10 transactions for each Agent. It display the raw data and
highlights the average of those measurements
Similar to top graph but the data is presented by Agent instead of Transaction. The graph will help
identify Agent issues
382
Gage R & R– Continuous DataExercise 1.17 (20 minutes)
1. Refer to your workbook.
2. Calculate Gauge R & R for a given example
383
An Engagement Team Lead (TL) considered 10 transaction and chosen 2 appraisers at random for Gage R&R study. The transactions were evaluated on “Correct” or “Incorrect” basis. For all the 10 transactions actual results (Standard) are also available with TL. 2 appraisers processed each transaction twice within gap of one week. The results are as follows. Study the Gage R&R.
TRANSACTION NUMBER
STANDARD RADHA KRISHNA
TRIAL 1 TRIAL 2 TRIAL 1 TRIAL 2
1 CORRECT CORRECT CORRECT CORRECT INCORRECT
2 INCORRECT CORRECT INCORRECT INCORRECT INCORRECT
3 INCORRECT INCORRECT INCORRECT INCORRECT INCORRECT
4 CORRECT INCORRECT INCORRECT CORRECT CORRECT
5 INCORRECT INCORRECT INCORRECT INCORRECT CORRECT
6 CORRECT CORRECT CORRECT CORRECT CORRECT
7 CORRECT CORRECT CORRECT CORRECT CORRECT
8 CORRECT INCORRECT CORRECT CORRECT CORRECT
9 INCORRECT INCORRECT CORRECT INCORRECT INCORRECT
10 CORRECT CORRECT CORRECT CORRECT CORRECT
Example: Gage R & R for evaluate appraiser process knowledge
Gage R&R for Discrete Data
384
Enter the data in Minitab worksheet.Test1 and test 2 results of same appraiser should be at one place, as shown
Note: Minitab 13 is used. Commands are similar to Minitab 14
385
Select Stat > Quality Tools > Attribute Agreement Analysis
Analyze Results
386
Analyze Results
Click on columns as shown
Enter 2, 2 , Radha, Krishna
Click on columns as shown
387
Percent Repeatable by appraiser(it should be >=80%)
388
Repeatability Vs Standard
389
Percentage ReproducibilityFor all appraisers(it should be >=80%)
Percentage ReproducibilityFor all appraisers Vs Standard(it should be >=80%)
390
KRISHNARADHA
100
90
80
70
60
50
40
30
Appraiser
Per
cent
Within Appraiser
KRISHNARADHA
100
90
80
70
60
50
40
30
Appraiser
Per
cent
Appraiser vs Standard
Assessment AgreementDate of study:Reported by:Name of product:Misc:
[ , ] 95.0% CI
Percent
Pictorial Representation
Since R&R is less than 80%, root causes to be identified and corrective actions to be taken. Re-conduct study to assess the improvement.
391
Case 1: Continuous Data
If Total gage R &R %SV < 20 % , Measurement system is acceptable
If Total gage R &R %SV between 20 % to 30 %
Some problem with measurement system, use with caution
If Total gage R &R %SV between > 30 %
Measurement system is unacceptable.
Rules for Gauge R&R study conclusions
Case 2: Discrete Data
If Gage R &R (Agreement) > 80 %
Measurement system is acceptable
Else, Measurement system is unacceptable.
392
1. Refer to your workbook.
2. Calculate Gauge R & R for a given example
Gage R & R– Attribute (Discrete) DataExercise 1.18 (20 minutes)
393
Process Capability
394
Common Process Capability Indices1. Potential capability Cp2. Achieved capability Cpk
Process CapabilityRefers to the inherent or natural variation of a process
Process Capability Cpk
A methodology to check whether the process have the capability to meet the customer requirements
Customer requirements are also expressed asLower Specification Limit (LSL) = 50 DaysUpper Specification Limit (USL) = 60 Days
Process Capability
395
Potential capability Cp: A measure of the ability/potential to meet the customer specifications
Example 1: Specification: 55 ± 5 DaysAllowed variation = 50 Days to 60 DaysNatural Variation = 52 Days to 58 Days
Natural Variation < Allowed variation
Hence Process have the capability to satisfy customer
Example 2 : Specification: 55 ± 5 DaysAllowed variation = 50 Days to 60 Days Natural Variation = 48 Days to 62 Days
Natural Variation > Allowed Variation
Then Process doesn’t have the capability to satisfy customer
396
Potential capability Cp:
If the data is normally distributed, then
Natural variation : Mean ± 3 SD
Example:
Mean = 55 Days & SD = 1 Day
Natural Variation = 55 – (3 x 1) to 55 + (3 x 1)
= 52 Days to 58 Days
397
Potential capability Cp:
Ratio of allowed variation to Total variation
Cp = Allowed variation / Natural variation
= (USL – LSL) / ((Mean + 3 SD) – (Mean - 3 SD))
= (USL – LSL) / 6 SD
A Process has the capability to meet customer requirements if
Allowed variation > Natural variation
(USL – LSL) > 6 SD Cp > 1
Preferably Cp should be greater than 1.34
398
Potential capability Cp: Example
The Time to Certify Agents in days is given in the table below. If the client requirement on Time to Certify Agents is 50 to 90 days, check whether the process has the capability to meet the client requirement ?
85 75 80 65 75 60 80 70 75 60
80 75 70 70 75 75 85 60 50 65
USL = 90 Days
LSL = 50 Days
Mean = 71.5
SD = 9.2
Cp = (USL – LSL) / 6 SD
= (90 – 50) / (6 x 9.2)
= 40 / 55.2 = 0.72
Conclusion ?
399
Potential capability Cp: Issues
• Cp checks only whether the process has the potential to meet the requirements
• Cp never checks whether the Process is actually meeting requirements
400
Potential capability Cp: Issues
Example:
Process: Training Process Characteristic: Time to Certify Agents
Specification : 55 ± 5 Days
Process 1 Process 2 Process 3
Mean 55 52 58
SD 1 1 1
USL – LSL 10 10 10
6 SD 6 6 6
Cp 1.66 1.66 1.66
401
Potential capability Cp: Issues
Process 1 Process 2 Process 3
Cp 1.66 1.66 1.66
Allowed variation 50 to 60 50 to 60 50 to 60
Total process variation 52 to 58 49 to 55 55 to 61
Cp = 1.66 for all 3 processes
all 3 process have the capability to meet customer requirement
But
only Process 1 is meeting customer requirement
Hence
Achieved Capability Index is developed
Example:Process: Training Process Characteristic: Time to Certify Agents Specification : 55 ± 5 Days
402
Achieved Capability Index Cpk:
Cpk = Min [Cpl, Cpu]
Cpl = (Mean – LSL) / 3 SD
Cpu = (USL - Mean) / 3 SD
Cpk checks whether the process is centered.
403
Achieved Capability Index Cpk: Graphical Representation
0
0.2
0.4
0.6
0.8
1
1.2
1 2 3 4 5 6 7
USLLSL
Mean + 3 SD 3 SD
ba
c d
Cpl = a / c
= (Mean – LSL ) / 3 SD
Cpu = b / d
= (USL - Mean ) / 3 SD
404
Achieved Capability Index Cpk: Example
0
0.2
0.4
0.6
0.8
1
1.2
6 7 8 9 10 11 12
126
Mean + 3 SD- 3 SD
33
3 3
Example:
USL : 12 LSL: 6
Mean : 9 SD : 1
Cpu = 3 / 3 = 1
Cpl = 3 / 3 = 1
Cpk = Min [1 , 1] = 1
Cpk = 1
405
Achieved Capability Index Cpk: Example
0
0.2
0.4
0.6
0.8
1
1.2
5 6 7 8 9 10 11
126
Mean + 3 SD- 3 SD
42
3 3
Example:
USL : 12 LSL: 6
Mean : 8 SD : 1
Cpu = 4 / 3 = 1.33
Cpl = 2 / 3 = 0.66
Cpk = Min [1.33, 0.66] = 0.66
Cpk < 1, Process doesn’t meet the customer requirements.
406
Achieved Capability Index Cpk: Example
0
0.2
0.4
0.6
0.8
1
1.2
6 7 8 9 10 11 12
126
Mean + 3 SD- 3 SD
33
3 3
Conclusion:
Cpu = 3 / 3 = 1
Cpl = 3 / 3 = 1
Cpk = Min [1 , 1] = 1
Cp = (USL – LSL) / 6 SD = 6 /6 = 1
When process Mean is at center of Specification then
Cpk =Cp
407
Potential capability Cp: Issues
Example:
Process: Training Process Characteristic: Time to Certify Agents
Specification : 55 ± 5 Days
Process 1 Process 2 Process 3
Mean 55 52 58
SD 1 1 1
USL – LSL 10 10 10
6 SD 6 6 6
Cp 1.66 1.66 1.66
Cpk 1.66 0.67 0.67
408
Relationship of Cp and Cpk
Cpk = 2
Cpk = 1.5
Cpk = 1
Cpk = 0
Cpk = -0.5
Cp =2
Cp =2
Cp =2
Cp =2
Cp =2
USLLSL
Mean
409
1. Refer to your workbook.
2. Calculate process capability for a given example
Process CapabilityExercise 1.19 (20 minutes)
410
Process Sigma Level Calculations
411
General Guidelines
1. Ensure that the data collected is free from measurement error
2. Ensure that the data is true representation of the population
3. Ensure that the process is stable
4. Ensure that the data is following Normal distribution.
5. If data is not normal, check for the transformed functions.
6. Still, data is non-normal, recheck the data or check whether the data fit in any non-normal distribution.
Process Capability for Continuous Data
412
Z Calculation for Normal Distribution Data
If data is normal follow the following sequence.
• Calculate Ppk using Minitab.
• Sigma multiple Long term of the process is = Zlt = (3 * Ppk)
• Sigma multiple Short term of the process is = Zst = (3 * Ppk)+1.5
• Note the DPMO also from Minitab output.
Process Capability for Continuous Data
413
Data Transformation
Typically one sided specification processes have skewed data.In such cases we transform the data points and specifications to convert data into a normal data.
Transformed data for Y may be in the form of Y2, Y3, Y0.5, Ln(Y) etc.
We can also use the Box-Cox transformation available in Minitab or apply instead of try out all the combinations.
Transformation used for Y is also applied to LSL and USL
However, test of normality should be done again to check if the transformed data has now become normal
414
Process Capability for Continuous Data Example 1
The cycle time (in Minutes) of each transaction in a day on both the shifts was collected. The SLA for the cycle time of the process is 60min.Calculate the process capability.
Enter the data in Minitab
Cycle Time Shift Cycle Time Shift50 1 60 251 1 55 250 1 49 255 1 53 256 1 46 252 1 51 248 1 50 252 1 41 251 1 51 249 1 58 252 1 54 250 1 57 256 1 50 252 1 49 252 1 41 2
415
Select Stat>Control Charts> Variable charts for Individuals> I-MR chart
Click Cycle time
Click OK
Stability Test
416
Observation
Indiv
idual V
alu
e
28252219161310741
60
55
50
45
40
_X=51.37
UCL=62.65
LCL=40.09
Observation
Movin
g R
ange
28252219161310741
15
10
5
0
__MR=4.24
UCL=13.86
LCL=0
I-MR Chart of Cycle Time_1
All the points in the I-MR control chart are within control limits. No significant trends also.
So, it is a stable process.
Note:If there are any out of control point, the related data to be analyzed and eliminate the assignable cause. Then, remove that data point and continue.
Stability Test
417
Cycle Time_1
Perc
ent
6055504540
99
95
90
80
70
605040
30
20
10
5
1
Mean
0.062
51.37StDev 4.206N 30AD 0.696P-Value
Probability Plot of Cycle TimeNormal
P-Value is great than 0.05.So, it is a normal Distribution.
Since the data is satisfying all the pre-requisites, we can calculate the process capability for this process.
Select Stat>Basic Statistics> Normality Test
Click ‘cycle time’ in ‘Variable’ field
Normality Test
418
Click Cycle time
Enter 60 in Upper Spec.
Enter 1
Click OK
Select Stat>Quality Tools>Capability Analysis>Normal
Capability Test
419
Select Stat>Quality Tools>Capability Analysis
6055504540
USLProcess Data
Sample N 30StDev(Within) 3.76009StDev(Overall) 4.24218
LSL *Target *USL 60Sample Mean 51.3667
Potential (Within) Capability
Overall Capability
Pp *PPL *PPU 0.68Ppk 0.68Cpm
Cp
*
*CPL *CPU 0.77Cpk 0.77
Observed PerformancePPM < LSL *PPM > USL 0.00PPM Total 0.00
Exp. Within PerformancePPM < LSL *PPM > USL 10836.64PPM Total 10836.64
Exp. Overall PerformancePPM < LSL *PPM > USL 20919.46PPM Total 20919.46
WithinOverall
Process Capability of Cycle Time
Ppk =0.68Zlt = (3 * Ppk) = 2.04Zst = (3 * Ppk)+1.5 = 3.45DPMO = 20919.46
Capability Test
420
Process Capability for Continuous Data Example 2
HR is working on reduction of recruitment cycle time. 30 data points are collected to set the baseline capability, as shown below. The SLA is 60 days. Calculate the baseline capability of the process.
Recruitment Cycle time Recruitment Cycle time50 5551 5550 5855 5356 4652 5148 5055 4151 5150 5852 5450 5056 5052 4952 40
Enter the data in Minitab
421
Select Stat>Control Charts> Variable charts for Individuals> I-MR chart
Click Recruitment Cycle time
Click OK
Stability Test
422
Observation
Indiv
idual V
alu
e
28252219161310741
60
55
50
45
40
_X=51.37
UCL=61.45
LCL=41.28
Observation
Movin
g R
ange
28252219161310741
12
9
6
3
0
__MR=3.79
UCL=12.39
LCL=0
11
I-MR Chart of Recruitment Cycle time
Some of the points are crossing control limits. No significant trends also.
So, it is a not a stable process.
Since there are some out of control points, the related data to be analyzed and eliminate the assignable cause. Then, remove that data point and continue.
Let us continue with the same data now.
Stability Test
423
P-Value is less than 0.05.So, it is a non-normal Distribution.
So look for the transformation.
Select Stat>Basic Statistics> Normality Test
Click ‘Rec. cycle time’ in ‘Variable’ field
Recruitment Cycle time
Perc
ent
6055504540
99
95
90
80
70
605040
30
20
10
5
1
Mean
0.018
51.37StDev 4.115N 30AD 0.911P-Value
Probability Plot of Recruitment Cycle timeNormal
Normality Test
424
Select Stat>Control Charts>Box-Cox Transformation
Click Recruitment Cycle time
Enter 1
Click on Options
Ensure dot on top option
Enter C2It store the transformed data in C2 column
Click Ok
Data Transformation
425
Lambda
StD
ev
5.02.50.0-2.5-5.0
4.2
4.0
3.8
3.6
3.4
3.2
Lower CL
Limit
Lambda
3.03
(using 95.0% confidence)
Estimate 3.03
Lower CL -1.26Upper CL *
Rounded Value
Box-Cox Plot of Recruitment Cycle time
Lambda Value = 3.03
Data Transformation
426
Select Stat>Basic Statistics> Normality Test
Click ‘Box-Cox’ in ‘Variable’ field
Box-Cox
Perc
ent
25000020000015000010000050000
99
95
90
80
70
605040
30
20
10
5
1
Mean
0.071
155934StDev 35099N 30AD 0.673P-Value
Probability Plot of Box-CoxNormal
P-Value is greater than 0.05.So, The transformed data is following normal Distribution.
Now let us calculate the process capability
Data Transformation
427
Select Stat>Quality Tools>Capability Analysis>Normal
Click Rec. cycle time
Enter 60 in Upper Spec.
Enter 1
Click on Box-Cox
Click Box-Cox button
Click others
Enter Lambda value 3.03
Capability Test
428
22500020000017500015000012500010000075000
USL*
transformed dataProcess Data
Sample N 30StDev(Within) 3.36268StDev(Overall) 4.15021
After Transformation
LSL* *Target*
LSL
*USL* 244230Sample Mean* 155371StDev(Within)* 29657.1StDev(Overall)* 35265.1
*Target *USL 60Sample Mean 51.3667
Potential (Within) Capability
Overall Capability
Pp *PPL *PPU 0.84Ppk 0.84Cpm
Cp
*
*CPL *CPU 1.00Cpk 1.00
Observed PerformancePPM < LSL *PPM > USL 0.00PPM Total 0.00
Exp. Within PerformancePPM < LSL* *PPM > USL* 1366.91PPM Total 1366.91
Exp. Overall PerformancePPM < LSL* *PPM > USL* 5872.39PPM Total 5872.39
WithinOverall
Process Capability of Recruitment Cycle timeUsing Box-Cox Transformation With Lambda = 3
Select Stat>Quality Tools>Capability Analysis
Ppk =0.84Zlt = (3 * Ppk) = 2.52Zst = (3 * Ppk)+1.5 = 4.02DPMO = 5872
Capability Test
429
We can use Capability Analysis (Binomial) if the data meet the following conditions.
• Each item is the result of identical conditions.
• Each item can result in one of two possible outcomes (success/failure, Go/No go)
• The probability of success (or failure) is constant for each item.
• The outcomes of the items are independent of each other.
Process Capability for Attribute Data
430
Process Capability for Attribute Data - Example
A Black belt is working on project to reduce the defective transaction in a process. He has collected data for 30 days as shown below.
Calculate the baseline Capability of the process.
Sl No Trasactions / day Defective Transations Sl No Trasactions / day Defective Transations1 52 9 16 45 142 55 11 17 53 133 51 16 18 55 124 47 11 19 49 105 50 9 20 51 76 49 11 21 46 87 54 9 22 55 98 48 13 23 50 89 53 11 24 53 1110 50 7 25 55 1011 45 10 26 45 1212 50 6 27 49 1213 52 11 28 53 1214 47 7 29 48 915 45 10 30 50 17
Enter the this data in Minitab
431
Select Stat>Quality Tools >Capability Analysis>Binomial
Click on ‘Defective Transactions
Click on ‘No. of Transactions
Target, by default 0.Value can be entered if there is any target.
Click OK
Capability Test
432
Sample
Pro
port
ion
28252219161310741
0.4
0.2
0.0
_P=0.2093
UCL=0.3819
LCL=0.0367
Sample
%D
efe
ctiv
e30252015105
22.8
21.6
20.4
19.2
18.0
Summary Stats
0.00PPM Def: 209302Lower CI : 189000
Upper CI : 230745Process Z: 0.8088Lower CI :
(using 95.0% confidence)
0.7364Upper CI : 0.8816
% Defective: 20.93Lower CI : 18.90Upper CI : 23.07Target:
Sample Size
%D
efe
ctiv
e
555045
30
20
10
35302520151050
6.0
4.5
3.0
1.5
0.0
Tar
Binomial Process Capability Analysis of Defective TransationsP Chart
Tests performed with unequal sample sizes
Cumulative % Defective
Rate of Defectives
Dist of % Defective
The P-Chart Verifies that the process is in a state of control.
In this case there is no out of control point.
The proportion defective is 20.93%.
Cumulative % defective is the running average of the percentage defective.
It verifies that you have collected data from enough samples to have a stable defective estimate.
The rate appears to be stabilizing around 21%
Capability Test
433
Defective rate plot verifies that the % defective is not influenced by the number of items sampled.
Data should appear randomly distributed.
Histogram of % defective displays the over all distribution of the % defectives from the samples collected
Sample
Pro
port
ion
28252219161310741
0.4
0.2
0.0
_P=0.2093
UCL=0.3819
LCL=0.0367
Sample
%D
efe
ctiv
e
30252015105
22.8
21.6
20.4
19.2
18.0
Summary Stats
0.00PPM Def: 209302Lower CI : 189000
Upper CI : 230745Process Z: 0.8088Lower CI :
(using 95.0% confidence)
0.7364Upper CI : 0.8816
% Defective: 20.93Lower CI : 18.90Upper CI : 23.07Target:
Sample Size
%D
efe
ctiv
e
555045
30
20
10
35302520151050
6.0
4.5
3.0
1.5
0.0
Tar
Binomial Process Capability Analysis of Defective TransationsP Chart
Tests performed with unequal sample sizes
Cumulative % Defective
Rate of Defectives
Dist of % Defective
Capability Test
434
Sample
Pro
port
ion
28252219161310741
0.4
0.2
0.0
_P=0.2093
UCL=0.3819
LCL=0.0367
Sample
%D
efe
ctiv
e
30252015105
22.8
21.6
20.4
19.2
18.0
Summary Stats
0.00PPM Def: 209302Lower CI: 189000Upper CI: 230745Process Z: 0.8088Lower CI:
(using 95.0% confidence)
0.7364Upper CI: 0.8816
%Defective: 20.93Lower CI: 18.90Upper CI: 23.07Target:
Sample Size
%D
efe
ctiv
e
555045
30
20
10
35302520151050
6.0
4.5
3.0
1.5
0.0
Tar
Binomial Process Capability Analysis of Defective TransationsP Chart
Tests performed with unequal sample sizes
Cumulative % Defective
Rate of Defectives
Dist of % Defective
Results:P-chart indicates that process is stable as there are no data points out of controlThe chart of cumulative % defective show that the estimate of the overall defective rate appears to be settling down around 21%.
Sample
Pro
port
ion
28252219161310741
0.4
0.2
0.0
_P=0.2093
UCL=0.3819
LCL=0.0367
Sample
%D
efe
ctiv
e30252015105
22.8
21.6
20.4
19.2
18.0
Summary Stats
0.00PPM Def: 209302Lower CI: 189000Upper CI: 230745Process Z: 0.8088Lower CI:
(using 95.0% confidence)
0.7364Upper CI: 0.8816
%Defective: 20.93Lower CI: 18.90Upper CI: 23.07Target:
Sample Size
%D
efe
ctiv
e
555045
30
20
10
35302520151050
6.0
4.5
3.0
1.5
0.0
Tar
Binomial Process Capability Analysis of Defective TransationsP Chart
Tests performed with unequal sample sizes
Cumulative % Defective
Rate of Defectives
Dist of % Defective
The process Z is around 0.8, which is very poor. This process could use a lot of improvement
Capability Test
435
Terminology
Unit:A unit is the tangible & measurable characteristic of a process input / output.Defects are observed / counted in the output characteristic of a unit (Denoted as Y)
Examples:Every Call received by a call center Agent: Unit= CallEach employee recruitment cycle time Unit = EmployeeEach transaction processed by agent Unit = TransactionThe transaction not meeting the customer requirement Unit: TransactionNon-availability of system Unit = System
Defect:• A defect is a failure to conform to requirements • Any type of undesired result is a defect.
A failure to meet one of the acceptance criteria of a customer. • A defective unit may have one or more defects.
Process Sigma Multiple for Discrete Data
436
Specification Vs CharacteristicSpecification is a customer-defined tolerance for the output unit value.There may be two sided specifications.Specification form the basis of any defect measurement exercise on continuous data
A characteristic is a customer-defined expectation on the output unit. Characteristic from the basis of any defect measurement exercise on discrete data There may be multiple characteristics defined on a single unit. It is also possible to have a combination of specifications and characteristic on an single unit
Specification : Continuous Data Characteristic : Discrete Data
Example: Transaction processing
Unit: Each transaction processedSome of the defect definitions may be
1) Transaction not completed before 24min = Specification 2) Transactions not submitted in to client server after processing = Characteristic
3) Transactions submitted with out filling up the amount = Characteristic
437
Opportunity for Defect:
Any critical characteristic which is routinely inspected before passing the item is an opportunity for defect.(or)Opportunity for the error in a process is the number of steps / task / actions in the process, where there is a possibility of committing error, that may result in a defect.Concept of OFD is applicable only when defect measurement is discrete.
Recollect the operational definition >>>>>>“Clarity is more important when developing and selecting the measures that will be used to determine the SIGMA PERFORMANCE of the process.
e.g. Operational definitions may determine if a team is to count all the defects on an invoice (required to calculate defects per million opportunities) or the total number of defective invoices (any invoice with any defect) or the type of defects encountered on an invoice (to eliminate the most common defects first). Each of these cases may require a very different approach for gathering the data”
438
• For example, if client wants to ensure that each transactions to be completed with in 20 min, it can be considered as specification and follow the continuous data path.
• If client is interested in controlling defective transactions, the entire unit is either good or Bad. A proportion can be calculated (Binominal).
• If operation head or client head is interested in minimizing the abandoned calls and team is interested in identifying the steps / task / actions in the process, First team suppose to map the process and identify the steps which results in abandoned calls and those steps can be considered as Opportunity for Defect.
• In some cases, client may scope the improvement area. In that situation,team can consider only that portion and identify the Opportunities for Defects.
Examples:
439
• If the measurement and improvement of process characteristics calls for noting all the defects, each detail of the process to be considered OFD.
• If there is no limit to the number of defects that can be counted, It is not possible to count the non-defects, poison distribution can be used.
• If operation head or client head is interested in minimizing the abandoned calls and team is interested in identifying the steps / task / actions in the process, First team suppose to map the process and identify the steps which results in abandoned calls and those steps can be considered as Opportunity for Defect.
Examples:
440
Exercise for DPMO calculation for Discrete data
The Inspection result for a set of 100 Purchase Orders (PO) are given in the Table below:
Cause of Rejection Number of Defects
Supplier Name Incorrect 1
Supplier Door # Incorrect 1
Quantity Higher than that in Indent 3
Quantity less than that in Indent 1
Price is higher than that in the Indent 2
Price is lower than that in the Indent 2
Number of Defects =
Number of Opportunities for Defects =
441
DPU: Defects Per UnitThe ratio of Number of Defects found to the total Number of Items Inspected
Number of Defects = 10
DPU = Number of Defects / Total Number Units Inspected = 10 / 100 = 0.1
Cause of Rejection Number of Defects
Supplier Name Incorrect 1
Supplier Door # Incorrect 1
Quantity Higher than that in Indent 3
Quantity less than that in Indent 1
Price is higher than that in the Indent 2
Price is lower than that in the Indent 2
442
DPO: Defects Per Opportunity
Ratio of total number of Defects to the total number of opportunities in the inspected lot.
DPO = Defects / (Opportunities x Total number of Units Inspected)
DPO = 10 / (100 x 4) = 0.025
Cause of Rejection Number of Defects
Supplier Name Incorrect 1
Supplier Door # Incorrect 1
Quantity Higher than that in Indent 3
Quantity less than that in Indent 1
Price is higher than that in the Indent 2
Price is lower than that in the Indent 2
443
DPMO: Defects Per Million Opportunity
DPMO = DPO x 1000000
DPO = Defects / (Opportunities x Total number Inspected)
DPO = 10 / (100 x 4) = 0.025
DPMO = DPO x 1000000 = 0.025 x 1000000 = 25000
Cause of Rejection Number of Defects
Supplier Name Incorrect 1
Supplier Door # Incorrect 1
Quantity Higher than that in Indent 3
Quantity less than that in Indent 1
Price is higher than that in the Indent 2
Price is lower than that in the Indent 2
444
Yield: Yield = e-DPU
DPU = Defects / (Total Number Inspected) = 10 / 100
= 0.01
Yield = e-DPU= e-0.01=0.99005 = 99 %
Cause of Rejection Number of Defects
Supplier Name Incorrect 1
Supplier Door # Incorrect 1
Quantity Higher than that in Indent 3
Quantity less than that in Indent 1
Price is higher than that in the Indent 2
Price is lower than that in the Indent 2
445
ZST: Short Term Sigma Value
Z is the Standard Normal Variate equivalent to DPO obtained from Z table.
DPO = Defects / (Opportunities x Total number Inspected)
DPO = 10 / (100 x 4) = 0.025
ZST = 1.96 From conversion tables
Cause of Rejection Number of Defects
Supplier Name Incorrect 1
Supplier Door # Incorrect 1
Quantity Higher than that in Indent 3
Quantity less than that in Indent 1
Price is higher than that in the Indent 2
Price is lower than that in the Indent 2
446
ZLT: Long Term Sigma Value
ZLT = ZST - 1.5
DPO = 0.025
ZST = 1.96 From conversion tables
ZLT = ZST - 1.5 = 1.96 - 1.5 = 0.46
Cause of Rejection Number of Defects
Supplier Name Incorrect 1
Supplier Door # Incorrect 1
Quantity Higher than that in Indent 3
Quantity less than that in Indent 1
Price is higher than that in the Indent 2
Price is lower than that in the Indent 2
447
1. Refer to your workbook.
2. Calculate process capability for a given example
Sigma level calculationsExercise 1.20 (20 minutes)
448
Some Helpful Hints:
• It is always preferable to deal with continuous data. Continuous data is measured on a continuum or scale.
• Collect the cause data along with performance data for initial quick wins. Once you determine the entire processes, collect the additional data related to those causes.
• Always evaluate the colleted data, before calculating the base line capability
449
Some Helpful Hints:
• Proportion Defective: The entire unit is either good or bad. A proportion can be calculated. Assume Binomial
• Count of defects: There is no limit to the number of defects that can be counted. Assume poison.
• Calculate Zlt value using Ppk (noted in Minitab output) instead of Cpk, as Ppk is represents the long term process capability
• Not all the percentages are discrete or count data. Eg. % system availability. If both the numerator and denominator are determined by measuring the % is considered continuous data.
450
ExerciseD & M Phase deliverables
Table 1: Raw data on Transaction time of express Teller
Woking day Transaction Times(sec)
Appraiser A Appraiser B Appraiser C
1 63 55 56 53 61 64
2 69 63 60 65 61 66
3 57 60 61 65 66 62
4 58 64 60 61 57 65
5 79 68 65 61 74 71
6 55 66 62 63 56 52
7 57 61 58 64 55 63
8 58 51 61 57 66 59
9 65 66 62 68 61 67
10 73 66 61 70 72 78
11 57 63 56 64 62 59
12 66 63 65 59 70 61
13 63 53 69 60 61 58
14 68 67 59 58 65 59
15 70 62 66 80 71 76
16 65 59 60 61 62 65
17 63 69 58 56 66 61
18 61 56 62 59 57 55
19 65 57 69 62 58 72
20 70 60 67 79 75 68
Deliverables
1 Type of data
2 Descriptive Statistics (Mean, Median, Mode, SD, Histogram)
3 Sample Size & confidence interval (Accuracy required 10 Secs)
4 Probability of getting transaction time between 55 to 65 seconds.
5 Probability of getting transaction time > 70 seconds
6 Normality Test
7 Process Capability ( LSL : 58 Secs, USL : 65 Seconds)
8 Sigma level calculation
9 Gauge R & R
451
1. Refer to your workbook.
2. Calculate process capability for a given example
Sigma level calculationsWeek 1 Define & Measure Phase
452
Thank You
Recommended