Sixgreen Courseware PDF

Lean Six Sigma
Green Belt
Courseware
Version 2.6
Copyright © Property of Firebrand Training Ltd www.firebrandtraining.com

08/10/2014
ORIENTATION
Lean Six Sigma
Green Belt
Welcome
Welcome to the
Lean Six Sigma
Green Belt
Training
Orientation v2.6 2
1
08/10/2014
Introductions
Be prepared to share with the class your:

- Name
- Designation
- Organisation
- Location
- Expectations for the course
- Summary of your project
Orientation v2.6 3
Programme Schedule
Day 1 - Define
Define the scope of the problem to be tackled – in terms of the customer
and/or business requirements and the process that delivers these
Day 2 - Measure
Map the “as-is” process and measure the current performance
Day 3 - Analyse
Analyse the gap between the current and desired performance, prioritise
problems and identify root causes
Day 4 - Improve
Generate the improvement solutions to fix the problems, implement them
and prevent them from re-occurring, so that the required financial and
other performance goals are met
Day 5 - Control
Ensure that the improvement continues Orientation v2.6 4
2
08/10/2014
FIREBRAND LEAN SIX SIGMA GREEN BELT

Certification Exam
When
Day 5, 2.30pm
Duration
1 hour (60 minutes)
Number of Questions & Format

30 questions, open book
Assessment
Questions test candidates on LEAN & DMAIC principles, process & precepts
Passing Score
75%
Orientation v2.6 5
ORIENTATION
LEAN
3
08/10/2014
Objectives of this module
At the end of this module, you will be able to

Understand the history and principals of Lean and Six
Sigma
Explain the differences between Lean and Six Sigma
Explain how Lean Six Sigma is deployed within a
business
Understand the role of a Green Belt project manager
Orientation v2.6 7
Lean Thinking
Lean emerged from post-WWII Japanese

automobile industry as a fundamentally
more efficient system than mass
production
Orientation v2.6 8
4
08/10/2014
History of Lean
Ford Rouge Plant
Indianapolis 300
Supermarkets
Toyota
Production
System
Orientation v2.6 9
Lean Thinking, continued

CRAFT MASS LEAN THINKING
PRODUCTION
Focus Task Product Customer
Operation Single items Batch & queue Synchronized
flow & pull
Overall Aim Mastery of craft Reduce cost & Eliminate waster
increase & add value
efficiency.
Quality Integration (part Inspection (a 2nd Inclusion (built
of craft) stage after in by design &
production) methods)
Business Strategy Customization Economics of Flexibility &
scale & adaptability
automation
Improvement Master-driven Expert-driven Worker driven
continuous periodic continuous
improvement improvement improvement
Orientation v2.6 10
5
08/10/2014
Lean Thinking, continued
Lean thinking is the dynamic, knowledge-

driven & customer-focused process
through which all people in a defined
enterprise continuously eliminate waste
& create value
Orientation v2.6 11
8 Forms of Waste
T
I
M
W
O
O
D
S
Orientation v2.6 12
6
08/10/2014
8 Forms of Waste - Continued

WASTE Description
Transport Moving people, materials and information around the
organisation
Inventory Any supply in excess of one piece flow
Motion Any movement of people that does not add any value
to the product or service
Waiting For people, machines, materials, information, etc
Over-production Doing things earlier / faster than the next process needs
Over-processing Effort that adds no value to the service from the

customer’s viewpoint
Defects Having to re-do work that wasn’t done right the first time
Skills Not utilising people’s experience, skills, knowledge,

creativity
Orientation v2.6 13
LEAN
Customer
Value
Map the
Seek Maximise Value
Perfection
Customer Stream
Value
While
Minimising
Waste
Pull Flow
Orientation v2.6 14
7
08/10/2014
Reducing Waste & Ancillary Activities

In order to increase the percentage of value adding activities,
the focus is to minimise the time and effort spent on the
waste and ancillary activities.
Understand value from the point of view of
the customer. It is essential not to spend
time on activities which do not add value
Waste for the customer.
Ancillary Objective
Activities Maximise the proportion of time
spent on value adding activities by
removing wasteful and ancillary
Value adding activities
Activities
Original Improved
Process Process
Orientation v2.6 16
In a traditional organisation, the value adding ratio= <10%
Lean in a nutshell
Jaguar Racing in the 1950’s & 1960’s
Time spent
Jaguar focussed on removal of the waste (braking)

Increased the value (acceleration) proportion
automatically Orientation v2.6 17
8
08/10/2014
ORIENTATION
SIX SIGMA
What is Six Sigma?
Six Sigma is a 5-phase problem solving methodology that

Understands a business problem
Translates it into a statistical problem
Solves the statistical problem
Translates it back into a business solution
Six Sigma is a data driven philosophy & process resulting

in dramatic improvement in product/service quality &
customer satisfaction
Using Six Sigma reduces the amount of defective
products manufactured or services provided, resulting in
increased revenue and greater customer satisfaction.
Orientation v2.6 19
9
08/10/2014
Definition of Six Sigma
What is Six Sigma?

Sigma is the symbol for Standard Deviation
Standard Deviation is a measure of the data variation
Standard Deviation is calculated from the data from the
s
process – it’s the Voice of the Process
Orientation v2.6 20
What is Six Sigma? … and what is a standard deviation?
7612
Standard
Deviation
7210
68 8
Variation
64 6
Mean
60 4
56
2
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Variation exists in all processes.
The principal aim of Six Sigma is to reduce variation to

operate consistently within customer expectations
Orientation v2.6 21
10
08/10/2014
Definition of Sigma Level

Sigma Level is a measure of process capability
Sigma Level requires customer specifications to calculate the
capability of the process.
These are derived from the Voice of the Customer (VOC)
Sigma Level states how many Standard Deviations lie between

the average and the nearest customer specification limit
VOICE OF THE CUSTOMER
VOICE OF THE PROCESS
LSL USL
s s s s
Orientation v2.6 22
Target Performance
Runway (Manchester Airport)
05
R
Touchdown target
How well can the pilot achieve the target performance?

We want to be able to measure the performance
and use it to compare the quality of pilots.
If we were to select a pilot for a new route to
SABA – how do we compare their performance? Orientation v2.6 23
11
08/10/2014
Why Use Six Sigma?
Variation Errors & Poor Dissatisfied Business

Defects Quality Customers Loss
Orientation v2.6 24
Is 99% Good Enough?
3.8-Sigma Level 6-Sigma Level

99% Good 99.99966% Good
• 20,000 lost articles of mail per • Seven articles lost per hour
hour
• Unsafe drinking water for almost • Unsafe drinking water one
15 minutes each day minute every seven months
• 5,000 incorrect surgical • 1.7 incorrect operations per

operations per week week
• 11.8 Million shares incorrectly • 4,021 shares incorrectly traded

traded on the NYSE every day on the NYSE every day
• 10,700 defects per million • 3.4 defects per million

opportunities opportunities
Orientation v2.6 25
12
08/10/2014
Y = f(x)
Y = f (x1;x2;x3;xn)
Outcome = function of the factors of the process
So what are the Factors for a x1 – coffee beans

Great cup of coffee? x2 – Volume of water
x3 – Volume of coffee
Cup of coffee = f (coffee x4 – hardness of water
beans; water) x5 – pH levels of water
xn – etc.
Orientation v2.6 26
Six Sigma Overview Summary

1. The objective of Six Sigma is to reduce process
variation such that a process continues to meet
customer expectations over time
2. To reduce variation it identifies then reduces the
statistically validated root causes of variation
3. Six Sigma performance means a near defect free
process (<3.4 defects per Million opportunities)
4. A Sigma level is a measure of capability for the
process to meet the Customer Specifications
5. Freeing a process from producing defects means
that capacity increases or throughput time
decreases (faster)
Orientation v2.6 27
13
08/10/2014
ORIENTATION
LEAN SIX SIGMA
Six Sigma v Lean in a nutshell
Imagine a leaking pipe and consider how

LEAN and Six Sigma work together to tackle
the problem……
Six Sigma tackles defects/rework, one of the

wastes Lean attacks
Orientation v2.6 29
14
08/10/2014
Lean Six Sigma Simulation
The coin simulation

Objective: To illustrate how to apply Lean Six Sigma principals
Your instructor will read the instructions of the simulation
Orientation v2.6 30
15
ENTERPRISE
Deploying Lean Six Sigma
in the business
Deploying Lean Six Sigma in the business
There are four key components critical for a successful Lean Six Sigma deployment
Leadership Right
Engagement people
Right
Customer
Projects
focus
Let’s take a closer look at each of these in turn starting with Customer Focus
Enterprise v2.6 2
1
Customer focus
Lean Six Sigma is customer centric. It strives to delight

customers by delivery of product/service to customer
specification time after time
Understanding requirements through techniques such as
Voice of the Customer (VoC) is critical to measure success
in this regard
Requirements are translated to measurable characteristics
called Critical to Quality (CTQ) which measure adherence
to these requirements
Delighting customers drives business profitability
Enterprise v2.6 3
Customer focus
The satisfaction of customer needs (internal

and external) drives quality, efficiency and
ultimately profitability
Cross-functional teams working together
focussed on a clear definition of customer
requirements will deliver increased
customer value and reduced waste
Enterprise v2.6 4
2
Leadership Engagement
Why should leaders lend their support to your LSS

deployment?
Enabler of strategy
Improve customer experience
Develop the problem solving capability of the business
Tackle their most challenging problems
Return on investment
Enterprise v2.6 5
Leadership Engagement
What should Lean Six Sigma leaders do?

Use process improvement tools and methods in daily work
Create a team from different parts or the organisation and/or
external organisations
Be responsive to support teams through tough patches
Get involved by participating in some events
Develop the capability of the teams
Hold the team and themselves accountable for the success of
improvement opportunities
Communicate widely on the progress and success of the deployment
Enterprise v2.6 6
3
Right Projects
For a new lean six sigma deployment to gain momentum,
project selection is critical
Too many projects being worked (resources spread too thin),
results in longer than needed cycle times
Too many marginal-value projects being worked = Low ROI
Enterprise v2.6 8
Right Projects
Consistently use prioritisation and selection

criteria
Related to driving Economic Profit and Revenue Growth
Actively manage projects-in-process

Stop working marginal value projects
Launch projects based on skill not resource availability
Enterprise v2.6 9
4
Right Projects
1 Identify Value
2 Identify Project
3 Screen Initial List of
4 Scope and Define
5 Prioritise List of
Levers Opportunities Opportunities Projects Defined Projects
 Identify value levers in  Translate Value Levers  Score each project as  Assign opportunities  Evaluate projects
the business into Opportunity Areas High / Med / Low for to project sponsors for using Evaluation
• Strategic  Translate Opportunity Benefit and Effort project definition Criteria
• Financial Areas into Project  Fill in Benefit / Effort  Complete Draft Project  Update Benefit / Effort
• Client Ideas Matrix Charters Matrix
• Operational (Process)  Select highest priority  Review plotted results
 Prioritise value levers opportunities for
 Prioritise projects
further analysis
 Schedule project
launches based on
resource availability
Enterprise v2.6 10
Right Projects – Identify Value Levers
Strategy
Voice of the
VOC Financial VOB
Client
 Client Needs provides focus on  Financial analysis drives to

critical client requirements and tactical business processes
drives to process performance. Process
 Considers key enablers
that span the processes, such
Process performance links to strategy, as Voice of our People
client, and financial levers.
Revenue Growth Economic Profit
Market Value
Insight: Tools such as Scorecards and Dashboards are used to translate
Strategy into Key Process Metrics in order to sustain improvement results
Enterprise v2.6 11
5
Right Projects – Identify Project Opportunities
Strategic Focus ROIC Opportunity Areas Project Ideas

Area/Value Driver
Simplify pricing options for ABC
Improve selling process
Streamline quotation process
Reduce delivery time for…

Proj #1
Improve delivery of services Decrease delivery time variability
Proj #2
Reduce Invoicing errors
Maintain
Proj #3
Relentless
Minimise Rework on Briefs
Client Focus Improve quality products Proj #4
Improve Initial Gathering of Requirements
Improve brochures
Improve client support
Improve client response time
Too broad to address Objectives that may be

directly with a Lean Six addressed with a Lean Six
Sigma project Sigma project
Enterprise v2.6 12
Right Projects – Screen Initial List of

Opportunities
Highly Desirable Opportunities
Benefit Projects in upper left are the most
desirable projects.
Potentially Desirable Opportunities

Projects in the upper right are
High
16
7
1 8 12 potentially desirable, but usually
require more analysis to ensure
3 good decision making. “Tie
breaking” variables such as
4 strategic fit, resource availability
and project type may also be
Med
17 10 employed.
9 13
6 “Potential Quick Hits”

While typically low in benefit,
these can be executed with little
14 effort .
Low
11 15 2
5
Least Desirable Opportunities
Projects in the lower right are the
Low Med High
Effort least desirable.
Enterprise v2.6 13
6
Right Projects - Scope and Define Projects
High priority project ideas are assigned to

Process Owners for project definition
Charters must be completed in enough detail to
enable final prioritisation
We’ll cover more on charters within the Define

module
Enterprise v2.6 14

Problem Statement
Problem Statement Purpose

Focuses the team on a process deficiency
Communicates the significance to others
The problem statement does not include any guesswork as to the
cause of the deficiency or what actions will be taken
A POOR Problem Statement A GOOD Problem Statement
Process rework is too high due to In 1999, sub process A had 480 sales
process A and will be reduced by returned, 58% of total returns,
analysing first and second level resulting in a profit impact of $2.9MM,
pareto charts. and customer dissatisfaction.
Enterprise v2.6 15
7
Project objective
Example 1 Example 2
A Poor Objective A Good Objective
Reduce returns by Reduce sub process A returns

implementing individual from 450 to 225 by year-end,
performance measures and resulting in
objectives a benefit of $1.5MM.
The Project objective does not state the cause of the deficiency or
what actions will be taken.
As it is progressed, the Project team will determine what areas need
to be improved.
Enterprise v2.6 16

Primary Metric
The primary metric is the yardstick that will be used to measure
the success of your Project
It must be consistent with the Problem Statement and Project
Objective.
It is plotted on a time series graph, with the following content:
Actual Performance
Baseline Performance (average over time or number of projects)
Target Performance
It should reflect 6-12 months of historical data and be updated during the
project
Enterprise v2.6 17
8
Secondary Metric
The Secondary Metric is the conscience that will “keep

you honest”
Otherwise, you could improve or optimise one portion of
the process at the expense of another
The Secondary Metric has no target
As with the primary metric, the data should reflect 6-12
months of historical data, and be updated during the
project
Enterprise v2.6 18

Contact Centre case study
The call centre’s leadership team identified a major gap to achieve world class
service level performance:
World Class Performance – 90% of calls answered within 30 seconds
The Industry Average – 70% of calls answered within 30 seconds
ABC’s Performance – 45.5% of calls answered within 30 seconds
% of Credit Card Call Centre Calls

Answered Within 30 Seconds, Last Year
0.5
0.49
0.48
Percent Calls
Ans Per Day
0.47
0.46
0.45
0.44
0.43
0.42
0.41
0 100 200 300
Rows
Enterprise v2.6 19
9
Contact Centre case study (continued)
Mr A Champion has asked you to support development of the

project charter by drafting
A problem statement
A project objective
Potential secondary metric(s)
Any further questions and/or research you may need to
complete the charter
Use the flipcharts to draft and present your work to the class
Enterprise v2.6 20
Right Projects – Prioritize List of Defined Projects

The project charter will drive the most appropriate project “vehicle”
Quick win Local project Large project Major change
Type Well defined Simple project Complex Large scale

issue with with no obvious problem with project/programme
known solution solution no obvious with high impact
solution and complexity
Timescale 1 week – 1 1 – 6 months 6 – 12 months 12 – 24 months
month
Sponsorship Local Local Head of Business unit
function executive
Improvement Stakeholders Green Belts Black Belts Master Black Belts
enablers Potentially with Potentially with
Green Belt Black Belt support
support
Ensuring that the correct projects are selected is critically important

Enterprise v2.6 21
10
Right Projects – Prioritize List of Defined
Projects
Project selection can be thought of as part art, part science
Alternative project
Project Define and
ideas scope charters
Programme Governance
Defects Quantitative and qualitative
Regular assessment of in- criteria used to select from
flight projects, opportunities, £ opportunity multiple project options
selection criteria and
resourcing Safety
Project charter
assigned
Enterprise v2.6 22
Right People -The Typical Flawed Approach
Continuous Improvement The Organisation

“Ping-Pong” Ball
No
“Beach Ball”
Integration
Investment in Time,
Money and… …Technical
Training
•Key Influencers in organisation are not
involved – this creates a lack of
understanding and encourages
resistance
Enterprise v2.6 23
11
Right People -The organisational view
KEY INFLUENCERS
1 Define the Strategy / Vision
LEAN SIX
CEO SIGMA
RESOURCES 2 Leadership Engagement
EXECUTIVES
1
DIRECTORS 3 Project Selection &
2 Sponsorship
3 People Selection and Skills

4 Development
MANAGERS
5 Deployment & Control
4
EMPLOYEES 5
Enterprise v2.6 24
Right People - Being a Green Belt
The objectives of a strong Lean Six Sigma Green Belt are to:
Be able to describe the Lean Six Sigma methodology
Deliver Lean Six Sigma projects utilising the right tools
and techniques at the right time
Support your business to build problem solving capability
Support Black Belts and Master Black Belts in the
execution of complex projects
To support you in this, there are a number of key support
roles required.
Enterprise v2.6 25
12
Right People - Lean Six Sigma Roles &
Responsibilities
Executives / Sponsor Champion & Process Owner

 Owns vision, direction, Champion &  Decides to launch
Executives/
business results Process project
Sponsor
Owner
 Leads Change
 Provides resources /
means
LSS Project Manager  Helps moving

roadblocks
 Validates solutions to
 Has been trained on Team be implemented
Members
Lean Six Sigma
 Has time dedicated to Coaching
leading projects using
Lean Six Sigma
methods and tools  From experienced
 Supports team in Project Project Project People
executing project plan Manager Coach  Supports project
 Provides support in managers and project
implementing teams
continuous
improvement elements
Enterprise v2.6 26
ENTERPRISE
In Summary….
13
Key learning points
The history of Lean and Six Sigma

The differences and complimentary
nature of Lean and Six Sigma
How to establish a Lean Six Sigma
deployment
The key roles and responsibilities
Enterprise v2.6 28
Recommended coach support points
When defining value from a

customer perspective
Developing your project charter
Developing your plan baseline
process performance
Sizing potential opportunities
Engaging key
project/programme
stakeholders for the first time
Enterprise v2.6 29
14
DEFINE
Lean Six Sigma
Green Belt
At the end of this module, you will be able to:

Define the business problem
Identify product families
Develop customer focused metrics
Establish your project team
Create your Define project charter
Understand some of the key change
adoption considerations when delivering
projects
2
LSS Green Belt DEFINE v2.6
1
Lean Six Sigma Project Structure
Lean Six Sigma projects are typically delivered through a 5 tollgate
DMAIC process:
Define
Measure
Analyse
Improve
Control
The Define phase is the first of these phases and starts with defining
the business problem
3
DEFINE
Define the business problem
2
Perception of Problems
A problem well defined is a problem half solved…
“If I had one hour to save the world, I

would spend fifty five minutes defining
the problem and only five minutes
finding the solution.”
Einstein
The Define phase of a Lean Six Sigma project is often

considered the most important
5
Scoping the problem

There can be multiple issues to be tackled to solve a
problem as generic as “maintain relentless customer
focus”. We need to break this down to something that we
can execute a project on.
6
3
Scoping the problem
To help us break the problem down we’re going

to illustrate the use of three simple tools:
Brainstorming
In/Out of frame
N/3
7
Brainstorming
Capture In scope/
Ask Loosely Clarify
initial out of
Question Affinitise ideas
ideas scope
For Brainstorming Let people Read out the When it looks Finally the ideas
to be effective – have time to ideas and start like there are need to be
a frame of generate some to make some no more ideas, sorted into
reference is ideas on their form of read through those that are
needed. This can own. When it grouping. You the ideas and to be taken
often be a simple looks like could let the clarify – forward or
question; things have team do this. practice those that may
slowed down, Let any “appreciative be eliminated
Eg. How can we enquiry” – do
gather the additional ideas
grow existing not give your
ideas on a flip be forthcoming
revenues? How own opinion –
chart
can we reduce ask how it
staff costs? relates to the
initial question
8
4
Brainstorming (cont’d) - Affinity Diagrams
1) Gather ideas around the Subject
2) Group into relevant categories
Category 1 Category 2 Category 3
9
In Frame / Out of Frame

Use this tool to enable agreement to be reached on the
scope of the change efforts
Draw a large rectangle on flip chart paper to represent the “picture frame”
Give each team member a packet of post-it notes and fine-tip paper
Each team member writes one idea per post-it
As a group discuss and stick in scope ideas “inside the frame”, out of scope “outside
the frame”
If people have any questions on where their post-it should go or are unsure whether
it should be in scope or not, they should place it on the picture frame
10
5
N/3 Technique
Use this tool to prioritise ideas based on voting.
This tool will reduce the number of alternatives to a small, manageable
number.
Count the Each Review to

Divide by
number of participant Vote ensure it
3 get N/3 votes make sense
ideas
This tool should be used after generating as many alternatives as possible
The “N” is the total volume of ideas.
11
Scoping the project
Defining your project scope

Objective: To illustrate the use of narrowing tools
Using your own project or use the example of a mobile

telephone company looking to improve customer service to try:
Brainstorming
In/Out of frame
Time - 45 minutes
N/3 Technique
12
6
DEFINE
Product families
What is a product family?
A product family is a group of products or services that

use the same or similar processing steps and equipment.
As a value stream represents the end to end flow of the
product, a value stream map cannot map more than one
product flow with any clarity.
14
7
Creating a product family matrix
Product families are identified using a product family

matrix.
List the process steps on the top

List the products on the left
Identify steps required for each
product
Group products that have similar
processing requirements
Identify families – groups that use
the same process
15
Creating a product family matrix (cont’d)
The listed process steps should be in enough detail to

show differences in products listed. To be in the
same family, we’re looking for approx 80% of the
processing steps to match.
Products should be listed in enough detail to show the

differences in processes. In a manufacturing company
this is fairly easy being the different products or
services the organisation sells, it can be a little more
challenging in a transactional environment.
16
8
Product family matrix example
The completed matrix below for an energy supplier
shows the identification of five product families.
While not a perfect match for every process step, the

Hot Key product family captures four “products”. 17
Selecting a product family

In the previous example of an energy supplier, five product
families were identified.
In order to decide what to work on first, for each identified
product family we need to asses criteria such as:
Disparity with customer expectations
Importance to company strategy
Opportunity for improvement
Ease or speed of implementation
Volume of units of product
18
9
DEFINE
Develop customer focused metrics
Four step approach
Step 1: Identify Customers

1 2 3 4
Customer Sources of Operating

RUMBA
Identification data Definitions
SIPOC KANO CTQ Tree
20
10
Customer Identification
Customer (noun)
1 a person who buys goods or services from a shop or business: Mr Harrison
was a regular customer at the Golden Lion.
2 [ with adj. ] a person of a specified kind with whom one has to deal: he's a
tough customer.
A Customer is a recipient of an output from a process

They can be either internal or external
They can be both a supplier and a customer
External customers are typically the end-user of the product or service
Internal customers are the recipient of the process output – the next
person in the process chain
21
Identifying customers
External Internal
Customers who directly receive our
Direct product or service
Secondary/ Customers who receive our product

or service through another party
Tertiary
Regulatory and policy setting
Indirect agencies that speak on behalf of
the customer, (eg. FCA)
22
11
SIPOC
Suppliers Inputs Process Outputs Customers
Who supplies What are the What are the Who receives
the inputs to the outputs from the outputs
information / process? List the process? from the
product that out all the List out all process?
moves “pieces” that Process the “pieces”
Name
through the are received that are
process? for the delivered as
process to a result of
perform. the process.
High level High level High level High level High level
process step 1 process step 2 process step 3 process step 4 process step 5
23
SIPOC Example
24
12
Table exercise: using SIPOC
SIPOC
Objective: How to use SIPOC to identify customers
and suppliers of a process
Prepare a SIPOC for one of the projects within

your table team
Time - 20 mins
25
Step 2: Gather VoC
1 2 3 4

RUMBA
SIPOC KANO CTQ Tree
26
13
What does a customer want?
Reliability Communication
Responsiveness Credibility
Competence Security
Access Understanding
Courtesy Tangibles
Performance Reputation
Timeliness Features
……………
27
Voice of the Customer
“If you don’t take care of

the Customer…….. someone
else will”
This module introduces the tools that can be used to better understand
customer requirements and needs;
Determine what a customer values
How to better understand a customer’s requirements
How to prioritise the requirements
28
14
Sources of VOC data
Customer Performance
Telephone Complaints relative to
Calls alternatives
Customer Internal Quality
Inspection Metrics – Scrap,
Meetings Errors, Rework
Focus Groups Sales Reports /

Feedback
Interviews Internal
Intelligence
Questionnaires
Casual
conversations
Personal Visits
Research
Sources for listening Formal

Surveys to the Voice of the Transactions
Customer
Think from the Customer’s perspective

29
Quality of Customer Information

Closeness to Customer
Interview
Focus Group
Survey
Customer
Complaint
data
Effectiveness of Data Gathering
30
15
Step 3: Analyse VoC
1 2 3 4

RUMBA
SIPOC KANO CTQ Tree
31
Analysing VOC data

Many of the methods for gathering the VOC means that the data
captured is often too abstract to be meaningful
Often the information is presented as complaints or solutions – it
needs to be analysed to ensure it is usable rather than just acting on
it as a requirement
We need to do a number of things to test out the requirement, to
ensure it makes sense – and we must have a way to prioritise these
requirements
We need to ensure we know how the customer defines Value
The defined requirements are then
known as the Output Characteristics
32
16
Kano model
Professor Noriaki Kano of Tokyo Rika University
“Attractive Quality and Must-be Quality”, Journal of
Japanese Society for Quality Control, Vol 14, no. 2
(1984).
Kano Analysis does not prioritize customer needs
 Instead it classifies needs
 These classifications can then prioritize the design
effort
Kano Analysis classifies needs, not features
33
Kano Model
Satisfied
 Attractive: unexpected
criteria that if met will
result in delight
Functional
One-dimensional:
Dysfunctional satisfaction is proportional
to performance
Must-be: basic criteria that

if not met will result in
Dissatisfied
 dissatisfaction
34
17
RUMBA - Test the customer requirements
R easonable You or your department can meet the requirement (it does
not violate company procedures, job accountabilities, etc.).
U nderstandable The customers verify that you understand what they are
requiring from you or your work group.
Measurable The specification is measurable - in some way you can

objectively determine the degree or frequency of meeting
the requirement.
B elievable Your co-workers will be agreeable to strive for that level of

achievement.
A chievable
You can meet the requirement. If not, you may need to
renegotiate the specifications as facts and actual data may
later dictate.
A want that meets all five RUMBA criteria is also a valid requirement.
35
Step 4: Define CTQs
1 2 3 4

RUMBA
SIPOC KANO CTQ Tree
36
18
Operational definitions
Consider the following scenario:
Your project goal - reduce the average time to process new customer
applications
Your primary metric - average # of days to complete application
Three fulfillment centres report the following data to your team:
Centre Average # of Days to process applications

1 3.3
2 2.4
3 2.5
What does this information tell you? What actions would you take?
37
The importance of operational definitions
The operational definition for each location:
Average # of Days to Day Request Is

Location process applications Received
1 3.3 1
2 2.4 0
3 2.5 0
Given this measurement information, would you still take

the action you identified previously?
38
19
The importance of operational definitions
Intended to remove ambiguity and ensure consistent
understanding of how data will be collected,
measured and evaluated
Specifies the exact criteria being measured

Provides an exact description of how to calculate
the metrics value
Provides instructions as to how to collect and
utilise the data
39
Be careful!!
The primary metric for most Green Belt
projects are monitored through existing
data collection systems:
Before collecting and analysing the data:
Check for different operational definitions of the metric

among locations/shifts/data collectors, and so on
Check for consistent applications of the operational
definition
40
20
CTQ Definition
Critical to quality is an attribute of a part, assembly,
sub-assembly, product, or process that is literally
critical to quality or more precisely, has a direct and
significant impact on its actual or perceived quality.
A measurable output of a product or a service that is

important to the Customer. Remember, this is from
a Customer’s point of view.
41
CTQ’s
Translate Create the Determine Define the
Identify
to the Key Output the Output Target &
the VOC Characteristic Measure
Issue tolerance
Voice of the CTQ / Target and

Customer Key Issue Metric
Characteristic Tolerance
I am always
Functionality:
transferred to Customer gets to
Want to talk to
three or more the correct person
the right person
different the first time
the first time
people.
Let’s discuss
I’m getting my
Functionality: Customer bill
bill
Consistent received same
at different
delivery day of month
times
of monthly bill
of the month.
42
21
Structuring CTQ’s
CTQ
CTQ CTQ
VOC
Align to Process Y’s

CTQ
CTQ CTQ
CTQ
CTQ CTQ
CTQ
It can be useful to structure the CTQ’s to best visualise them 43

CTQ Tree
Identify Break high

List Customer Look for links
level req’s Assemble
requirements common to Process Ys’
/ needs
into more
from SIPOC the tree
groupings detail
Why use it?

Helps the team move from the broad and often vague high-level
requirements and needs into detailed requirements
Helps to ensure there is a direct relationship between the CTQ’s
and the Process Y’s
44
22
Table exercise: Creating CTQs
Objective: To translate VOC to CTQ from
Work in groups of 3-4 and review the following ‘Voice
of the Customer’ statements (from a Sandwich shop):
“It takes ages before I get my sandwich”
“I really don’t like your seats”
Complete VOC to CTQ template on next slide

Time - 5 mins
45
Exercise: Translate VOC to CTQ
Voice of the CTQ / Target and

Key Issue Metric
Customer Characteristic Tolerance
It takes ages
before I get my
sandwich
I really don’t
like your seats
46
23
VOC & CTQ Summary
Understanding customer requirements is critical for all

Lean Six Sigma projects
There are numerous primary, secondary and tertiary
sources of VoC
Customer needs can be classified using Kano analysis
and prioritised
VoC should be translated to CTQs
CTQs should be measurable following an agreed
operational definition
47
Voice of the Business (VOB)

Derived from two primary drivers
What is the profit to What is the customer

be delivered from the experience to be
process? delivered?
What are the primary performance How does the customer expect to be
guidelines we want to achieve? treated when dealing with us?
Unit costs How do we differentiate from our

competitors?
Transaction costs
Can we make negative
Manpower levels
interactions seem positive to the
Asset utilisation customer?
Material costs How do we make ourselves easy
to work with?
48
24
Critical to the Business (CTB)
Stakeholders
Controllership Employee
Regulatory Safety
Work Council Morale
Critical
Unions to the Satisfaction
Legal Business
CTB
Shareholder Community
Net Income Environmental
Return On Investment Local Relations
EBIT
 Often these interests result in constraints or boundaries that are imposed on

the process
49
VOC vs VOB
Lean Six Sigma is based on the belief that business
benefit is derived from delivering customer
requirements – it’s customer focused
Many of the techniques for hearing and analysing VOC can

and should be applied to VOB
CTB’s are more often constraints than output targets:

The business may not have defined targets for processes
The business may not have defined the customer experience it wants to
deliver
50
25
Customer Requirements Summary
• Understanding & clarifying

customer requirements is
a 4-stage approach:
1. Identify Customers (SIPOC)
2. Gather their requirements (survey, interviews etc)
3. Analyse the customer information (Affinity & Kano)
4. Define the CTQ (target & tolerance)
• Translating VOC into CTQ provides

a method to measure whether the
customer requirements are being
achieved
The CTQ makes the VOC measurable 51
DEFINE
Project Charter
26
The project charter
The project charter is the primary output document of

the Define phase.
It pulls together all the components of Define
The problem statement
Business case
Scope
Goal statement
Team roles, responsibilities and time commitment
High level timeline of tollgate delivery
Could be thought of as a “contract” between the

project team and the business
53
A basic project charter – Page 1 of 2
54
27
A basic project charter – Page 2 of 2
55
The problem statement

The problem statement should identify what problem this project is focused
on solving. The problem statement should contain:
Who - Who is being impacted? (What customer)

What- What is the issue that is impacting the customer
Where- Where do the customers encounter the problem with the
process? (e.g.. on the phone with NE call centre or on line
bill pay set up screen. etc...)
When- When in the process does the customer experience the
problem?
Impact- What is the impact to the customer when the problem
occurs? (e.g.. calls are being dropped, can't complete bill
pay set up etc...)
Use SIPOC to identify the customer and the output that they receive
56
28
Business case
The business case should provide a compelling case to the Line of Business as
to why this problem needs to be solved.
It should contain:
From a business perspective, why should we do this?
How does this project align with other business initiatives/strategic
plan?
What is the focus for the project team?
What impacts will this project have on other business units and
staff?
What benefits will be derived from the project?
How are the value of the benefits quantified? (e.g.. increased
revenue, lower expense, cost per unit, cost avoidance, strategic
value)
Ensure you enlist the support of Finance in quantifying and validating the
potential financial benefits for your project! 57
Example of a Business Case
58
29
Project scope
The scope should consider:
What authority do we have?

What processes are we addressing?
What is not within scope?
What are the starting and ending points of the process?
What components of the business are included?
What components of the business are not included?
What, if anything, is outside of the project boundaries?
What constraints must the team work under?
In/Out of frame, brainstorming and N/3 are important tools here!
59
Goal statement
The goal statement addresses the key project success metrics, the primary and
secondary metrics (covered in the LSS Enterprise module).
What are my measures of success that are aligned to the project objective?
What are the goals for primary and secondary metrics?
How much does the primary metric need to change for your project to be
considered a success?
Other items to be included

How long will it take you to complete this project? (A summary project
timeline should be completed as shown on page 2 of the Charter)
The goal of the tangible and intangible benefits of the business case?
60
30
DEFINE
Team
The Lean Six Sigma project team

Revision:
In the Enterprise module we covered the following core
roles within the Lean Six Sigma project team
Other important roles include Project Finance Certifier

and Operational Risk
62
31
Individual vs Systemic Efficiency
Efficiency is not
derived from
individual efforts,
but by optimising
the collective
63
The characteristics of a good core team member
A good core team member should:

• Have operational expertise in one or
more sub-processes of the high-level
process flow
• Be familiar with the problem
• Be motivated to fix the problem
• Commit to remain on the project
until its completion
• Have the support of his or her manager
64
32
Document the agreed team roles within the
Project Charter
The Project Charter should document these project roles within

the “team selection”:
Project #: 000000000
Project Chart er
Required information for each Problem Statement
team member:
• Name Business Case Summary Project Scope
• Department
• Role on the project team Goal Statement Team Selection
Team Leader Organi zat i on Black Belt 100%
• Approximate percentage of time

Project Champi on Organi zat i on Champi on 100%
Coach Organi zat i on M BB 10%
Team M ember Organi zat i on SM E 20%
required on the project

Team M ember Organi zat i on SM E 20%
Team M ember Organi zat i on Role 20%

1 (01/ 06)
65
Forming a team
FORM TEAM
There is more to forming a team than gathering a group
of people together
All teams will “go
Interdependence
Social relationships development
through” these stages.

PE
The Belts’ role is to be

RF
OR
aware of this and help

M
Cohering the team through.

NO
IN
G
RM
IN
Note that the inclusion

G
ST
Conflict of a new team member

OR
at any stage will cause

M
IN
the team to return to

FO
G
RM
“forming”
IN
Dependence Work organisation development

G
Orientation Information Organisation Problem

flow solving
66
GB M1 06 Forming the Team Iss 3.0 © The Six Sigma Group 16
33
DEFINE
Change Adoption
Why is Change Adoption Important?
E=QxA
E = Effectiveness of the solution
Q = Technical quality of the solution

A = Acceptance of the people
68
34
The change curve
COMMITMENT
Happiness
Moving on
DENIAL
Morale & Competence
Gradual Acceptance
Fear
Threat
Understanding
Guilt
Anger / hostility
DEPRESSION
Defeat
Time
69
Original source Elizabeth Kubler-Ross (1969) LSS Green Belt DEFINE v2.6
Change targets
• People who are directly impacted by the change; way of

Impacted working, organisation etc
Populations • Typically front-line, staff, middle management etc
• Usually managed on a group basis
• Key individuals involved in the project or who have a

Key vested interest in the project
Stakeholders
• Typically top-management
(individuals)
• Usually managed on an individual basis
Work Councils,
• Organisations involved in the management of social &
Unions &
legal aspects of the change
Authorities
70
35
DMAIC Change Management Strategy
Define Measure Analyse Improve Control Reinforce
Identify impacted
Assess skill gaps and set-up training
Impacted population
populations Understand benefits and resistance Set-up accompanying structure and action plan
to change to minimise resistance to change
Involve / Inform management and teams
Key Identify stakeholders

Stakeholders
Understand stakes and attitudes
(individuals)
Engage stakeholders
Understand social impact Inform / Consult Work Councils Keep Work Councils
of the project / Unions Unions informed
Work Councils
/ Unions Review legal
requirements 71
Change adoption
Impact Training
Impacted
populations analysis plan
Key
Leadership
Stakeholders
engagement
(individuals)
Stakeholder
mapping
Work Councils / Change history

Unions assessment
Communication
Communication
plan 72
36
Identify & Engage Key Stakeholders
 Identify the key stakeholders associated with the project
 Evaluate their influence over the project
 Assess how much they support the project
 Prioritise communications actions in order to engage stakeholders and to

maximise their positive impact on the project
73
Resistance distribution
“What’s in it for me” “Need to touch to believe”
“Up for Change – no

matter what”
“See it as MY role”
Early Late
Innovators Adopters Adopters Resistors
10% 40% 40% 10%
74
37
In relationships, Synergy and Antagonism
co-exist...
Synergy
• Synergy > Antagonism :
1 it’s easy to work together !
1
«red line»
• Synergy < Antagonism :
3 2 easy to disagree with each
? other !
2
3 • Synergy = Antagonism :
I am still undecided
Antagonism
We’d like to do something together (the Bridge), but ...
the Bridge
The Gap
... we have different priorities, interests and constraints

(the Gap)… LSS Green Belt DEFINE v2.6
75
4 levels of Synergy and Antagonism can be

expressed at the same time
Synergy «Blue lines» =

under the line (+1, +2 or -1, -2)
Synergy and Antagonism
are passive qualities.
Devoted +4
Active
Synergy
Co-operative +3
Only on a
Interested +2 specific item,
Passive at a given time !
Synergy
Minimalist +1
Antagonism
-1 -2 -3 -4
Passive Active
Antagonism Antagonism
Indifferent
Disinterested
Reluctant
Hostile 76
38
Stakeholders mapping – Principles
Synergy Antagonism attitudes
Intention credit / common game Personal game
+4
Synergy -4
He/she identifies themselves with He/she leverages absolutely
the project team in order to help everything to impose his/her
reach a common objective, even own solution. If he/she does not
devoted
if the team members are reticent succeed in doing so, he/she will
break off the relationship !
co-operative +4
+3 -3
He/she takes the initiative to He/she uses more important levers
develop synergies, seeks to impose his own solution. If they
consensus, offers suggestions, do not succeed in doing so, he/she
uses some of his own levers to gives in and only to superior power
interested +3
support the project team. If they and for a given period

follow him/her, then synergy will
increase in their relationship.
+2 -2
minimalist +2
He/she is interested in what the He/she uses certain levers and

project team want, say or do. He tries to obtain the best possible
consults, discusses and works but compromise for themselves via
does not seek to go beyond what negotiation. He/she can give in at
+1
is strictly necessary… -1 disinterested -3 hostile

the end of the day…
+1 indifferent-2 reluctant -4 -1
Rather favourable, he/she limits He/she hesitates because they are
his/her relationship with the
Antagonism fairly indecisive. He/she will rally
project team to the strict to a contrary point of view in a
essentials… passive manner, with someone who
is more decisive than them
77
Stakeholder mapping : an essential tool to cope with

Antagonism
• Passive : they do the minimum Synergy

• Undecided : they are ready to get involved if
they get something in return
Committed
devoted
Constructive Torn
• Hostile : they will give in only when
apart
co-operative +4
confronted by a stronger power ; they can

decide to stop the confrontation
• Opponent : they are driven by
interested +3
the satisfaction of their own interest

• Committed : they support the cause
minimalist +2
• Constructive : they support the project,

and may offer positive criticism Passive Hostile
• Torn apart : they have equal amounts
+1
of love and hate... -1 disinterested -3

indifferent-2 reluctant
hostile
-4
Antagonism
78
39
Stakeholder management – The Allies strategy
Once stakeholders are identified,
then elaborate a strategy for each
Synergy category
Allies management Strategy

Torn
devoted
Constructive
Committed
apart • Spend 2/3 of your time with your "Allies" :

co-operative +4
− Praise their efforts, accept them for what they

are, support them, spend time with them…
and in return, be demanding !
− Surround the Passives and the Undecided,
motivate them by offering them a joint stream
interested +3
to work on, compromise on issues of lesser

importance…
− Propose / suggest (at any moment) to the
Passives to rally to a project, thus increasing
minimalist +2
their synergy
Passive Hostile
• Use your Allies to convince, ignore or exclude your
+1
opponents :
-1 disinterested -3 hostile
indifferent-2 reluctant -4 − With 1/3 of your time (maximum), contain
the attempts of the Opponents and Hostiles
to rally the Undecided and the Passives to their
Antagonism side, by making counter-offers.
Any positioning is only meaningful : − If necessary, neutralise the Hostiles, maybe by
changing the scope of your project…
 at a given date
 on a given subject / topic
79
Example of Stakeholders management strategy
Synergy
Involve / Support Support

devoted
+4
Federate
Delegate Commit
Involve
Seek their opinion Optimise
Support and build new invested time
Delegate options with them
co-operative
specific
tasks Negotiate
+3
Negotiate
Show
enthusiasm
Use key
Allies
interested
Treat well to convince

+2
or exclude
them
Explain
Impose Secure Isolate

minimalist
Do not leave Get rid

+1
unattended… of them
if necessary
Antagonism
indifferent -2 reluctant -4
80
40
Where should we map our stakeholders?
Where people are located on the stakeholder map depends on
the specific project/point of application and the precise
moment in time
As you us the tool..
Don’t try to guess, ignore or analyse the reasons why....
...rather measure the facts, focus on behavioural observation,
Many positions seem irrational but...
… there is always a rationale in the individual interests of different partners !
Build your perceptions through cross information, exchange points of view with
team members
Define key actions with project management
Don't think people will stay where you saw them the last time !
Things, actions and positions move… all the time !...
Because that's what Change Adoption is all about…LSS Green Belt DEFINE v2.6 81
Communications Plan
Use a communication plan to ensure stakeholders are kept properly informed
and engaged throughout the change process. Ensure you adjust your
communication strategy based on the issue under discussion and where the
stakeholder is on the stakeholder map
82
41
Table exercise: Stakeholder mapping
Objective: To produce a stakeholder map
Working in groups of 3-4, read the briefing document
on the next slides.
You are a project manager about to kick-off a cross-
functional project on the bid process
1. Map the stakeholders positions
Time 2. Think and plan the top 5 actions to get a better
45 mins + 15 picture of the stakeholder positions
mins 3. Are there any immediate communication needs?
discussion
4. Use the communication plan template to plan
any communication needs
83
Synergy
devoted
Committed
Constructive Torn
+4
apart
co-operative
interested +3
minimalist +2
Passive Hostile
+1
indifferent -2 reluctant -4
84
Antagonism
42
Table exercise: Stakeholder mapping - briefing
• NOTE this is an invented set of characters; there is no similarity to any of

the people involved in the real project
Name Role/Organisation
Leonard Sponsor
Cameron Business Line leader
Giles Head of Credit
Scott Credit Approval Analyst
Ray Bid Manager
Julie Sales Lead
85
Leonard Sponsor
Leonard has worked at BFB for 15 years. His wife also used to
work for BFB where she was a successful Lean Six Sigma Project
Manager and was recently recruited by another Financial Services
company to head their Lean Six Sigma programme.
Leonard has become increasingly interested in Lean Six Sigma as
he has seen it succeed. He has been with BFB long enough to have
seen many other improvement initiatives start and fail.
Before Relationship Management he used to be Head of HR at
BFB. His preferred business style is to reach consensus smoothly
and always put the client first.
He is not a detailed person; always talks about the big picture;
tends to delegate rather than direct.
86
43
Cameron Business line leader
Cameron has 25 years experience in the Financial Services

industry and has successfully identified and built a new
broker/dealer client segment at BFB. Whilst initially profitable,
in the last year the smaller clients have struggled in the post-
crash environment and in turn some have become loss-making for
BFB.
Cameron is commercially and politically astute. He knows there
needs to be radical change to the bid process to ensure new
clients brought in are profitable for BFB. He believes great client
service and a cross-functional bid approach are key to this.
He reports locally to the CEO and to the Global Business Head.
Both are close to retirement age and he knows that if he can
successfully turn-around the segment he could be in line for an
Executive position.
Cameron exudes old-school charm which the clients love and he is
a gentleman to work with at BFB.
87
Giles Head of Credit
Giles has a Masters in Economics from the Sorbonne.

He has taken great intellectual interest in the recent crash in
Financial Services and has written several papers on the topic
which have been published in The Economist.
Giles’ economic expertise is acknowledged and respected within
BFB and he is based in the Head Office so he is on hand to advise
the Global Business Head.
At the weekend he goes sailing with the Head of the BFB Group.
This relationship has helped secure him several senior credit
positions within the group despite his lack of leadership &
communication skills.
His direction to the Credit Approval team is to be risk-averse and
follow the procedures but he also personally looks into some
Credit applications which his team have advised against for his
old colleague Julie (Sales Lead).
88
44
Scott Credit Approval Analyst
Scott has worked in Financial Services for 18 years, 10 at BFB.

His previous role was as a Sales Lead where his fingers were burnt
in the crash when his client went bust and cost the organisation
£2.5million when they defaulted on a loan that he had secured
credit approval for.
Scott took a career change to the Credit Approval team where he
is passionate about protecting the bank from Credit Risk.
His experience in Sales and Credit make him ideal at working with
the client and Sales leads to come to a Credit agreement that
works for both parties. But with some of the smaller broker
dealers, Scott knows that their risk is too high for BFB’s credit
policy and clearly explains this from the start.
Scott works hard and gets frustrated when Julie contacts Giles
after her client application has been declined. Not following the
procedure wastes everyone’s time and frustrates the client.
89
Ray Bid Manager
Ray has worked for BFB for 5 years. He keeps a low profile but is
well respected by his team and works hard and diligently.
He used to be a Police Officer where he enjoyed working in an
environment of strict rules which were followed and great team
spirit.
His role is to compile the bid based on the information provided
to him by Sales and the Business Line. Conditions of the bid are
provided by Credit, Legal and Compliance.
The Request for Proposal (RfP) process works to strict market
deadlines so Ray depends on the information being provided to
him in a timely manner. In reality his team spend a lot of time
chasing and end-up working long hours and weekends before the
deadline.
As a result, quality of the bid suffers and Ray feels disenchanted
by the organisation, struggling to maintain morale in his team.
90
45
Julie Sales Lead
Julie is the rising star of the sales team. She started her career in
Head Office and made it her business to become and stay well
connected with the senior players.
Julie has had some early success with winning broker/dealer
business. Following implementation some of these deals have not
turned out to be as profitable as forecast which has not gone
unnoticed by the Business Line or Credit department.
Julie is the consummate sales person and focuses on the win.
Individualist by nature she does not follow procedures but makes
calls to her network of senior players to make things happen. This
causes friction with her colleagues that make up the deal team
but the behaviour seems to be tolerated and even encouraged by
senior management.
Julie’s ambition is to be the Sales Lead on high profile deals and
eventually transfer back to Head Office to take on the role of
Global Business Head when the incumbent retires.
91
Change adoption - Summary
E=QXA
Change adoption techniques should be
used at every stage of the change journey
Use mapping tools to adjust your
communication approach with different
stakeholders
Remember that stakeholders perception
of change moves over time be prepared
to adapt your approach
92
46
Lean Six Sigma simulation: Round 2
The coin simulation
Objective: Use the tools covered in this module to prepare
A project charter for the process. Your instructor has some

information but will only give it to you if you ask the
correct questions!
Re-run the simulation implementing 1 quick win.
60 mins
93
When agreeing the scope of

your project
Defining your problem
statement
Gathering VoC data and
translating it to CTQs
Drafting your Define tollgate
project charter
94
47
Key points of the module
Defining the business problem is of

critical importance to the success of
any Lean Six Sigma project
The Project Charter pulls together
the key components of the Define
phase
Change adoption efforts should start
as early as possible and continue
through to completion of your
project
95
48
MEASURE
Lean Six Sigma
Green Belt
Use various mapping techniques

Identify potential root causes
Understand the basics of data
Conduct an attribute MSA
Baseline process performance relative to
customer requirements
Create and execute a data collection plan
2
LSS Green Belt MEASURE v2.6
1
Six step approach
1 2 3 4 5 6
Sources of Special and

Value stream Data collection
C&E diagram Data types measurement common
mapping planning
variation cause
Activity of the Planning an

C&E matrix Shape Control charts Sample size
product MSA
Block Operational Calculating Sampling

FMEA Centre
diagrams definitions capability strategy
Detailed Conduct a Calculating

Spread
process maps discrete MSA stability
Analyse MSA
Normality
results
Graphs for
comparison 3
MEASURE
Map the process
2
Step 1: Map the process

mapping planning
variation cause

product MSA

FMEA Centre

Spread
Analyse MSA
Normality
results
Graphs for
comparison 5
Preview of the lesson
In this module we’ll discuss the need to document the

existing process as a first step to improve it. We’ll review:
Different mapping techniques
How to create them
6
3
Types of visual maps
Value Stream Map
See the whole process, from
beginning to end. Focuses on
the journey of a “product” from
order to delivery
Block / Chevron Diagrams Define Measure Analyse Improve Control
High Level, low detail,

typically showing phases
Process Map Request Transmit Conduct Are

Transmit
Complete
Prepare Documents for
Detailed, with different symbols for
services
documents
Documents for
Peer Review
Peer
Review
documents
acceptable
Technical
review
Technical
review
representing different types of

activities
Do Docs
No Yes Store Docs Compile
meet
final
Customer
package
Req’ments
7
Styles of maps
Value stream map (VSM)
A value stream map is all of the actions, both value-

creating and non value-creating required to bring an
order to delivery
A VSM is therefore a “picture” of the process from a
products point of view
Consistent use of icons are needed to create a clear

picture of the value stream.
8
4
The company view of itself
Executive
Project
Sales Commercial Procurement Production
Controls
Strategy Strategy Strategy Strategy Strategy
Objectives Objectives Objectives Objectives Objectives
Team Team Team Team Team
Companies are usually organised vertically – for a transaction to be completed, it flows

up and down each department before being passed onto the next department.
9
The customers’ view of the company

Customer
Company
Project
Sales Commercial Procurement Production
Controls
The customer views and experiences only a proportion of the interactions required in
the production of a product / processing a transaction / provision of a service.
Customers are not concerned with the handoff between teams or their individual
performance. They are concerned by the end result and their experiences along the
way. 10
5
Typical VSM icons
Activity
Tasks
Customer/Supplier Staff
Dedicated Process Shared Process
I
Information Flow Courier
Work in Progress / Supermarket
Inventory
Electronic Withdraw Kanban Production Kanban

Information Database
Flow (Excel/Access, etc.)
Tuesday
+ Thurs
Pull Flow
Push Flow Finished Goods Flow 11
Transport
Creating a value stream map
A “current state” value stream map is prepared to help build a

picture of how the existing process delivers value to the end customer
To create a current state value stream map:
Define a unit of product

The team should follow a “product” through the entire value stream
(physically if possible)
Post-it notes and pencil should be used to sketch out the first drafts
allowing you to refine as you learn more
Use the standard icons as shown on the previous slide
Collect “real” data, don’t rely only on process documentation or
existing reporting
12
6
Typical sequence to building a VSM
Identify Customers Identify customers and define value
& define value Include a clear definition of value from the customer perspective
Demand on the process incl quantities, mix, batch size and working hours
Transport
Transport
Identify activities Delivery frequency, mode of transport
Supplier and Identify activities

supplier info
The activities/processes through which the process is flowing
Transport from
A picture of the flow including parallel processes and branches
supplier
Supplier and supplier information

Data collection What is delivered, delivery frequency, pack sizes
Information flows
Monday
Material flow data + Thurs
Friday
Receive Build Pack

Timeline
13
Key measures and definitions

Identify Customers
& define value Cycle Time CT The rate at which product is produced at
any point in the value stream
Transport
Value Adding Time V/A The time during which value is added for
the customer
Identify activities
Waiting Time WT The time work in progress is waiting for
the next operation to become available
Supplier and
supplier info
Lead Time LT The elapsed time between a piece of
work entering the value stream and the
Transport from
supplier time it is released (to an internal or
external customer) or is completed.
Data collection
Processing Time PT The elapsed time from when the product
enters a process until it leaves that
Information flows process
Uptime UT % of planned operating time the process

Material flow data
operates assuming product is available to
work
Timeline
14
7
Key measures and definitions
Identify Customers
Work in Progress WIP Number of items in process or waiting at
& define value
(Process) each activity
Transport Pack size / Pack-out Volume of product required by the
quantity customer for movement or shipment
Identify activities
Scrap rate % of total product that does not meet
customer requirements and must be
Supplier and
supplier info discarded
Rework rate % of total product that does not meet
Transport from
supplier customer requirements and must be
worked on again
Data collection Defect rate % of total product that does not meet
customer requirements and includes both
Information flows
scrap and rework rates
Number of people Number of people per process on an FTE
Material flow data (full time equivalent) basis
Available time Time that the VSM can run continuously

Timeline
TAKT time TAKT Available time / Customer demand

15
Typical sequence to building a VSM

Identify Customers
Information flows
& value Information flows inform each process what to do next. Examples include
schedules, priorities and forecasts
Transport
Material flow data
Information on whether the product whether in raw material, finished goods
Identify activities or part-completed is pushed or pulled through the value stream.
Supplier and Timeline

supplier info How long the product has taken to journey through the value stream.
Information to be added to the VSM includes available time, production rate
Transport from
supplier per day, processing time for each process and inventory wait time
Data collection
Information flows
Material flow data
Timeline
16
8
Shingo model
The key principal of the Shingo model is:
Analyse Separately, Design as One
Separately analyse the Design a process solution

activity of the product, that integrates the
staff and equipment to elements
locate the waste in a
process
17
Activity of the product (AoP)

We start with AoP as up to 80% of the 50 – 80% of the
improvement opportunity lies with what improvement
opportunity exists here
happens to the product:
Activity of the Equipment AoE 1 AoE 3
(AoE)
Prepare Pre-Sales Collect Gain Set-up Fund Complete

Prepare Pre-Sales Collect Gain Set-up Fund Complete
Package
PreparePackage
Pre-Sales Information
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
PreparePackage
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
Prepare Package
Collect Approvals
Gain Account
Set-up Account
Fund Account
Complete
Information Approvals Account Account Account
Activity of the Product Prepare
Prepare
Package
PreparePackage
Pre-Sales
Pre-Sales
Collect
Collect
Information
Collect
Gain
Approvals
Gain
Approvals
Gain
Set-up
Account
Set-up
Account
Set-up
Fund
Account
Fund
Account
Fund
Complete
Account
Complete
Account
Complete
Package Information Approvals Account Account Account
(AoP) Prepare Pre-Sales
PreparePackage
Pre-Sales
Collect
Package Collect Information Gain
Information
Gain
Approvals Set-up
Approvals
Set-up
Account
Account
Fund Complete
Account Complete
Fund
Account
Account
Account
Package Information Approvals Account Account Account*
Activity of Staff (AoS) AoS 1 AoS 2 AoS 3 AoS 4 AoS 5 AoS 6
The product flow (AoP) reveals the overall level of waste in the
process. Its journey is impacted by both the people and equipment.
18
9
The four possible activities of a product
There are four activities that can happen to a product at

as it moves from one end of the process to the other. A
product can be:
Stored (Non-Value-Adding)
Transported (Non-Value-Adding)
Inspected (Non-Value-Adding)
The
Product Processed (Value-Adding or
Non-Value-Adding)
How did we define value-adding activity?

19
The structure of AoP template
The AoP analysis worksheet consists of three sections:
Section 1 Section 3 Section 2

Describes what Lists and categorises the Lists the cycle times for
happens to the average cycle time for each the activity, and displays
product activity of the product descriptive statistics
(stored, transported, 20
inspected or processed) LSS Green Belt MEASURE v2.6
10
The Structure of AoP Template, Continued
The second tab of the template summarises the AoP data in

tabular and graphical formats and will provide you with a
baseline of the value add time in the process
The tabular format
lists the total
processing time
for each activity
category and its
percentage of
contribution to the
total lead time for
the product.
21
Why and when spaghetti diagrams are used in the step
Spaghetti diagrams depict the physical travel or movement

of a product, helping you identify waste across the
production environment whether that be an office or
manufacturing facility
The AoP study “quantifies”
the travel or movement of the product.
A Spaghetti Diagram shows a product’s travel
distance and pattern as it flows through a
single cycle of the process.
Use a Spaghetti Diagram when the product
moves physically through the process.
Spaghetti Diagrams can also track staff movement.
22
11
Quantify the current inventory levels throughout the
process
Inventory is:
The raw materials, work in progress and
finished goods in a process that are not
yet delivered to the customer
Inventory can exist in a physical or virtual state
What problems might occur my holding too much inventory?

How much inventory is just enough?
23
How inventory is identified and quantified
Identifying and quantifying inventory is

straightforward.
As you follow the product through the process,

locate and count each inventory type:
Raw materials inventory (before and after each
process step)
Work-in-progress (the number of products at
each process step not fully transformed into
the end product or service)
Finished goods inventory (any completed
products that are being stored or held)
The count of each inventory type is recorded

on a Inventory Map.
24
12
How inventory is identified and quantified, Continued
Virtual inventory is more difficult to identify and quantify.
Virtual inventory exists in the form of

electronic files.
For example, a list of pending applications
stored on a computer server
Another example are IT change requests
which are work in progress virtual inventory.
Review the information systems used
at each process step for virtual inventory
25
Inventory maps
An Inventory Map graphically depicts the volume of

inventory in the process at a given point in time.
The inventory is shown on a

physical layout of the workplace.
Icons indicate where and how much
inventory exists at the given point in time.
“One Scalable Icons

Size Fits
All”
Icon 17 25 5 1
26
13
Baseline Rolled Throughput Yield (RTY)
Yield is the percentage of products that are judged as

“good” from the total number of products processed.
The “product” is the reference product from the Value Stream Map.
Number of “Good Products”

Yield = x 100
Number of Products Processed
The yield of a process is the yield of the process steps multiplied together
27
How Yield is calculated
100 Products Number of “Good Products”

Yield =
Number of Products
Product Flow Processed
Step 1 Yield1 = 100 / 100 = 1
100 The traditional yield

calculation does not
3 Scrapped Step 2 7 Reworked Yield2 = 97 / 100 = 0.97 account for defects
fixed through rework!
97
3 Scrapped Step 3 10 Reworked Yield3 = 94 / 97 = 0.97

The effort to fix defects in
a process is often called
94 the “hidden factory”.
2 Scrapped Step 4 5 Reworked Yield4 = 92 / 94 = 0.98
Process Yield =
Process Yield =
92 Good 1(0.97)(0.97)(0.97)(0.98) = 0.92
Products
1(0.97)(0.97)(0.98)=0.92
or 92%
After Rework Or 92%
28
14
RTY is the preferred measure of product quality
The Rolled Throughput Yield (RTY) reveals the rework that

occurs in a process (the hidden factory).
RTY is the probability that a single unit can pass
through a series of process steps free of scrap or
rework
Number of Products Processed – (Number of Products

Scrapped and Reworked)
RTY =
It is the product of the yields of the individual steps of a process in

sequential order.
29
How RTY is calculated

Number of Products Processed – (Number of Products
100 Products Scrapped and Reworked)
RTY =
Product Flow
Step 1 RTY1 = (100 – 0) / 100 = 1
100
3 Scrapped Step 2 7 Reworked RTY2 = (100 – (7+3)) / 100 =

90/100 = 0.90
97
3 Scrapped Step 3 10 Reworked RTY3 = (97 – (3+10)) / 97 =

84/97 = 0.87
94
RTY4 = (94 – (2+5)) / 94 =

2 Scrapped Step 4 5 Reworked 87/94 = 0.93
92 Good RTY = 1(0.90)(0.87)(0.93)= 0.73

Products After or 73%
Rework
30
15
Definition of a process
Dictionary
A series of activities, changes or functions bringing about a result
Practical
A manner or order in which a task is carried out repeatedly
A systematic way of reaching an objective
A Process is not the same as a procedure

A procedure is usually a written set of steps / instructions which
when repeatedly followed targets consistency of the performance
of a specific function.
SOP – can refer to either Standard (or Standing) Operating
Procedures
31
Block Diagrams
A block diagram is the simplest type of map. It provides a quick and

uncomplicated view of the high level process
Only rectangles connected by lines are used in this type of map. The
rectangles represent major activities and the arrows indicate
direction of flow
Block diagrams can be used to simplify large and complex processes
32
16
Example block diagram
Establish Obtain Issue

Perform Develop
System Vendor Construction
Calculations Drawing
Requirements Information Package
Block Diagrams
Normally contain only rectangles
Help show process scope
Identify major steps and flow
Contain 5-7 Blocks; fit on one page
Block Diagrams are typically used for presentations or

management overviews
They do not provide the detail needed for diagnostics or
problem solving 33
The block diagram scopes the detailed process map
The block diagram can provide context:
Invoice Initial follow- Collection

Litigation Sale of debt
customer up letters
While the detailed process map can be used to break these steps down further
34
17
The levels of process detail
The detailed process map breaks down the sub-processes in the block diagram:
Block Diagram
Sub-process Sub -process Sub-process
Invoice 1. Initial follow 2. Collection

3. Litigation Sale of debt
customer up Letters
Detailed
Process Map
Pass account 3.1 Check
Funds 3.4 Issue
to litigation customer Paid? Sell debt
available? litigation letter
department details
3.2 Charge off

debt A/c up to date
35
Symbols used in detailed process maps

All Process Maps use symbols to depict the flow.
Symbol Description
Start and End Symbol
Action
Decision Point
Direction of Flow
36
18
Identifying boundaries for your process
Boundaries define the start and end points of the process:
Start End
Pass account 3.1 Check

Funds 3.4 Issue
to litigation customer Paid? Sell debt
available? litigation letter
department details
No
3.2 Charge off

debt A/c up to date
37
Functional process maps
Process maps can be built in a functional (swim lane)

format:
Functional (Swim Lane) Process Map
Organization 1 Organization 2 Organization 3
No
Yes
No
No
Yes
38
19
Components of a detailed process map
Detailed Process Maps build on the process flow by

depicting input variables, outputs, specifications and/or
targets and key metrics. Outputs from
the process
actions (the
y little y’s)
Approved Loan
with Payment
Amount
Determine Loan
x’s Payment Amount
SOP–Loan Target
Application Payment Amount
Input
N–Loan Amount Determined within 1
variables for
the process N–Interest Rate Day
actions C–Length of Specifications Key metrics,
from Receipt of
(the x’s) Loan and/or targets such as time,
Application cost, yield, CTQ
C–Approver for the input
variables and requirements,
outputs etc.
39
Steps to develop a detailed process map

1. Determine the format for the Detailed Process Map eg vertical or horizontal
2. Identify the flow of steps and decisions within each sub-process.
3. List the outputs for each step.
4. List the inputs for each step.
5. Classify the input variables as Noise, Controllable or Standard Operating
Procedure.
6. List the current specifications and targets for the process inputs and outputs.
7. Evaluate the Detailed Process Map to ensure that all criteria have been met.
Your detailed process map will be complete when you can answer the following two
questions adequately:
Who is responsible for doing the work to complete each process step and what
activities do they perform to complete that work?
How does the product or service go from initial input to final output?
40
20
Example of a detailed process map
Customer Completed
Inquiry Call
Request and Yes

Greet Inquiry
Obtain
Customer Resolved?
Information
C: VRU C: VRU
N: Call Volume SOP: VRU scripting No -- route call to agent
SOP: VRU Menu
N: Customer issue
N: Customer information Greet/Verify Resolve
N: Call Volume Customer Inquiry
C: staff
C: staff
SOP: Telephone
SOP: Telephone
SOP: Call Centre System
SOP: Call Centre System
SOP: Computer
SOP: Personal Computer
N: Customer issue
N: Customer issue
C: Call Centre staffing
SOP: VRU routing algorithm
C: Call Centre staffing
Survey
Customer
SOP: Automated Survey System
41
Class Activity: Mapping Exercise
Objective: to practise mapping
Team A: draft a value stream map for a process

you know well highlighting what information
you’d collect to complete it
Team B: create a Block diagram and swim lane
detailed process map for one of the following;
Time - 45 minutes
(incl 10 min de-brief) Your project
The process to book on this course
42
21
Summary of key learning points
VSM’s are the key maps to understand the whole of the process.
They include how the materials, information or product flows
through the process. [“Cradle to Grave” or “Soup to Nuts”]
Capture key metrics such using Activity of the Product, Spaghetti

Diagrams and Inventory Maps as you go
Block diagrams are used to summarise a large process.
Process Mapping (flowcharting) is used to drill down into a process,

allowing it to be analysed for interactions etc. This takes time and
therefore can be costly so care should be taken to ensure that the
scope is carefully defined
43
MEASURE
Y = f(x)
22
Step 2: Identify potential root causes

mapping planning
variation cause

product MSA

FMEA Centre

Spread
Analyse MSA
Normality
results
Graphs for
comparison 45
In this module we’ll discuss how to identify potential root

causes using the following techniques:
Cause & Effect diagram
Cause & Effect matrix
FMEA
46
23
What is Y = f(x)?
The “Y” in this simple formula relates to the output of a
process:
At a point in time
Over a period of time
At a point in the process
Overall for a set of process steps
Examples include:
Total time required to complete all process steps (lead time)
Percentage of deliveries achieved according to customer demand
Percentage of pizzas delivered within 15 minutes of order 47

What is Y = f(x)?
The “x” is the input upstream process indicator
In-process measures
Upstream from the customer of the process
Taken at key points in a process to assess performance and
intervene before customers are impacted
Knowing how these inputs (Xs) drive the process output (Y), allows
us to predict and control process performance.
For this reason, the process output (Y) is sometimes referred to as
the dependent variable while the input (X) is referred to as the
independent variable
48
24
Developing outcome indicators
Customer Requirement = Outcome Indicator

Helpful guidelines:
Measure data at intervals
Describe how to measure
Measure defects/non-conformance
Avoid using averages alone
Understand customer specifications
49
Avoid using averages alone as

outcome (Y) indicators
Pipe Spool Deliveries
(What is the average delivery time of each supplier?)
Shipment 1 Shipment 2 Shipment 3
Fabricator 1 8 weeks 8 weeks 8 weeks
Fabricator 2 4 weeks 6 weeks 14 weeks
50
25
Understand Customer Specifications
Customer Customer
Outcome (Y) Indicators
Valid Requirements Specification(s)
Y1 = % pizzas delivered
>30 min.
30 minutes or less from
Fast Delivery
order placement to delivery
Y2 = Actual pizza delivery
time
>10 min. from "out-of-
oven"
Hot Pizza At least 140 F
Y4 = Actual time from "out-
of-oven" to delivery
51

Consumer Cashier Cook Delivery
Hungry
Ordering
Call in
Order
Take
Order
Make
Pizza Y1 = % Pizzas delivered
Making/Baking
>30 minutes
Cook Y2 = Actual delivery
Pizza
time
No
Correct? >10 minutes from
Yes “out of the oven.”
Y4 = Actual time from
Delivering
Deliver
Pizza
“out-of-oven” to
Y1
customer.
Eat Y4 Y3 Y2 52
26
Identifying the vital few inputs (x’s)
Cup of coffee = f (coffee beans; water)
Identify on the map where the

measures will be applied
Document
CREDIT AGENCY
Documents/day
What are the factors?
Couriers
3/ day
Y1
1/jour
X1 – coffee beans
TIMES
8:00, 12:00
& 3:00 Telephone
requests
Confirmation
of
X2 – qty of coffee
Document (urgent) acceptance
X3 – hardness of water

faxes of file
8H 4H 4H 2H 1H 4H 4H
MAIL INTERNAL PREPARE VERIFY INDICATE VERIFY INTERNAL MAIL
X4 – pH levels of water

CENTRE MAIL DOCUMENTS CREDIT EXPIRY DATE DOCUMENTS MAIL CENTRE
Jean Deborah x2
Susan/Catherine John/Jeanne Susan/Catherine John/Jeanne Deborah Jean
Sort Files Sort by

client /5secs
Prepare file
45 secs
verify docs
10secs/doc
verify docs
10secs/doc x3
Check /3min Sorti /30s Post
File /30s
X5 – Chlorine content of water

Insert in
Rack /5secs Date Stamp X reference Assign Assign
2 secs file /4min insurance /2min insurance /2min
Insert in X despatch Estimate Estimate
x1 Folder /10secs file /10min credit /2min credit /2min
Etc, etc, etc
53
Identifying the vital few inputs

Our process mapping will have identified a number of input variables but not
all of these will have an impact on the process output
We use one of the following narrowing tools to answer the question, “Which
are the potential vital few inputs (x’s)?”
Cause-Effect Diagram (Fishbone Diagram)

Cause-Effect Matrix (C&E Matrix)
Failure Mode Effects Analysis (FMEA)
The potential vital few inputs (x’s):
Are the short list of x’s that the team identified is believed to have a critical
influence or effect on the Y
Will be statistically confirmed in the Analyse phase
54
27
Overview of the C&E Diagram
The C&E Diagram:
Is a visual brainstorming tool to identify potential causes for your
problem
Also known as a fishbone or ishikawa diagram
Process/Procedures People Measurement System
Problem or Y
Equipment/Machines Environment Materials
55
Prioritising the causes

Methods for prioritising causes:
Categorise as:
High or low control to make or influence change for the identified
cause
High or low impact upon the stated effect
56
28
Prioritise causes based upon impact to the project
Priority Matrix
High Consider
Cumulative Start Focus
Impact Here
High Control, High Control,
Control
Low Impact High Impact
See
Don’t Bother
Champion
Low
Low Control, Low Control,

Low Impact HIgh Impact
Low High
Impact
57
Steps to identify the potential vital few inputs (x’s) with a

C&E diagram
1. Record the Y for the project in the head of the diagram, stated as a
problem or effect.
2. Determine the categories for the causes on the spines of the diagram
(e.g., consider People, Materials, Measurement Systems,
Equipment/Machines, Process/Procedures, Environment)
3. List the causes and attach them to the appropriate spines.
4. Identify the potential vital few causes by using a Priority Matrix.
5. Align the potential vital few causes with the inputs from the Process Map.
58
29
Example of a Cause-Effect diagram
Process People
Workplace Organization Tenure
Marketing Events Technical Training
Visual Managment Typing Speed
Routing Algorithm Computer Skills
Fax Locations Spirit Training
Statement Cycles Coaching Frequency
Reference Preference Hiring Score
(Manual vs. Web)
VRU Scripting Sales Training
Policies & Procedures Customer IQ
Computer Desktop Soft Skills
Soft Skills Training
Traiing
Customer Service Experience Speed of Answer
is Too Slow
Systems hardware
Infrastructure Capability Monthly Econ. Cycles Previous Customer Experiences
Switch Algoritms
Systems availability Day of the week Time of Day
Network Stability
Network Speed Customer Mood/Attitude Long Term Economic Cycles
Computer Desktop Economic Events
World Events
Natural Disasters
Systems Environment
59
Example of a C&E diagram, Continued

Priority Matrix
H
Systems availability Network stability Typing Speed
Previous customer experience Technical training Computer skills
Customer service experience Customer IQ Coaching Frequency

Fax locations Visual management
Control
Workplace organisation Computer desktop
Natural disasters Economic events Time of day Day of week
Switch algorithms Network speed

Sales training World events
Customer mood/attitude Systems hardware Infrastructure capability Tenure
Statement cycles VRU scripting

Hiring score Soft skills training
Marketing events Spirit training Routing algorithm Reference preference
L
L Impact H
60
30
Overview of the C&E matrix
The C&E Matrix
The C&E Matrix generates an individual score for each of the identified
inputs (x’s) on the Process Map.
The score defines the collective impact of the identified inputs (x’s) upon
all key customer CTQ requirements for the project.
The potential vital few inputs (x’s) are the inputs on the matrix with the
highest scores.
Output Variables
Metric 1
Metric 2
(Y's)
Output
Input Variables (X's) Ranking
10 8 Score % Rank
Input 1 3 3 54 7.89%
Input 2 9 90 13.16%
Input 3 3 9 102 14.91%
Input 4 3 1 38 5.56%
Input 5 3 30 4.39%
Input 6 1 3 34 4.97%
Input 7 0 0.00%
Input 8 3 30 4.39%
61
Steps to identify the potential vital few

inputs (x’s) with a C&E matrix
1. List the project metrics.
2. Prioritise each metric.
3. List the inputs (x’s) in the rows of the matrix.
4. Assign a correlation rating for each x and metric pairing.
5. Cross-multiply the priority and correlation ratings to obtain the score for
each x.
6. Identify the potential vital few inputs based on the score
62
31
Example of a C&E Matrix
63
Identifying potential vital few inputs
Identifying the potential vital few

Objective: To use narrowing tools
Based on the process of arriving to class on day 1:
• Choose to utilise either the C&E Diagram and Priority
Matrix; or the C&E Matrix to identify the potential
vital few inputs
Time - 30 minutes • Discover the potential vital few inputs (x’s) for the
following problem:
Why do we arrive to class late?
64
32
Overview of FMEA
Failure Mode Effects Analysis (FMEA)
An FMEA is a systematic way to evaluate and prioritise the potential failures

of a process to prevent them from occurring.
Failure Modes The things that go wrong (or fail) at a process step
Effects The impact of the failure mode upon the customer
Causes The sources of variation with the inputs that result in

the failure mode
Analysis FMEA prioritises the x’s results based on three
factors:
• Severity of the effect on the customer
• Occurrence of the cause
• Detection or prevention of the cause or failure
65
The FMEA template
66
33
Steps to identify potential vital few inputs (x’s) with FMEA
1. List the process steps in the Process Function column of the FMEA template.
2. Enter the potential failures for each process step in the Potential Failure Modes
column.
How can this process step fail?
3. Enter the effects on the customer of a failure in the Potential Failure Effects
column.
4. Enter the possible causes for the failure in the Potential Causes of Failure
column.
5. Enter the methods and systems used to currently detect the failure in the
Current Process Controls column.
What do we have in place to either detect or prevent the cause or
failure mode so that the effect does not occur?
6. Assign a rating to each effect by using a scale from 1–5 in the Severity
(SEV) column.
The higher the severity rating, the greater negative impact to the customer.
67
Steps to identify potential vital few inputs (x’s) with FMEA,

Continued
7. Assign a rating to each potential x by using a scale from 1–5 in the

Frequency of Occurrence (OCC) Column.
The higher the occurrence rating, the more frequently the cause occurs and will
result in a failure mode.
8. Assign a rating to each potential x by using a scale from 1–5 in the
Likelihood of Detection (DET) Column.
The higher the detection rating, the more likely the customer will experience
the effect.
9. Calculate the Risk Priority Number (RPN) for each potential x.
RPN = SEV x OCC x DET
10. Identify the vital few inputs (x’s) for validation in the Analyse
phase.
Prioritise by: • High-occurrence and low-
• High RPN Value detection ratings
• High Severity Ratings
68
34
Example of an FMEA
69
Analysing the FMEA
RPN
70
60
50
40
30
20
10
0
Associate Slow typing skills Associate Associate does not Customer does not Call Center Associate's Call Center Switch overloaded
inexperience inexperience with understand Call provide sufficient System is "slow" telephone System is down
system center System information malfunctions
70
35
Building an FMEA
Failure Mode Effects Analysis

Objective: To prepare an FMEA
Based on the process of arriving to class on day 1:
• Create an FMEA and be prepared to present back to
your classmates.
Time - 15 minutes
71
Guidelines on when to use the narrowing tools

The C&E Diagram
Advantages Limitation
• It is quick and relatively simple to • Only one CTQ can be evaluated with
complete. one fishbone.
• It organises the causes into
categories.
72
36
Advantages and limitations of narrowing tools, Continued
The C&E Matrix
Advantages Limitation
• It quantifies the influence of each x • Does not consider level of control
on the project metrics.
• It uses the inputs from the Process
Map.
• It can evaluate multiple CTQs.
73
Advantages and limitations of narrowing tools, Continued
The FMEA
Advantages Limitations
• There is more rigour in identifying • It is time consuming to complete.
and analysing inputs in a FMEA
when compared to a C&E Matrix
and a C&E Diagram
• It is process focused. The flow of
the tool follows the flow of the
process.
• It can evaluate multiple CTQs.
74
37
After mapping out the processes
Step 1: Identify the key output metrics (Y’s)

Step 2: Use the appropriate tool to identify the potential
few inputs (x’s)
C&E Diagram
C&E Matrix
FMEA
75
The identification and

measurement of your process
outputs
The type of narrowing tools used
and the rationale for their
selection
The potential vital few inputs for
your project Y
76
38
MEASURE
Basic Statistics
Step 3: Basic statistics

mapping planning
variation cause

product MSA

FMEA Centre

Spread
Analyse MSA
Normality
results
Graphs for
comparison 78
39
During this lesson we will cover the following basic

concepts of:
Determining the type of data

Assessing the shape of the data
Measures of central location
Measure the spread
Assess the normality of data
Creating graphs for comparison
79
Types of Data
80
40
Continuous Data
It is measurable data.
It is data that can theoretically be subdivided and measured in
smaller and smaller units.
Based on required precision, we decide the number of decimal
places.
Small amount of data is required to draw valid conclusions
Examples:
The length of a cable
Thickness of a wire
Downtime of a LAN server
The average repair time of a machine
81
Discrete data
It is countable, indivisible data

Whole numbers only—no decimal places
Usually related to defects or proportions – after the error has had
an effect
Illustrates the trends only over a longer time period
Large samples sizes are required in order to draw valid conclusions
Examples:
The number of times a schedule date is missed
The number of people absent
The number of customers in various market units
Percent of products defective
82
41
Data Classification
Continuous or discrete?
1.The frequency of failure of a machine
(number of times)
2. The number of hours an installed circuit
board lasts
3. The number of damaged telephone
circuits in a business
4. The number of undelivered bills per Time - 5 minutes
month
5. The daily fuel consumption of repair
trucks
6. The dollar differences from work order
budgets and work order actuals
7. The percentage (%) of phone calls
greater than 30 seconds
8. Shoe size? 83
Continuum of data possibilities

It is most desirable to be able to measure your Y on a continuous
variable scale.
Other types of data can certainly be analysed. As you move from
the ideal, more samples will be required to show and detect
improvement.
The data at your disposal can be in many forms: Less Desirable
Binary – classified into one of two categories

Unordered categories - no rankings
Ordered categories – rankings/ratings
Count – counted discretely
Continuous – on a continuous scale
More Desirable
84
42
Descriptive statistics
Speed of Call Answer

Glasgow Centre
30
Shape of the data (Bell-Shaped or 25

skewed)
20
Frequency
Centre of the data (mean, median or 15
mode)
10
Spread of the data (range, standard 5
deviation or variance)
0
7.50 8.25 9.00 9.75 10.50 11.25 12.00 12.75
Time (secs)
85
Assessing shape - Histogram

What does the shape of this Histogram suggest?
Speed of Call Answer

Glasgow Centre
30
25
20
Frequency
15
10
0
7.50 8.25 9.00 9.75 10.50 11.25 12.00 12.75
Time (secs)
86
43
Parts of a Histogram
A histogram is made up of three components:

Vertical or Y-Axis:
Indicates the scale for the frequency of the
bars Speed of Call Answer
Glasgow Centre
30
Horizontal or X-Axis: 25
20
The scale of values into which the data
Frequency
values fit 15
10
Data values grouped into intervals
5
0
7.50 8.25 9.00 9.75 10.50 11.25 12.00 12.75
Bars: Time (secs)
Denote frequency of the data within the

grouped intervals
Provide indication of the shape 87
Advantages and limitations of using a histogram

Using a Histogram
• A Histogram displays data in an • A Histogram cannot distinguish
easy to interpret graph. exact values; it creates intervals or
• It shows the general shape of data. “buckets” from the data points.
• You can place specification limits on • It is not meaningful for small data
a Histogram to show what portion of sets; at least 30 measurements are
the data that is within requirements. needed.
• It will obscure any time differences
within your data sets.
Use Histograms to answer the following questions:

What is the shape of my continuous data?
Does my process perform to specifications?
88
88
44
Measures of the central location
The three measures of centre:
Mean – the average of the data

Median – the middle point in the
data
Mode – the most frequently
occurring value in the data
89
Formula for calculating mean

Calculating mean:
Steps for calculating mean:

1. Calculate the sum of all the data.
2. Divide by the number of data points to calculate the mean.
x Sample mean also

known as x-bar
 Population mean
90
45
Calculating the median
What is the median?
Is the middle value of the ordered observations

Half of the measurements are greater than the median and half are
smaller than the median
~
x Sample median
Steps to calculate median:

 Population median
1. Order the numbers from smallest to largest.
2. If the data set includes an odd number of data points, choose the point
that is exactly in the middle.
3. If the data set contains an even number of data points, choose the two
numbers that are in the middle and average them.
91
The impact of shape upon centre

The Histogram below plots the wait time, in seconds, at a
hotel reception desk.
The mean wait
time is 100
seconds. The
median wait time
is 74 seconds.
Which one should we

use?
Based on the shape of the data, what might we conclude

about these wait times?
92
46
The impact of shape upon centre, Continued
Another wait time example:
This Histogram represents the

wait times for another hotel. The
mean wait time is 200.1 seconds
and the median is 200.1 seconds.
There is no/little difference

between the two statistics for
centre of the data.
Why is this?
93
Guidelines for use of mean and median
Mean or Median?
If the data is bell-shaped, use If the data is skewed, use the

the mean to describe the centre median to describe the centre of
of the data. the data.
94
47
Mode
Mode is the value in the data that occurs with the

greatest frequency.
Uniform Normal
Skewed Bimodal
95
Spread of data - variation
Runway (Manchester Airport)

05
R
Touchdown target
How consistently do the pilots perform?
We want to be able to measure the

performance and use it to compare
the quality of pilots.
If we were to select a pilot for a new
route to SABA – how do we compare
their performance? 96
48
Measures of spread
Range – difference between the

largest and smallest values in the
data A
1 2 3 4 5
Variance – how far each data

point is from the mean B 1 2 3 4 5
Standard deviation – square root

of the variance C
1 2 3 4 5
97
Range
Range = x Largest –x smallest
98
49
Limitations of Range as a measure of spread
What are the ranges for data sets A, B and C?
A
1 2 3 4 5
B 1 2 3 4 5
C
1 2 3 4 5
Range doesn’t account for each point in the data; it only

evaluates the extremes.
99
Variance and Standard Deviation
Variance does account for every point in the data.
Variance evaluates the squared difference of every point in the data from the
mean.
Variance is not practical since it would be in squared units.
To remove a squared term, take the square root.

The square root of the variance is the standard deviation.
We can use improvement software such as Minitab to calculate variance or
we can calculate it manually!
100
50
Variation - Variance
Variance is a way of describing how much the data
is dispersed around the mean.
As the values have been Squared – the units stay in
the same units, but now a “Squared” unit.
S2 =
1
Mean
Max So VARIANCE is the average of the squared

differences from the mean
s 2 for Population
Variance
for Sample
S2 Variance LSS Green Belt MEASURE v2.6
101
Standard Deviation
The trouble with the VARIANCE – is that the units
don’t really make sense. If we are working with
time (seconds) – what exactly does seconds2 mean?
Our final step is to calculate the STANDARD

DEVIATION. This is a way of describing how much
of the data is dispersed around the mean. As we
squared the values to get the VARIANCE – we apply
the Square Root to ensure the answer is in the
Standard
deviation same units that we started with.
The Point
of
inflection:
where the
High Standard Deviation
curve goes
from
Concave to
Convex
S S
Mean
102
Low Standard Deviation LSS Green Belt MEASURE v2.6
51
Lets work through an example….
A B C
A
1 2 3 4 5
Mean 3 3 3
B 1 2 3 4 5
Median 3 3 3
Range 0 4 4
C
1 2 3 4 5
Variance 0 2 2.5
Standard 0 1.414 1.581
Deviation
103
Have a go!!!
For the data sets shown

calculate:
1. Mean
2. Median
3. Mode
4. Range
5. Variance
6. Standard deviation
104
52
Normal Distribution Curve
Characteristics of the Normal Curve:

• It is symmetrical.
• The mean, median and
mode have the same
value.
• There is an equal
number of values on
either side of the mean.
105
Normal Curve properties
-1σ 68.26% +1σ

-2σ 95.46% +2σ
-3σ 99.73% +3σ
-4σ 99.9937% +4σ
-5σ 99.999943% +5σ
-6σ 99.9999998% +6σ
106
% of the data that falls within +/- x standard deviations from
LSS the mean
Green Belt MEASURE v2.6
53
Normal Distribution example
Handle time (seconds)
Text Text Text Text Text Text
68%
95%
99.73%
35 40 45 50 55 60 65
We have an average handle time of 50 seconds and a standard deviation of

5. If the distribution of the handle times is normally distributed, then we
should expect to see:
• 68% of our calls between 45 and 55 seconds
• 95% of our calls between 40 and 60
• 99.73% of our calls between 35 and 65 seconds
If you were the call centre manager, how could you use this information?
107
Graphs for comparison - Boxplots
Components of the Boxplot

Highest Value
25% of
the
data
3rd Quartile
(75%)
Median 50% of
the
Mean
data
1st Quartile
(25%)
25% of
the
data
Lowest Value
The Boxplot shows the range values and gives a simple

picture of the variation
Note the distribution in not symmetricalLSS Green Belt MEASURE v2.6
108
54
Boxplots for comparison
Boxplots are frequently used to compare one distribution to

another.
They are especially useful for identifying differences in median, and
spread as well as for highlighting the presence of outliers
Boxplot of Normal, LSkew, LongTail, Flat, RSkew, Bimodal

100
80
What can we say about
60
the following boxplots of
airport queue time?
Data
40
20
Normal LSkew LongTail Flat RSkew Bimodal

109
Using Boxplots
• Provide a general indication of • Can be misleading for data sets
centre and spread with less than 10 values
• Identify potential outliers that are
present
• Ideal for comparing different
groups of data
Use Boxplots to:

• Compare certain characteristics of the data
• Plot continuous data or to compare a continuous Y for discrete x’s
110
55
Scatter Plot
A Scatter Plot illustrates the relationship between two continuous
variables.
Components Scatterplot of Minutes vs Units
1. Vertical or Y-Axis 120
2. Horizontal or X-Axis:
3. Data Points 100
80
Minutes
Tests the relationship
between a continuous Y 60
and a continuous x
Displays the direction (as x 40
increases, does Y increase

or decrease?), shape 20
0 1 2 3 4 5 6 7 8
(linear or nonlinear) and Units
strength of the relationship
111
Using a Scatter Plot
 Show a relationship between two • Do not show the time order in
variables which data was collected
 Can be used for continuous data
or discrete count data
 Display the direction and strength
of the relationship between two
factors
Use Scatter Plots to:

• Examine how two continuous variables are related (indicate strength, shape
and direction).
112
56
Understanding data basics:

• Determine the data type.
• Assess the shape, centre and spread of the data.
• Assess normality
• Create graphs for comparison.
113
When your data is not normally

distributed
Comparing sets of data with small sample
sizes
Drawing conclusions from your graphical
analysis
114
57
MEASURE
Measurement Systems
Analysis
Step 4: Assess measurement systems

mapping planning
variation cause

product MSA

FMEA Centre

Spread
Analyse MSA
Normality
results
Graphs for
comparison 116
58
During this lesson we will cover the following basic

concepts of:
Why analysis of measurement

systems is important
Types of measurement system
variation
How to create an attribute study
How to analyse the results
117
The additive nature of process and measurement

variation
Total
True Process + Measurement = Variation
Variation Variation (Observed)
(s2 True Process ) (s2 Measurement ) (s2 Total)
What You Need to What Masks Your What You See

Understand to Understanding of the Through Your
Improve Performance True Process Measurements
Variation
118
59
Measure measurement variation first
Why quantify measurement variation first?

Total
True Process + Measurement = Variation
Variation Variation (Observed)
(s2 True Process ) (s2 Measurement ) (s2 Total)
What You Need to What Masks Your What You See

Understand to Understanding of the Through Your
Improve Performance True Process Measurements
Variation
Difficult to determine the real problem and its cause.

Faulty measurement system produces data leading to bad decisions.
Bad decisions waste time and money.
119
Total observed variation

Observed variation is made up of two components.
Which process:
1. Has the largest total observed variation?
2. Has the largest measurement variation?
3. Has the largest process variation?
4. If you were to choose which process was in better shape statistically, which one
would you choose? Why?
There are four sources of measurement system variation 120

60
Sources of measurement variation
Measurement Variation
Accuracy
Precision
Repeatability
Reproducibility
Stability
Linearity
121
Definitions: Sources of measurement variation
Accuracy Accuracy
Precision
Centering of measurement data around a known
standard Repeatability
Reproducibility
Stability
150 Linearity
100 True Value
50
Left Right
It is a measurement of how close the average of multiple

measurements of an event is equal to the true value.
122
61
Definitions: Sources of measurement variation, Continued
Precision Accuracy
Precision
Distribution of measurements around the average
Repeatability
Reproducibility
Stability
150
Linearity
100 True Value
50
Left Right
Precision tells how clustered the individual measurements are around

the target value.
123
Definitions: Sources of measurement variation,

Continued
Repeatability Accuracy
…measures the extent to which the same Precision

person gets the same measurement results Repeatability
using the same measurement system. Reproducibility
Stability
Linearity
To determine if someone can get the same

measurements by using the same
measurement system repeatedly, we
collect data repeatedly.
Since the event does not change, any

change in the measurements must be due
to changes in the measurement system.
124
62
Continued
Repeatability
Accuracy
…measures the extent to which the same person gets
the same measurement results using the same Precision
measurement system. Repeatability
Reproducibility
Reproducibility
…measures how several people or systems repeatedly Stability
measure the same event and we look for differences in Linearity
the results among the people or systems.
… quantifies the differences in results between the

appraisers
125

Continued
Stability Accuracy
...is the capacity of your measurement Precision

system to produce the same measurement Repeatability
results over time when measuring the same
Reproducibility
event or item
Stability
Example Linearity
The tendency of measurements to
differ depending on the time of day or
the length of time the appraiser is
working or measuring.
126
63
Continued
Linearity Accuracy
… measures how well your system Precision
maintains its performance over its range of
Repeatability
measurement values (for continuous data)
or measurement categories (for discrete Reproducibility
data) Stability
Examples Linearity
Can a scale measure with the same
accuracy and precision the weight of a
mouse versus the weight of an
elephant?
Are we more prone to accuracy and
precision issues when we evaluate
applications with 10 fields versus
applications with 100 fields?
127
Causes of measurement variation

Draw on our personal work experience and as a class brainstorm
five causes for poor:
• Accuracy
• Repeatability
• Reproducibility
Use flipchart paper and be prepared to

present back
Time - 20 minutes
128
64
What is Measurement Systems Analysis?
Measurement System Analysis (MSA) is:

An analytical procedure to determine how much of the total
observed variation comes from your measurement system
There are two different types of MSA:

• Discrete
• Continuous
129
Discrete MSA overview
Outcomes • Measurement variation calculated in terms of

accuracy, repeatability and reproducibility
Limitations • Discrete measurements do not describe the
extent to which an item falls in its category
(e.g. how good or bad the item is)
Common Findings • Problems with standards or operational
definitions (vague, not understood or applied
correctly by appraisers)
Frequency of Use • More common than Continuous MSA
130
65
Continuous MSA overview
Outcomes • Repeatability and Reproducibility

Limitations • Not as applicable for service industry
measurement systems
Common Findings • Problems with the design of the
measurement devices
• Problems with the use of the measurement
devices
Frequency of Use • Less common than Discrete MSA particularly
within an office environment
131
Recap class activity

Instructions:
Answer the following questions
Time - 10 minutes
Difference?
1. What term is used to characterise the
difference between the true value True Value (or Average of its Actual
Standard) Measurements
(or standard) of a measured item and
the average of its actual
measurement results?
2. Which has better accuracy, the plot 150
on the left or right? 100 True

Value
3. Which has better precision, the plot
50
on the left or right? Left Right
132
66
Questions
4. Which source of measurement

variation are we checking for in
this example?
5. Which source of measurement

variation are we checking for in
this example?
133
The Discrete MSA procedure
1. Confirm the key discrete measures that require an MSA.

2. Confirm the operational definition for each measure.
3. Design the discrete MSA study.
4. Develop data-collection sheets for the discrete MSA
study.
134
67
Confirm the measures that require an MSA
Conducting an MSA procedure can be lengthy and

resource-intensive.
Therefore, the first step is to review the various
measurement activities within your
project/process and determine which ones truly
require an MSA.
Consider the following criteria:
• Significant Risks: In other words, major costs

or problems will occur if the measurements are
subject to errors
• High Levels of Measurement Error: Your

observations and experience tell you that the
measurement system is not at an acceptable
level of accuracy, repeatability or
reproducibility.
135
Operational definitions
The operational definition describes:
The measure being analysed

The items that will be measured in the study
The categorisations used for the measure
How the standard for each item will be determined
136
68
Design the Discrete MSA study
Key design considerations for a discrete MSA:
The number and • Recommend a minimum of 30 items

type of items to be • Try to distribute items equally among
measured categories
The number of • Typically three appraisers
appraisers • Use more appraisers if issues of reproducibility
are suspected
The number of • Two is sufficient
measurement trials
The organisation • Randomise trials to prevent memorising
of the • Avoid situations where appraisers may be
measurement trials influenced
137
Develop data collection sheets for the Discrete MSA study
Data collection sheets:
Define the order in which the

items will be presented to each
appraiser
Provide structure for the study
Are used to record the actual
measurement results during the
study
Are provided to each appraiser as
separate sheets
We’re now ready to conduct a study!

138
69
Conduct the Discrete MSA Study
Key things to remember:
Randomise the presentation order.

Maintain consistent measurement
methods.
Code the entry results for discrete
data to simplify data recording.
Example: P = pass, F = fail
139
Discrete MSA Practice Activity

The vowels exercise
Objective: To illustrate the use of a discrete MSA
10 images will be displayed on the screen at the front of the
classroom.
As you view each image, determine whether the image passes or
fails.
A screen passes if it contains no vowels. It fails (is defective) if it
contains a vowel.
30 mins
Record your measurement results in the first trial column of the
Data-Collection Sheet shown on the following slide.
Record your findings quickly since a new image will be displayed
on the screen every four seconds.
After we complete the first trial, there will be a second trial, and
you will record your answers on the following slide.
The findings from some of the class will be used to asses the
measurement system.
140
70
Data recording sheet 1
141
Data recording sheet 2
142
71
Data recording sheet - combined
143
The “Standard”
144
72
Analyse our discrete MSA results
Our assessment of the measurement system should review

the between (reproducibility) and within (repeatability)
sources of variation.
The following “rules of thumb” should be used when assessing the adequacy of our
measurement system
• % Study Variation Criteria:
 Good: <10%
 Marginal: 10% to 30%
 Bad: >30%
• Divide the points by the total number of observations.

• Math Breakdown: 1 – (22/30) = 1 - .7333 = 0.2667 or 26.67% Study Variation.
• How good is the MSA? What do we need to fix?
145
How did we do?
From our 10 samples, how many did

we get complete agreement
(reproducibility and repeatability)?
What is our measurement system
variation?
Improving the measurement system: 20 mins

Identify and implement the
appropriate solutions to reduce the
level of measurement variation.
Repeat the MSA study to validate
the impact of any solutions to
reduce the level of measurement
variation.
146
73
Measurement variation can negatively impact your
decision making.
The sources of measurement variation:
Accuracy
Precision
Repeatability
Reproducibility
Linearity
Stability
MSA is a procedure to quantify the level and source of
measurement variation
There are two different MSA procedures:
Discrete MSA
Continuous MSA 147
Designing the MSA Studies

If you have continuous data
Analysis and interpretation of results
Diagnosis of root-level causes for
excessive measurement variation
148
74
MEASURE
Process stability and
capability
Step 5: Stability and capability

Value stream Sampling
mapping strategy
variation cause
Activity of the Planning an Data collection

C&E matrix Shape Control charts
product MSA planning
Block Operational Calculating

FMEA Centre Sample size
diagrams definitions capability

Spread
Analyse MSA
Normality
results
Graphs for
comparison 150
75
In this module we will learn how to

“Check” the stability of data before using it to assess
the inherent performance of a process
“Compare” the capability of process performance
against the desired performance
You’re at a point in DMAIC where you need an accurate

and predictable base line measure of your problem by
using the project metrics.
The metric data needs to be assessed for its stability.

If the data is influenced by special causes or unusual
circumstances, your base line measure will be
inaccurate.
151
Stability - Introduction
What is process stability?
Process stability is the ability of the process to perform in a

predictable manner over time
A project metric is stable when it varies in a predictable manner over
time or is in a state of statistical control.
In a Lean Six Sigma project, stability is determined by:
Evaluating the project’s primary/secondary metrics in the Measure
phase
Managing the confirmed vital few inputs (x’s) of a process in the
Control phase
Stability is determined by examining whether common or special causes
of variation are present.
Stable processes experience only common causes of variation.
Unstable processes experience special causes of variation.
152
76
Understanding Common and Special Causes of
Variation
Variation can be common cause or special cause.
Common Causes of  Are present in all processes

Variation  Are comprised of the variation inherent to the
(systematic, random, combination of all process elements (people,
normal, in-control or equipment, environment, methods and materials)
expected, natural)  Produce random, predictable fluctuations in the x or
project metrics over time
When only common cause variation is present, the
process is stable and predictable.
Special Causes  Are large or unusual differences in variation due to
some “hiccup” in the process
(abnormal, non-
random, out-of-control When special cause variation is present, the process is
or unexpected) unstable and unpredictable.
153
The two causes of variation
How you interpret variation …
Common Cause Special Cause
Focus on step Type 1

Common
changes to Tampering
Cause
processes (increases variation)
True
variation
type …
Special Type 2
Investigate causes
Cause Under-reacting
154
77
Introduction to Control Charts
What are control charts and

why are they useful?
Display a project metric in the order it

is collected over time
Are used to determine whether a
process measure is stable (in a state of
statistical control)
155
The two categories of Control Charts
Control Charts can be used for continuous or discrete data.
Category Type of Charts Displays

Continuous I-MR Individuals and Moving Range
Xbar and R Xbar and Range
Xbar and S Xbar and Standard Deviation
Discrete P Proportion defective
NP Number defective
U Defects per unit
C Number of defects
156
78
The components of a Control Chart
All Control Charts have three components:

Performance over time
A centre line
Control limits (LCL and UCL) Data Points
Upper Control
Limit (UCL)
Over Time.
Center Line
Lower Control
Limit (LCL)
157
Understanding Control Limits, Continued
How control limits work:

0 .0
0 .2
0 .3
0 .4
0 .1
• We should expect to see

all of our data points (99
to 100%) fall between
m-3s
the control limits with no

m-2s
patterns.
• If a data point falls
m-1s
outside the limits, we

99.73%
95%
68%
m
detect patterns or both,

m+1s
these are signals of out-

of-control or special
m+2s
cause conditions.
m+3s
158
79
Control limits are not specification limits
Control limits should not be confused with specification limits:
• Control limits represent the Voice of the Process (VOP).

They tell you how the process is expected to perform when no
special causes of variation are present.
• Specification limits are the Voice of the Customer (VOC).

They tell you what your customers (internal or external) want
from the process.
159
The two indicators of a special cause of variation
1. Any point outside a control

I Chart limit T
for Answer
70 1
60 1
UCL=56.18
Individual Value
50
40
30 Mean=30.91
20
10
LCL=5.640
0
0 50 100 150
Observation Number
160
80
The two indicators of a special cause of variation
Continued
2. A non-random pattern of data points within the control

limits
Shifts Trends
A series of A series of data
points above points that
or below the continuously
centre line increase or
decrease
0 10 20 30 40 50 0 10 20 30 40 50
Cycling or
Alternating
Non-random
patterns seen in
the data points
0 10 20 30 40 50 0 10 20 30 40 50
161
Applications of Control Charting in DMAIC
The two primary applications for Control Charts in DMAIC
Phase Application
Measure Phase • To assess the nature of the process

variation – are the metrics’
performance over time stable?
Control Phase • To control the vital few inputs (x’s)
• Control Charts are an integral part
of the control plan
162
81
Selecting the best control chart for the Measure Phase
How do I choose the correct chart for my project?
Questions to ask:
Is my data discrete or continuous?
If the data is discrete, do I want to monitor the number of defects or
the number of defective items?
Defect: A flaw (not meeting customer specifications)
Defective: A product or service with one or more defect
Control Charts used to assess stability of the primary metric
Type of Data Measurement Chart

Continuous Continuous Value I-MR Chart
Discrete Defects C Chart
Defective Items NP Chart
163
Structure of the Individuals and Moving Range Chart
The Individuals-Moving Range (I-MR) Chart is comprised

of two charts:
Individuals Chart monitors the
individual observations over time.
Moving Range Chart monitors the
change of the individual
observations over time.
Is this process stable or not?

Why?
What is the appropriate next
step?
164
82
The C Chart
The C Chart is used when your metric involves a count of the

number of defects:
Is this process stable or

not?
Why?
What is the appropriate
next step?
165
The NP Chart
The NP Chart is used to plot the number of defective

items/units for a constant subgroup size.
Differs from the C Chart:
• The C Chart plots the number
of defects (flaws).
• The NP Chart plots the
number of defectives (an item
or unit that contains one or
more defects).
Is this process stable or
not?
Why?
What is the appropriate
next step?
166
83
The importance of identifying special causes
Special cause variation indicates a need to investigate:

How you interpret variation …
Points outside of the control limits Common Cause Special Cause
Trends Type 1
Common Focus on systemic
Tampering
Cause process change
Shifts (increases variation)
True
Cycles variation
type …
Special Type 2 Investigate special
Alternating patterns Cause Under-reacting causes
M-L5-37
Identify when the special cause occurred and identify the factors
that contributed to it.
Don’t quantify process capability before understanding whether

the process is statistically stable. 167
Definition of Process Capability
Process capability is the ability of a stable process to

meet customer requirements.
We can use capability analysis to compare the

performance of very different processes.
168
84
Picturing capability
A universal business metric that can be used to
indicate the capability of the process (VOP) to
meet an output specification (VOC) s1
Voice of the Process
It is the number of standard deviations that can fit LSL USL

between the mean and the nearest specification
Performance that is outside of the specification is Voice of the Customer

a “defect” – not meeting the customer
specifications
LSL USL
LSL USL s1
Defects
(Out of specification)
Process Capability
169
The first measure of capability: Sigma Level
Sigma Level is a measure of how Sigma Yield Percent

well a process performs relative Level Defective
to its requirements. 1 30.9% 69%
The higher the Sigma Level, the 2 69.1% 31%
more capable the process. 3 93.3% 6.7%
The appropriate Sigma Level 4 99.38% .62%
depends on the importance and 5 99.977% .023%
cost of poor quality for the 6 99.99966% .00034%
characteristic.
The higher the sigma level, the more capable the process.
The appropriate sigma level depends on the importance and cost of

poor quality for the characteristic.
170
85
The second measure of capability: DPM
We tend to use DPM when dealing with discrete measures of capability.
The relationship between DPM and Sigma Level:
Sigma DPM
Level
1 691,462
2 308,538
3 66,807
4 6,210
5 233
6 3.4
DPM is the number of defectives in every million units.
Note that it is “defectives per million” rather than “defects

per million.”
171
Calculating sigma with continuous data

First we calculate the Z-score for the process, then add 1.5 sigma shift.
The “+1.5” represents short-term and long-term views of the process.
Average Handle Time

Average handle time = 50 seconds
Standard deviation = 5 seconds
15 Data is normally distributed.
Customer requirement = 57 seconds.
Frequency
10
What is the sigma level for 57

5
seconds?
0
Value of Interest  Mean

40 50 57 60
Z 
Standard Deviation
Z  1.5  Sigma
Note that the z-score can only be used when we have a normal distribution,
172
otherwise we need to use another distribution or DPMOLSS Green Belt MEASURE v2.6
86
The Sigma Calculator
The Sigma calculator quickly determines the Sigma Level

and DPM for your data.
173
Steps to calculate sigma level for continuous data
Three Requirements
1. Data must be normally

distributed.
2. The process must be

stable.
3. You must have a valid

measurement system.
174
87
Steps to calculate sigma level for continuous data
Steps to use the Sigma Calculator for continuous data:
1. Enter the average (x bar).
2. Enter the standard deviation (s).
3. Enter the specification

limits for the project metric.
4. The calculator displays these

values automatically.
175
Calculating sigma with discrete data
Defectives per million opportunities (DPMO) can be used to quantify

capability for discrete output metrics
DPMO = D X 1,000,000
N
When:
D= total number of defective units in the sample evaluated: a

defective unit does not meet the customer specification
N= Number of units evaluated
176
88
Steps to calculate sigma level and DPM for discrete data
Steps to use the Sigma Calculator for discrete data:
Calculating Sigma Level for Discrete Data

1. Enter the number of
Enter Number of Items Evaluated 30
items in the sample.
Enter Number of Defective Items 2
2. Enter the number of
This is your Sigma Level 3.00
defective items.
This is your Yield 93.33%
This is your DPM 66,667 3. The calculator
displays these values
automatically.
177
Table exercise: Quantify process capability

Capability Analysis
Objective: Practise using the sigma calculator to calculate the
sigma level and DPMO for the following examples:
1. Time to process mobile phone application:
− Historical average is seven minutes (420 seconds)
− Historical standard deviation is 1.5 minutes (90 seconds)
− Upper specification limit is 10 minutes (600 seconds)
− There is no lower specification limit
Time - 5 mins
2. Percentage of transactions completed within seven minutes
at banking centres:
− Total number in sample = 1,000 transactions
− # defective from sample = 73 transactions during 7 minutes
3. Lead time to issue orders:
− Historical average is 20 days
− Standard deviation is 10 days
− Upper specification limit is 25 days
− There is no lower specification limit 178
89
• Assessing the stability of the metrics using

control charts driven by the type of data and
the metric you want to control
• Remove special causes but avoid tampering
by reacting to common cause variation
• Determining the capability of the process by
reference to customer requirements LSL USL
s1
Process Capability
179
Choosing the appropriate control chart
Sampling strategy for control charts
Analysis and interpretation of results
Determining the best approach when

continuous data is not normal
Interpreting sigma level and DPM data

relative to the goals for your project
metrics
180
90
MEASURE
Data collection and sampling
Step 6: Collect data

mapping planning
variation cause

product MSA

FMEA Centre

Spread
Analyse MSA
Normality
results
Graphs for
comparison 182
91
Preview of lesson
A Lean Six Sigma project team working in a 24 hour car

manufacturing facility has identified one of the project
metrics as end-to-end lead time.
The team wants to create a baseline measure for the

metric before beginning its analysis.
What decisions would the team need to make to ensure

that the data it collects is useful and reliable?
183
Preview of lesson
In this module we’ll discuss the need for data within Lean
Six Sigma projects as well as:
Determine what to measure
Determine sample size
Select the appropriate sampling

strategy
184
92
Data collection overview
The need for data

Teams need data to speak with facts
“In God we trust, everybody else brings data”
Data is the “facts or figures from which conclusions can be

drawn”
Information collected about a product, service, process, person or machine is
called data
Data, when properly organised and analysed, provides useful information and
serves as the basis of decision making and action
185
Data collection for hypothesis testing
In addition to quantifying the process baseline, we gather data to

hypothesize on the drivers of process performance (Y)
Y= f(x1; x2; x3…..xn

Examples include;
Shift Location on Unit Valve Model #
Time of Day Inside/Outside Instrument Tag #
Day of Week Left Side/Right Side Equipment #
Week of Month Top Side/Bottom Side Cable #
Season of Year Leading Edge/Trailing Edge Raceway #
Employee Orientation Design
Name Temperature Revision #
Grade Humidity Supplier
Training
Experience
186
93
Data Collection
Error
Process Process Process Process
Customer
Step A Step B Step C Step D
Factors Factors Factors Factors = Outcome
x1 x2, x3 x4 x5, x6 = Y
Where did the error occur? (which process step)

Could we have caught it before the customer “found” it?
How much “wasted” effort did we spend working on the item that was
erroneous?
How much will this dissatisfaction cost us in the long term?
The outcome is a function of the factors within the process
187
Components of a typical data collection plan
2 3 4 5 6 7 8 9 10
The Purpose for The Data to The Who, How, Where and When
Collecting the Data Collect
4 Who will collect the data? 9 Strategies to
1 The question(s) you want to 2 How/where 5 How/where recorded? eliminate bias
answer recorded
6 Related conditions to
3 Measure type/data record 10 How the data will
type be displayed
7 Frequency/timing
8 Sampling notes
188
94
Finding the right data
Hrs from receipt to delivery x1

x2
Cycle time Day of Week
x3
y1 Month
Start time
x4
On time
delivery Available Hours x5
Quality
Score x6
Stage in process (%, steps)
Processing
Y # Errors x7
errors Team
y2 x8
Individual
System x9
% downtime
Errors
189
Example completed data collection plan for contact centre

PREPARED BY Purpose of Data Collection DATE CREATED DATE REVISED
Jake Calc Baseline credit card operation performance for service level, 1st call resolution, and handle times 5/29/2006
DATA Who, How, Where, and When
Sampling Notes
Related Frequency /
(any conditions or
How/where conditions to Timing. How How will the data be displayed? What
Measure Type / Who will collect characteristics How will you
What recorded? record ie. Time of often and at what graphical tools will be utilized to summarize
Data Type data? about the sample minimize bias?
(attach form) Day, month, shift, intervals will the the data collected so it can be analyzed
that should be
location etc data be collected
noted)
Random
sampling will
% calls Jake C using Excel report from Randomly sample
Date and time Sample 100 calls provide Data will be used for stability and capability
answered within Discrete: binary automated automated the 100 calls for
Call volume every 2 hours representative studies
30 seconds system system each time period
sample for this
metric
SMEs will
Total sample size
evaluate each
of 369 stratified
call. Prior to
based on call
evaluating, all
Date and time type:
Q&P will randomly SMEs will review Data will be used for baseline capability for
% first call Record results on Associate 185 Balance
Discrete: binary Sally S and Q&P record calls based the established secondary metric. Data can support pareto
resolution attached form Call type Inquiry
on VRU responses standard for 1st analysis
Call volume 111 Change of
call resolution.
address
MSA will be
19 Drop terms
performed for
56 Drop rate
each SME
Total sample size
of 369 stratified
based on call
Stratfied random
Date and time type: Data will be used for baseline capability for
Will use the same sampling will
Record results on Associate 185 Balance secondary metric. Performace by call type
Call handle time Continuous Sally S and Q&P samples for % first provide a
attached form Call type Inquiry and associate will be plotted using boxplots
call resolution. representative
Call volume 111 Change of &/or dotplots
sample
address
19 Drop terms 190
56 Drop rate LSS Green Belt MEASURE v2.6
95
Sampling requirements
There are two basic needs for data:

Estimation
• Your focus is to collect enough representative data to give you a “good”

estimate of the parameters of the process. “Good” is defined by confidence
level and margin of error.
• If historical data is not available, a data collection plan should be created to
collect the appropriate data.
• The rest of this module deals with sampling for estimation.
Hypothesis Testing
• Your focus here is to collect data on the Y and different Xs to determine if
there are differences in how the Xs affect the Y, or if different values or levels
of an X affect the Y.
• Here again we need enough representative data to be able to prove the
significance of the effects of the Xs.
• We will cover sampling for testing in the Analyse phase
191
How much data to collect
Descriptive Statistics
Descriptive Statistics Characterises
Shape
Centre
Spread
Population Data Summarises
Percentile
Proportions
Rankings
Sample Inferential
tist ial
Statistics –
Data
Sta rent
ics
typically used to
infer information
e
Inf
about populations
from sample data
192
96
The concepts of populations and samples
Population Sample
Collection of all objects or individuals Collection of measurements

with at least one common characteristic selected from the population
193
Population Parameters vs. Sample Statistics
Population Sample
Data describing the whole of the Data describing a sample of the data
population are known as PARAMETERS
An entire group of objects that have been are known as STATISTICS
made or will be made containing a The group of objects actually
characteristic of interest. measured in a statistical study.
“Population Parameters” “Sample Statistics”
 = Population Mean
x= Sample Mean
s = Population Standard Deviation
s = Sample Standard Deviation

X
s
s
194
97
The value of sampling
Why do Lean Six Sigma projects use samples instead of

populations?
Costs less to collect a sample than an entire population

Requires less time to collect than a population
Enables you to collect data when it is impossible or impractical to
evaluate the entire population
195
Populations vs. Samples
You are looking at the effectiveness of online orders and you have
access to every online transaction for the last six months.
Do you have a population or sample? Why?
You need to estimate the average income of football season ticket

holders in Glasgow, UK?
What would be the population in this example?
What would be a sample?
Would you collect population or sample data? Why?
196
98
Sampling can be applied to populations and
processes
Sampling from populations

A relatively fixed, static base
Typical for surveys or baseline studies at one specific point in time
Examples: registered voters and employee satisfaction surveys
Sampling from processes

A dynamic flowing base
Typical for studies evaluating performance over time
Examples: Sampling in-process applications to determine time it
takes to perform a process step or sampling process bake time within
a galvanising plant
197
Factors affecting sample size

Four factors determine how much data is needed:
Factor Effect
Data Type Discrete data requires larger sample sizes
than continuous data.
Required Confidence Level The required sample size increases as the
confidence level increases.
Margin of Error The required sample size increases as the
margin of error decreases.
Variation in the Population Variation has an impact on the sample size.
or Process The sample size increases as the variation
increases.
What factor might you think is missing?

198
99
Example for factors impacting sample size
We have a survey measuring job satisfaction specifying

a 95% confidence level and with a margin of error of
±3%.
The survey results indicated an satisfaction rate of 65%.
This means we have a 95% chance that the true
population satisfaction rate falls between 62% and 68%.
• What is the data type?

• What type of sampling are we using (population or process)?
199
Using the sample size calculator

Continuous Data
2 1. Select the Sample Size
4
Continuous tab.
2. Enter the estimated
standard deviation for the
process or population.
3. Establish the acceptable
3 margin of error for your
5
sample.
4. Determine the required
confidence level for your
sample results.
5. Locate the sample size
requirement for the
confidence level and
6. margin of error.
Determine if the sample
size is feasible.
200
100
Using the sample size calculator, continued
Discrete Data
1. Click the Sample Size
Discrete tab.
2
2. Enter the estimated
4 defective proportion for the
population.
3. Establish the acceptable
margin of error for your
sample.
3 5 4. Determine the required
confidence level for your
sample results.
5. Locate the sample size
requirement for the
confidence level and margin
of error.
6. Determine if the sample size
is feasible, given project
restraints.
201
Class activity: using the sample size calculator

Sample size
Objective: How to use the sample size calculator to quantify
data collection efforts
For the following two scenarios:
Which type of data is being collected?
How confident do you need to be?
What is the margin of error?
Time - 10 mins What is the estimated standard deviation for the process?
What is the required sample size from the Sample Size
Calculator?
Scenario 1: You need to estimate the average Scenario 2: You need to estimate the percent
call length in the call centre. of calls in the call centre that were within the
Typical call length is from 30 seconds to three call-handling specification of three minutes.
minutes. The standard deviation is The estimate should be accurate within + or -
30 seconds. A confidence level of 95% is 2%. You believe that 90% of the calls meet the
required. call handling specification. A confidence level
of 95% is required.
202
101
Is sample size all that matters?
The 1936 Presidential Election:

The Story of Alf Landon
Source Sample Predicted Winner

Size
Literary Review 2 million Alf Landon
George Gallup 2,500 Franklin D.

Roosevelt
How could a sample of 2 million voters be less

accurate than a sample of 2,500 voters?
203
The four methods to select your sample data

Sampling strategies
Sampling from a Sampling from a

Population Process
Random Sampling X
Stratified Random X
Sampling
Systematic Sampling X X
Rational Subgrouping X
204
102
Random Sampling
Population Sample
Application Population sampling
XXXXXXXXXX
XXXXXXXXXX XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
Easy to understand and
XXXXXXXXXX
Advantages explain
XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
Not an appropriate
XXXXXXXXXX strategy for process
sampling
Every member in the population If there are segments
Limitations
has an equal chance of being within the population, they
selected. may not be represented in
the random sample
205
205
Stratified Random Sampling

Population
Type of Call Units Sample

Population sampling
Balance Inquiry XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
Application
XXXXXXXXXX
XXXXXXXXXX XXXXXX
XXXXXXXXXX
XXXXXXXXXX
All segments within the
Change of Address
XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX Advantages population are represented
Drop Terms
XXX X Not an appropriate strategy
for process sampling
Drop Rate
XXXXXXXXXX
XXXXX Limitations
XXXXX
Population is divided into distinct

segments (or stratified layers)
and then a random sample is
taken within each segment
206
103
Systematic sampling
Population or process
Process Sample Application sampling
XXXXXXXXXXXXX XX
Can maintain the time

order of the data for
Sampling frequency is based
upon a specified number of
Advantages process studies
Easy to explain
observations or units.
For example, we will select every

Potential for bias based on
5th survey for detailed review in a Limitations the systematic scheme.
customer satisfaction survey
207
Rational subgrouping
Process Process sampling

Application (Statistical Process Control)
Maintains the time order, so

you can observe trends
Provides insight as to
Advantages
sources of variation in the
process
More complicated to design

Sub-grouping applies to how and conduct the rational
many samples we need to Limitations subgrouping sampling
collect at each defined time strategy
interval (frequency).
208
104
Class activity: sample strategy
Sample strategy
Objective: How to adapt how you select the calculated
sample size based on the nature of the population
In the slide that follows answer the questions posed and be
prepared to share your thoughts with your classmates
Time - 5 mins
209
Scenario: Sampling Strategy

A call centre has had issues with meeting the customer requirements for speed
of answering calls. The call centre team decided to take a sample of its calls to
estimate the percentage of calls that were within the call handling
specification of three minutes (primary metric) as well as the average call
length (secondary metric).
Four types of calls are handled by the call centre:
Balance Inquiry (50%)

Change of Address (30%)
Drop Terms (5%)
Drop Rate (15%)
1. Which strategy would you recommend to collect the data?

2. How would you change the strategy if the team is interested in how the
call length performs over time?
210
105
Every piece of data, potentially costs

time and money to collect, analyse and
to store. It is important that the
benefit of collecting data outweighs the
cost
Having a solid data collection plan is
vital
Collecting samples of data can save
time and money
We need to be aware of sampling risks
and ensure these are considered when
choosing sample size and sampling
strategy
211
When determining the appropriate

sample size when there is a small
population size
Calculating sample size when the

current proportion defective is
unknown
Choosing your sampling strategy
212
106
Key points of the Measure phase
Use mapping techniques to help understand

the current process
Identify potential root causes
Understand the basics of data
Baseline the performance of your
the process relative to customer
requirements
Create and execute a data collection plan
ready to move into the Analyse phase
Ensure appropriate care is taken when
calculating sample size and selecting
sampling strategy
213
Lean Six Sigma simulation: Round 3
The coin simulation

Objective: Use the tools covered in this module to prepare
The current state value stream map for the process.

Highlight the location of any potential opportunity area
Re-run the simulation implementing 1 more change.
60 mins
214
107
ANALYSE
Lean Six Sigma
Green Belt
At the end of the module, you will be able to:

Create a future state process map
Quantify demand and capacity for the
process
Analyse the key equipment and activity
constraints
Perform correlation and regression analysis
Understand basic hypothesis testing
2
LSS Green Belt ANALYZE v2.6
1
Analyse Phase overview
Create the Create

Activity of Load and
future state Demand Correlation hypothesis
equipment Balance
map tests
Prioritise
Changeover Alpha and
improvement Takt time Pull systems Regression
analysis Beta risks
efforts
Interpret test
Capacity Visual Mgt
results
Activity of
Layout
staff
3
ANALYSE
Future state value stream
mapping
2
Future State VSM
Create the Load and Create

Activity of
future state Demand Balance Correlation hypothesis
equipment
map charts tests
Prioritise
improvement Capacity Pull systems Regression
analysis Beta risks
efforts
Activity of Interpret test

Visual
staff results
Management
Layout
5
Creating the future state
Having mapped the existing process as part of the Measure phase we’re
now going to create the future state Value Stream Map (VSM).
A future state VSM should be made up of value-added activities that

produce:
What the customer wants

How much the customer wants
When they want it
Where they want it
A lean value stream has the shortest lead time, highest quality and
lowest cost possible
6
3
Approach
Ideal state
What does
ideal look
like?
Future state
How close
can we get
to ideal?
Quick wins
What quick
wins can we
put in now? 7
Current state LSS Green Belt ANALYZE v2.6
Ideal State VSM
An ideal state value stream map is a

theoretical process created free of any
constraints on time, money or even
current technical feasibility.
So why create it?
One of the principals of Lean is continuous

improvement.
The creation of an ideal state value stream map ensures, that after
delivery of the project solutions, the Process Owner is challenged to
strive closer to this vision avoiding stagnation.
Creating the ideal map will also help those taking part to think in a
creative and innovative way about how value could be delivered.
8
4
Future State VSM
The Future State VSM is created by tackling
some of the problems currently faced by
the Current State while aiming as close to
the Ideal State VSM as possible.
Future State 4
As we solve the problems of the current
state, a number of iterations of future Future State 3
state could be created.
Future State 2
The remaining sections of Analyse training
is focused on finding the root causes of Future State 1
these issues so that solutions can be
generated to reduce or eliminate these
root causes.
9
Prioritising improvement efforts

The amount of
change required to
move from the
current state to
future state is
normally too great
to be done in one
step.
Breaking the value stream up into project loops makes change more
digestible.
But where should we start?
10
5
Prioritising improvement efforts, continued
There is no prescribed rule for prioritising which loop should be worked on

first but the following should be considered:
Begin as close to the customer as possible so improvements are visible
to the customer sooner
Focus on the pacemaker process or cell, since this will set the pace to
takt (see next module) for the full value stream
Work on the processes within your control before branching out to
suppliers and customers; your internal work may change requirements
Learn as you go and apply what you learn; feel free to adjust the
future state design as you learn more
Ask which loop has the best chance of making the greatest
improvement considering the customer and business?
11
Quick Wins
Just mapping the value stream alone may yield some obvious
improvement opportunities. These opportunities are sometimes
referred to as “low hanging fruit” or “quick wins”.
Teams should be prepared to
pursue these opportunities – the
return on investment can be very
high.
Typical criteria for quick wins are:
Easy to implement (no significant co-
ordination or planning required)
Easily reversed if not successful
Fast to implement
Cheap to implement
Within the teams control 12
6
Create an Ideal State to give the team something to aim for and
avoid stagnation after the project solutions are implemented
The Future State map is created based on what is realistically

achievable by tackling the existing issues with the current
process
Create project loops to make the scale of change more digestible

for the business and easier to manage for you
Seek to identify and implement quick wins, you can get an early
return on investment and start the momentum for change.
13
ANALYSE
Analyse time and work
7
Analyse time and work
Create the Load and

Activity of
future state Demand Balance Correlation
equipment
map charts
Prioritise
Changeover
analysis
efforts
Activity of
Visual
staff
Management
Layout
15
In this module we’ll discuss:
Quantifying customer demand
Takt time
Quantifying capacity to meet demand
Standard work
16
8
Customer demand
A customer demand requirement specifies:
• What …
• How much …
… your customer base requires
• When …
• Where …
17
There are three different types of demand

Value demand is the type of
demand process owners want
placed on their processes. It’s
what the process was set up to
deliver.
Examples include production orders in a
manufacturing facility or customer deposits
at a bank.
Failure demand is considered Moveable demand is

“bad demand” as you want to demand that can be placed
minimise the resources spent on an alternative process
dealing with it. for the benefit of the
Customer Complaints and many types of business and/or customer.
customer inbound calls represent The growth in online customer
significant investment by organisations servicing is a great example of how
to deal with this type of demand this area has grown
When mapping processes be aware of the “hidden factory” – processes set up

18
to deal with errors in the standard process (failure demand).
9
Identifying types of customer demand
For the following examples, identify the demand being

placed on each process
Classify the demand as: 1) value 2) failure or 3) moveable
demand
For the value and failure demand, identify opportunities to
convert the demand to an alternative process that is more
effective or efficient.
5 minutes For each opportunity, describe how the customer and/or

organisation handling the demand could benefit.
1. Coffee/tea served at a staff restaurant

2. Ordering a new mobile phone
3. Logging a software fault
4. Requesting a replacement credit card
5. A quality inspection function at the end of a manufacturing line
19
Quantifying demand
Quantifying the demand on resources requires us to:
Identify the different demands
Create an operational definition of each demand and

time interval
Determine the size of the time interval between each

demand measure
Quantify how much data is to be gathered
Collect the data
20
10
Quantifying demand
Customer demand is unlikely to occur in an even pattern.
Understanding when peaks and troughs occur is important in the
management of resource.
We need an appropriate increment of time to acquire each demand
measure
An interval too large or too narrow hides the true pattern and
variability of demand.
Talk to people who work in the process, the Process Owner and your
Coach to determine:
- The current volume of demand
- How demand varies over time
- The size of a time interval (such as hours, days or weeks) to
depict the variability in demand
- The relationship of cycle time to customer lead time
21
Determining how much data to gather

We need to collect enough data to capture a minimum of one demand
cycle.
• Customer demand is often
cyclical.
• A demand cycle is the time
before each pattern begins to
repeat itself.
Demand Cycles
• Knowing whether the demand data occurs in cycles and the length of
each cycle requires historical data and/or experience with the
process and its past demand patterns.
• If no historical data exists, talk to the Process Owner, process Subject
Matter Experts (SMEs) and/or your Coach to decide on the
appropriate number of samples. LSS Green Belt ANALYZE v2.6
22
11
Demand volume is derived from historical data and
forecast volumes
Historical view Forecast
Analysis of historical Analysis of demand

demand up to and forecasts and any
including current other factors which
demand have arisen since
publication of the
forecast
There is no guarantee that your estimate for demand will be

accurate but taking a combination of historical and forecast
demand volumes is a good place to start 23
Takt Time
Takt comes from the German word for rhythm.

It can be thought of as the rate the process has to run at to keep up
with customer demand.
24
12
Takt Time (cont’d)
We calculate the takt time as follows:
Available Time
Takt Time =
Demand Requirement
Available time to run the process doesn’t include breaks, lunch etc. It’s
the amount of actual time available to produce output.
In order to increase takt time (slow production rate), what could we

do with breaks, lunch etc?
Takt time does not account for variation in demand which may be
caused by changes to:
customer demand requirements
the hours of operation
25
Takt Time v Cycle Time

We look to synchronise takt time and cycle time
Cycle time is a rate measure quantifying how long a process

step takes to produce one unit of output
If the cycle
time is: Then:
Less than the Work is being completed faster than required. The
Takt time waste of overproduction, inventory or waiting will occur.
Greater than The process is unable to meet the pace of customer
the Takt time demand.
26
13
Takt time
For the following example,

1. Calculate the takt time of the process
2. Calculate the process cycle time
3. Advise on next steps
5 minutes Work hours are 09:00 to 17:00 with one hour for lunch
Customer demand is 200 units per day
Output rate is 220 units per day
27
Capacity
Customer demand and process capacity should match like

two pieces of a puzzle.
• Customer Demand – Defines Customer Process
how much is needed Demand Capacity
• Process Capacity – Defines

how many can be produced
Once both are quantified, they are compared

to acquire a baseline of how the capacity differs
from the demand.
28
14
Approaches to quantify capacity
Capacity is expressed based on the degree to which resources are

dedicated to the process.
Processes with dedicated resources:
Capacity is expressed in terms of the ability to process units in a given period of time
(3,135 applications per day)
Capacity is expressed for the process as a whole
Processes with shared resources

Capacity is expressed in terms of available time
Capacity is expressed for a resource or resource group
(Six operators working between 9:15 a.m. and 9:30 a.m. on Monday provide 90 minutes
of capacity)
29
Capacity with dedicated resources
Is the average of the actual output

levels of the product/service 8:00 - 9:00
Monday
110
Tuesday
117
Wednesday
116
Thursday
107
Friday
107
collected from the process 9:00

10:00
-
-
10:00
11:00
109
111
114
109
113
107
115
112
116
112
11:00 - 12:00 105 112 117 110 115
per unit of time 12:00
1:00
-
-
1:00
2:00 114 109
Lunch
113 104 110
2:00 - 3:00 117 112 111 108 108
-
Uses the same sampling
3:00 4:00 113 113 101 106 116
4:00 - 5:00 109 109 108 117 113
5:00 - 6:00 103 110 118 80 115
approach as the Demand 6:00 - 7:00 109 110 117 109 107
Analysis
Is compared with the
average demand
level identified Capacity =
110.4 units per
from the Demand Analysis hour
30
15
A process with shared resources
The capacity is measured for each resource group:

A resource group is an employee or piece of equipment (single or group)
capable of performing similar work.
The resource group’s capacity is its “ability to do work per interval
of time.”
• A Banking centre measures capacity in 15-minute increments
• Capacity is Tellers Tellers on Tellers at Time Teller
the ability to perform Present Break Station Interval Capacity
9:00 - 9:15 5 5 X 15 min. = 75
“x” amount of teller 9:15 - 9:30 5 5 X 15 min. = 75
work per 15-minute

9:30 - 9:45 5 5 X 15 min. = 75
9:45 - 10:00 5 1 4 X 15 min. = 60
increment. 10:00 - 10:15 5 2 3 X 15 min. = 45
10:15 - 10:30 5 2 3 X 15 min. = 45
10:30 - 10:45 5 5 X 15 min. = 75
10:45 - 11:00 5 5 X 15 min. = 75
11:00 - 11:15 5 5 X 15 min. = 75
11:15 - 11:30 5 5 X 15 min. = 75
11:30 - 11:45 5 5 X 15 min. = 75
11:45 - 12:00 5 5 X 15 min. =
31 75
The four strategies to balance demand and

capacity
There are four general strategies to balance the demand and

capacity of a process:
• A Provide strategy
• A Match strategy
• An Influence strategy
• A Control strategy
You need to determine which strategy, if any, is used for your

process.
Keep in mind that a combination may be used.

32
16
The Provide Strategy
Provide sufficient capacity at all times to meet peak demand.
• Sets the capacity equal to the peak demand
• Assumes enough capacity is available to meet demand
PROVIDE Capacity Strategy
6000
Relative Demand
5000
4000
Demand
3000
Capacity
2000
1000
0
1 2 3 4 5 6
Time Periods
33
The Match Strategy
Anticipate the expected demand to match capacity to demand.

• Sometimes possible with employees
• Is difficult to accomplish with equipment and facilities
MATCH Capacity Strategy
6000
Relative Demand
5000
4000
Demand
3000
Capacity
2000
1000
0
1 2 3 4 5 6
Time Periods
34
17
The Influence Strategy
Attempts to influence customer demand patterns to obtain a good
utilisation of resources.
• Influences demand by moving demand from the highest demand
period to the lowest
• Influences lowest demand by having fixed hours of operation
• Reduces the amount of demand variation
INFLUENCE Capacity Strategy
6000
Relative Demand
5000
4000
Demand
3000
Capacity
2000
1000
0
1 2 3 4 5 6
Time Periods
35
The Control Strategy
Sets capacity to equal the average demand.

• Uses some means of accommodating (controlling) excess demand
until capacity is available
• Uses queues to accommodate excess demand
CONTROL Capacity Strategy
6000
Relative Demand
5000
4000
Demand
3000
Capacity
2000
1000
0
1 2 3 4 5 6
Time Periods
36
18
Activity of the Staff (AoS)
Activity of the staff studies require us to

1. Optimise the activity of staff engaged in
process steps that will remain in the
future state process
2. Identify and standardise the one and
best way to complete activities
3. Define the activity of staff on new
process steps introduced by the future
state design
37
Optimise the activity of staff (AoS) for the Current-State Process Steps in
the Future-State Product Flow
An activity of staff worksheet can be used to:

Identify each work element
Time each work element
Categorise activities for each element
Work element Categorise Times 38

19
How AoS studies are conducted
AoS studies are comprised of five steps:

1. Identify those employees to participate in the study
2. Determine the number of cycles and the sampling strategy for the
study
3. Prepare a data collection plan to acquire the activity times for
each step
4. Conduct the study and complete the AoS worksheet
5. Determine the baseline measures of non-value-adding time the
employee experiences.
We will look at steps 1 and 4 in more detail.
39
How AoS studies are conducted, continued
Step 1: Identify the employees to participate in

the study.
• Identify the staff position/group
performing the step.
• Include a representative sample of the
best, average and poorest performing
employees.
• All employees may have best
practices that can be carried forward.
• Secure any approvals to observe the employees.
• Communicate the purpose of the study to the affected
employees.
40
20
How AoS studies are conducted (Cont.)
Step 4: Conduct the study.

• Capture all work elements that comprise the process step
• A work element is the smallest increment of work that can be
transferred from one person to another.
• The level of detail for the activities depends on the cycle time
for the step.
• A step with a cycle time of minutes will have activities defined
in seconds.
• You may encounter “foreign activities” during the study.
• Foreign activities are not part of the designed work for the
process step (employee adjusts their glasses, equipment
failures, etc.). They are special cause variation.
• They should be captured in the study.
41
The primary methods to acquire the AoS data
The data for the AoS worksheet can be acquired by:

• Videotaping the employees performing the process
• Observing the process in-person, using a stopwatch to acquire the
time data
• Having employees self-record the AoS data
42
21
Stacked bar – current state
Creating a single stack of time ordered activities from the Activity

of Staff worksheet immediately provides a visual on potential
opportunities for improvement
This stacked bar chart

illustrates employee work
elements within a
correspondence process
43
The definition and importance of standard work
Standard work is defined as:

An agreed-upon set of work procedures that establish
the best, most reliable methods and sequence
for each step and employee in the process.
• Standard Work reduces variation in how work is performed

• Standard Work alone does not result in “work standardisation”
• Standard Work provides …
• A foundation for continual improvement
• A common process language among all employees
44
22
Use mock-ups to define employee activity in
new process steps
A mock-up is:
A model of a proposed process step or activity
in a controlled environment.
Mock-ups are used to:
Test and improve the design of a process step in a safe
environment
Enable the acquisition of cycle time data for a process step
Use mock-ups when:
The process step is new
The sequence of the process steps is new
The risk employed with the new process step is high
45
Stacked bar – optimised
Implementing standardised work processes, delivering mail

directly to the correspondence team significantly reduced
employee activity
46
23
Demand analysis is used to quantify the pull on

resources from the customer
Takt sets the rate at which output should be produced
to keep up with customer demand
Capacity is the volume of staff or equipment available
to meet demand
Conduct an Activity of Staff (AoS) study to identify
opportunities to improve and standardise work content
47
Quantify the length of a demand

cycle
Quantifying sample size for AoS
studies
Quantifying capacity for processes
with shared resources
48
24
ANALYSE
Equipment Analysis
Equipment Analysis

Activity of
equipment
map charts tests
Prioritise
analysis Beta risks
efforts

Visual
staff results
Management
Layout
50
25
Lesson objectives
Identify and classify the equipment in the process that

requires an Activity of the Equipment (AoE) study
Conduct and analyse an AoE study
Apply the ICE (Identify, Convert and Eliminate) method to
reduce equipment changeover time in a process
Develop standard work for the use of process
equipment
Select and apply the appropriate equipment optimisation
strategy to reduce the cycle time of an equipment
activity
51
Activity of the Equipment (AoE)
Equipment activities, like employee activities, include

waste.
The tools and procedures in this section will help you identify and
eliminate equipment waste
Equipment waste is important to eliminate because it:
• Results in lost employee time/costs, including injuries
• Prevents you from meeting customer demand without heroic employee
efforts
• Creates defective products, leading to inspections
• Generates a need to process in batches versus one-piece flow
• Builds up WIP inventory
• Frustrates employees and customers
52
26
Identify the equipment within the process that
require analysis
There are three general types of equipment in a process.
1. Equipment dedicated to one employee performing:
• One/multiple steps in a process
• Steps in many processes (A personal laptop or a telephone)
2. Equipment shared among multiple employees performing:
• One/multiple steps in a process (A shared copier, fax
machine)
• Steps in many processes (Shredding equipment, a shrink
wrapping machine)
3. Equipment where the customer is the operator.
• For example, petrol pump, cash machines and telephone IVRs
53
How to determine which equipment to analyse
• Identify the equipment using the Value Stream Map

• For each piece of equipment ask:
• “If the equipment was unavailable for any reason,
would it impact the ability of the process to
produce the desired results?”
• “Does this equipment affect the ability of the
process to meet its Takt time?”
• “Does the equipment negatively affect the
customers in the process?”
An AoE study is recommended if you answer

“yes” to any question
54
27
Conduct an Activity of the Equipment (AoE) study for
the selected equipment
AoE studies typically require the help of equipment SMEs.

• Most project team members lack detailed knowledge of the
equipment in the process.
• The employees who operate the equipment daily are ideal SMEs.
• Be sure to review their roles/responsibilities as a SME.
• Include their managers in any discussions of roles and
responsibilities.
55
The general activities of the equipment in a process
There are 10 general equipment activities that are

observed and timed during an AoE study
56
28
The structure of the AoE template
The AoE Analysis Worksheet has the same structure as

the AoS worksheet
57
The structure of the AoE template
Like staff studies, AoE studies should be planned

carefully and executed well.
1. Plan the AoE study with equipment operators/experts
2. Develop the data collection strategy for the AoE study
3. Analyse the AoE study to identify waste in the usage of
equipment. Identify the root causes of the waste
4. Optimise the use of the equipment at process steps
5. Develop standard work for use of the equipment
58
29
How to identify equipment monuments in a
process
A process monument is:
A resource that serves more than one process step that
cannot be moved or decentralised
due to cost, environmental or control issues
• Monuments can be equipment, facilities,

organisations or employees.
• Monuments occur when:
• Many process steps feed into one resource
• One resource feeds many process steps
• You often have to work around a monument
in the short-term until a longer-term solution
is created.
59
How are your tyre changing skills?
Assume one day next week

you get a flat tyre during
your drive home from work.
From the time you stop your

car on the shoulder of the
road, how long will it be
before you are on the road
again?
60
30
How the ICE method is used to reduce
equipment changeover time
Identify, Convert and Eliminate (ICE) is a method to reduce

the changeover time for a piece of equipment.
61
Table exercise: ICE
Objective: Understand the practical

application of the ICE method
Think of a process from your organisation in

which the changeover from one product or
service is important.
Why is it important?
Time - 15 mins
What are the high level steps and could
any of these benefit from the ICE method?
62
31
Other equipment optimisation strategies
The Lean equipment philosophy is simple:
Equipment should be available to run when needed,

but should be run only when needed
ICE
Preventative
Pull Systems
Maintenance
There are six key strategies to support

Equipment
the Lean equipment philosophy Optimization
Six Sigma
Downtime
Improvement
Logs
Methods
Poka-Yoke
Some businesses believe it is wasteful to have idle equipment

(even when there is no customer demand)! 63
An AoE study is used to quantify the level of

equipment waste
If a monument exists in the process, plan to work
around it in the short term until a longer-term
solution is created.
ICE is a method that converts set-up and changeover
activity to reduce equipment downtime
ICE is one of six methods used to get the most out of
equipment within the process
64
32
Quantifying sample sizes in AoE

study
Working with monuments in a
process
Further information on other
strategies to optimise the use of
equipment.
65
ANALYSE
Refine the Future State VSM
33
Refine the Future State VSM
Create the Create

Activity of Load and
future state Demand Correlation hypothesis
equipment Balance
map tests
Prioritise
improvement Takt time Pull systems Regression
analysis Beta risks
efforts
Interpret test
Capacity Visual Mgt
results
Activity of
Layout
staff
67
After completing this lesson, you will be able to:

• Use Load Charts and Balance Charts to balance the cycle times of
the process steps
• Design a pull system to flow the product continuously through the
process steps
• Determine where and how visual management methods should be
applied in the future-state process
• Develop a conceptual design of the physical layout of the future-
state process
68
34
How to refine the initial process design to
achieve continuous flow
A major focus of Lean is the design of a continuous flow
system for the process.
• The continuous flow system is the conductor of the process.
• It connects the “white space” between process steps, ensuring
that all steps work in unison.
• A well-designed system:
• Hinders the creation of waste in
a process (since the majority
occurs in the white space)
• Minimises interruptions to the
flow of the product White Space – a Primary Source of Waste
• Reduces the time to complete

the product
69
Balance the standard work content of the future-state

process
Work content balancing is:
A method to compare and adjust the cycle times of the individual
steps in a process so that the cycle times of all steps:
1. are similar and
2. fall just below the Takt time for the process
•Balancing is performed after you have optimised the

design of the process steps in the “Activity of ” studies
70
35
Factors to consider when structuring the standard work
content
There are many strategies to structure the work content within a
process. Two car manufacturers illustrate the extremes:
Henry Ford’s Approach Volvo’s Approach
Narrowly structure the work Broadly structure the work

content to produce a car content
• Thousands of employees • At one time, Volvo employed
were involved a production model using a
in the team of 10
production of employees to
a car build a car
What are some of the advantages and disadvantages of each

approach?
71
Factors to consider when structuring the standard work

content
Factor Does the work …

Geographical/ Need to be performed at certain physical sites or
Environmental controlled environments?
Vocational Require unique employee skills?
Lead Time Need to be performed in parallel or around the clock to
reduce lead or cycle time?
Equipment Have unique equipment requirements?
Risk Mitigation Need to be performed by different employees for a
“check and balance” within the process
Training Allow for efficient new employee ramp-up?
Job Enrichment Need to increase or decrease to mitigate employee
turnover?
72
36
How to estimate the number of employees
required for the process
The number of people required for a process can be

estimated once the Standard Work content has been
defined:
Number of people required = Sum of the work content

Takt time
* Round up to the next whole
number.
How work is assigned to these employees will depend on

the work structure and the extent to which the cycle
times balance for the process steps.
73
How Load Charts are used and constructed

A Load Chart depicts the extent to which the process
steps cycle time balances to takt time.
74
37
How balance charts are used and constructed
A Balance Chart is used to analyse and “balance” the work

content at the activity level of a process.
• The Balance Chart uses time data at the activity level for each
process step.
• Activity-level data is needed to shift the work content among the
process steps
• The activity cycle times are added to arrive at a step-level cycle
time.
75
How balance charts are used and constructed (Cont.)
76
38
How the charts are constructed for process
steps with multiple employees
A five-step process has a Takt time of 12 minutes and the cycle

times below.
77

steps with multiple employees (Cont.)
The Load Chart should display the overall cycle time of

the work content at each process step.
Load Chart 23 minutes at Step 2
30.0
25.0 2
20.0
Time
15.0
1 1 1 1
10.0
5.0
11.5 23.0 11.0 11.3 10.7
0.0
Step 1 Step 2 Step 3 Step 4 Step 5
Process Steps
78
39
steps with multiple employees (Cont.)
The Balance Chart should display the work content

assigned to each employee at each process step.
11.5 minutes for employees A

Balance Chart and B at Step 2
12.5
12.0
11.5
Time
11.0
11.5 11.5 11.5
11.3
10.5 11.0
10.7
10.0
Record issue Research issue Research issue Make correction Report back to Document
customer issue/resolution
Process Steps
79
How the charts are used for process steps

with equipment resources
Apply the following rules for process steps with equipment

resources.
Use the: To display:
Load The overall cycle time for each process step. The cycle
Chart time includes the equipment’s run time and the time to
complete any employee-performed tasks while the
equipment is not running.
Balance The equipment and employees as separate resources.
Chart This will show the elements constituting the cycle time of:
 The equipment and how it compares to the Takt time
 The employee operating the equipment and how it
compares to the Takt time
80
40
Design the Pull System to flow the product through
the process steps
As a general rule, “Flow when you can, batch when you

must.”
Batching
One-Piece Flow
Batching might be required when:
• The design of any materials, supplies or equipment hinder one-piece processing

• The product is easier to process (or monitor) in batches.
• The movement of one product at a time is cost or time prohibitive
• The suppliers of an input cannot deliver the input when needed to achieve one-
piece flow at the consuming process step
• The software used in the process automatically batches the product and changes
to the software are not possible
81
How pull systems work
Pull systems use signals to control when a product (or

batch) is processed at each step in a process.
• Pure pull systems are used for Upstream
commodity items such as cash Step 1

in the ATM channel
• In-process pull systems are used to
control the work flow of unique items Step 2
such as loan applications
Step 3
Product Flow
Pull Signal
Step 4
Downstream
82
41
The rules of a pull system
The employees or equipment at a process step should work

only when: Signal
• A signal to work is Process

received from the Input
Step
Output
downstream step
Input Status Process Status Output Status
• There is product to Not Open Cannot Work Not Open
work on from the Open Cannot Work Not Open
upstream step
Open Cannot Work Open
Not Open Work Open
In all other circumstances, the employees and equipment at

the step remain idle.
The signals are usually visual cues such as
signs, cards, flags, level indicators and electronic messages.
These signals are known as Kanban 83
The process characteristics that maximize the value of a

Pull System
A pull system is best applied to a process that is:

• Reliable. Dependable equipment, a flexible, multi-skilled
workforce and no delays due to materials movement
• Organised. Well-structured with minimal physical travel for the
product
• Repeatable. Work content performed consistently, with clearly
defined and understood Standard Work
• Balanced. Cycle times of steps are balanced within Takt time. Lot
sizes are small; ideally, the process uses one-piece flow
Pull systems are most easily applied to high volume, repetitive processes.
They can be applied to one-of-a-kind processes
(such as software or new product development).
84
42
The concept of Standard Work in Process (SWIP)
Standard Work in Process (SWIP) is:
SWIP is usually required because:

• It may not be possible to perfectly balance all steps
• The cycle times of the process steps will always vary
• Monuments can exist in some future-state designs
• Customer demand will vary
• Supplier delivery performance can vary
85
The different approaches to determine the SWIP

quantity
The trial and error approach.

• You perform a root cause analysis of the
WIP at each step:
• Every piece of inventory exists for a reason
(some are valid; most are not).
• Most processes suffer from too much
(rather than too little) WIP.
• Begin with the average WIP level at the step. Repeatedly ask, “Why does this
exist?”
• You determine a reasonable starting WIP for a trial and error process.
• You slowly remove (or add) WIP to the process step until the process flows
continuously and efficiently.
86
43
The different approaches to determine the SWIP
quantity (Cont.)
An alternative approach is to calculate the cycle stock and

safety stock requirements using statistical methods.
• The SWIP quantity is calculated using the following formula:
• SWIP Inventory = Cycle Stock Inventory + Safety Stock Inventory
Cycle Stock
How much I need to meet demand at the process step
Safety Stock
How much extra I want to cover variation
87
Visual Management
Visual management systems create the visibility to

operate and manage the process.
• They display vital process information
so you can see what’s happening allowing
you to:
• Respond to issues rapidly
• Avoid wasting time and energy trying
to figure out things
• They control or guide the actions in the process to prevent errors
or mistakes from occurring.
88
44
Where Visual Management systems are normally applied
Visual management systems are used to display.

1. Important status and failure messages to employees and
customers
2. Vital information about the performance of the process (or
process step)
3. Who is trained (or has demonstrated the ability) to complete the
steps, activities or tasks in the process
4. Samples of items with defects (or prototypes of items that are
processed correctly)
5. The Standard Work for a process step
89
The scope of Visual Management design activities

in the Analyse Phase
You create a conceptual design of any visual displays or

controls for the process.
A conceptual design is a sketch or written description of the

system and how it would work.
• The visual management systems will be evaluated as part of the

pilot test in the Improve phase:
• Simple displays/controls will be developed as designed.
• More complicated, costly displays/controls will be mocked up.
90
45
Layout for the future-state process
A process layout is:
The arrangement of the materials, equipment and employees (such

as their workstations) used within a specific process
L L M M D D
D D
L L M M
L L M M
G G
L L G G
A A
A A G G
91
When process layouts are important
Layouts are very important in physical processes.

• Depicting the layout helps minimise:
• The transportation of the product from one process step
to the next
• Movement of materials, people and equipment within
each process step
• Depicting the layout helps optimise the ability for employees
to:
• Share information
• Assist each other
92
46
When process layouts are important, continued
Layouts are less important in virtual processes.

• A layout of a virtual process is usually not required.
• It may be impractical to rearrange (or co-locate) the
employees or equipment in a virtual process.
• The VSM and the “Activity of ___” studies depict the
movement of the product in a virtual process.
• You may need to define the location of any centralised
electronic files in the process to facilitate employee access.
93
The types of process layouts: Process-Oriented
A layout in which the employees and equipment performing the same

process steps are grouped together.
Step 3
Each group is usually organised Step 1
as a separate department having
different managers and separate
performance metrics. Step 5
Step 6
What are the advantages
and disadvantages of using
a process layout?
Step 2
Step 4
Process-Oriented Layout
94
47
The types of process layouts: Product-Oriented
(Work Cell)
A layout in which the employees and equipment performing the process

steps are co-located into an end-to-end continuous workflow called a
work cell.
S1 S2 S1 S2
S3 S3
Each group is usually S1 S2

Work Cell 1
S4
Work Cell 4
S4
organised as a separate S3 S6 S5 S6 S5
department having S1 S2 S1 S2
different managers S4 S3 S3
Work Cell 2 Work Cell 5
and separate S6 S5 S4 S4
performance metrics. S6 S5 S6 S5
S1 S2 S1 S2
S3 S3
Work Cell 3 Work Cell 6

What are the advantages S4 S4
and disadvantages of using a S6 S5 S6 S5
work cell?
Product-Oriented Layout
95
Other types of process layouts
Office • Designed to maximize the utilisation of space while

Layout providing employees with cubicles for some level of privacy
and quiet.
Retail • Organises a sales and service area to respond to or drive
Layout specific customer behaviors.
• Banking Centres and grocery stores are examples.
Warehouse • Optimises the usage of space, while minimising the effort to
Layout store and retrieve the materials.
• Examples are storerooms.
Fixed • The product stays in one location. The employees in the
Position process, and supplies move to the product.
Layout • Examples are the construction of a building.
96
48
When to organise the process layout in a work cell
Work cells are the preferred layout for processes that

produce a single product or product family.
• Each cell initially may process a narrow range of highly similar
products. Over time, the products may be broadened.
• Work cells are self-contained with all necessary equipment and
resources.
• There are five work cell configurations:
1. U-Shaped (employees on outside)
2. U-Shaped (employees on inside)
3. Straight Line
4. L-Shaped
5. Converging Line
97
Refine your initial future state value stream design so that

product flows from one end of your VSM to the other in a smooth
uninterrupted flow
The use of Load and Balance charts help you balance takt with
work content and resourcing requirements
Pull systems ensure you minimise over-production and should be
included in future state design where possible
Visual Management guides the actions in the process to prevent
errors or mistakes from occurring.
Layout is important to minimise the transportation/movement of
employees, equipment and product
98
49
Creating pull systems

Calculating SWIP
Designing your layout
99
ANALYSE
Correlation and regression
50
Correlation and Regression

Activity of
equipment
map charts tests
Prioritise
analysis Beta risks
efforts

Visual
staff results
Management
Layout
101
After completing this lesson, you’ll be able to:
Use a Scatter Plot to assess the shape,

strength, direction and presence of
outliers for a relationship between a
continuous x and a Y
Understand how predictive equations for
a continuous x and a Y using simple linear
regression are created
102
51
The purpose of correlation and regression
analysis in the Analyse Phase
Correlation and regression analysis are used to confirm the

significance of continuous x’s:
Potential Vital Few Input (x)
The method used depends on the (from the Measure phase)
type of data for the Y: Discrete Continuous
• Logistic regression analysis is used
Discrete
Discrete
when the Y is discrete. Hypothesis
Logistic
Regression
• Correlation analysis and Testing
Project Y
regression analysis are used
Continuous
when the Y is continuous. Continuous
Correlation and
Hypothesis
Regression
Testing
103
The basic principles/approach of correlation

analysis
Correlation analysis is a method to evaluate a potential

relationship between a continuous input (x) and a Y.
• The relationship is depicted graphically 60

through a Scatter Plot
• Limitations of using correlation analysis: 50
- Does not provide a predictive equation
- Does not confirm causality 40
60 70 80
104
52
How scatter plots are used
Scatter plots are used to evaluate the direction of a

potential relationship:
• As the x increases, what happens to the Y?
55 No Relationship Positive Relationship
54 60
53
52
51 50
50
49
48
47 40
46
45
46.5 47.5 48.5 49.5 50.5 51.5 52.5 53.5 60 70 80
Negative Relationship
60
50
40
105
60 70 80
How scatter plots are used, Continued
Scatter plots are used to depict the shape of a relationship:

• What shape best describes the relationship between the x and the Y?
106
53
How Scatter Plots are Used, Continued
Scatter plots are used to depict the strength of a relationship:
• How tightly grouped are the data points along the shape?
107
How Scatter Plots are Used, Continued
Scatter plots are used to identify suspicious outliers

• Are there any data points that do not fit the shape suggested by a
majority of the data?
108
54
Table exercise: using Scatter Plots
Objective: Interpretation of scatter plots
Review the Scatter Plot below to determine the
relationship between volume of calls answered (x) and wait
time (Y).
Time - 5 mins
What conclusions can be reached regarding the direction, shape and strength of the
relationship? Are there outliers?
• Direction _________________
• Shape _________________
• Strength _________________
109
• Suspicious Outliers _________________ LSS Green Belt ANALYZE v2.6
Regression
What is Regression Analysis?
• Develops a mathematical equation to quantify the relationship.
• Enables you to predict the Y based on
the value of the input (x).
• The model includes a
Coefficient of Determination (R2):
• Expresses the percentage of total variation
in the Y that can be explained by the
changes in the input (x)
• Regression analysis, like correlation, cannot prove causality
110
55
How and when regression analysis is used,
Continued
An example of a mathematical model:

• You can develop a predictive equation for the
total amount of change in your pocket:
• The Y is the dollar amount of change you have
in your pocket.
• The x’s are the coins (quarters, dimes,
nickels and pennies).
The predictive equation:
• Y (total $) = 0.25 x (number of quarters) + 0.10 x (number of dimes) + 0.05 x
(number of nickels) + 0.01 x (number of pennies)
• R2 = 100%
What would happen to R2 if you did not include the pennies?

How about the quarters?
111
The types of regression analysis
There are three types of regression analysis:
Method: Used to develop a predictive equation for:

Simple Linear One continuous Y variable and one continuous
Regression x variable.
Multiple Linear One continuous Y variable and multiple (more
Regression than one) continuous x variables.
Logistic Regression Discrete Y variables.
112
56
Interpreting the simple linear regression analysis
Minitab will create a line of best fit through the collected data to create
a predictive equation, allowing you to estimate the Y for a given value of
x
113
How to apply the predictive equation

From the previous slide, assume call volume was on average 4,900. You
can use the regression equation to estimate the wait time:
Wait Time = -37.2 + 0.0214 x Average call volume

• Wait Time = -37.2 + 0.0214 x 4,900
• Wait Time = -37.2 + 0.0214 x 4,900 = 67.66
If your goal is to reduce the average wait time to less than 68 seconds,
the average call volume should be kept as low as possible and no higher
than 4,900 calls.
What things should management consider before taking action?
114
57
How to apply the predictive equation (Cont’d)
The management team now can

consider:
• Is it possible to achieve this reduction in
calls?
• Does the cost of reducing call volume to
achieve this improvement justify the
benefit?
• Is there a true causal relationship between
typing speed and handle time?
115
Use correlation to understand the direction, shape and

strength of a relationship between a continuous x and
continuous y
Regression analysis creates a predictive equation to
quantify this relationship
116
58
Reviewing your planned testing

strategy
Reviewing the Correlation
results
What to do if you have more
than one continuous x
What to do if your Y is discrete
117
ANALYSE
Hypothesis testing
59

• Describe the two primary applications of hypothesis testing in all Lean Six
Sigma project types
• Prepare null and alternative hypothesis statements to confirm the impact
of a discrete x upon a project Y
• Explain how the p-value is used to reach a statistical conclusion for all
hypothesis tests
119
Hypothesis testing

Activity of
equipment
map charts tests
Prioritise
analysis Beta risks
efforts
Activity of Visual Interpret test

staff Management results
Layout
120
60
Approaches to Problem Solving
The Typical Approach The Six Sigma Approach
Business Problem Business Problem Statistical Problem
Business Statistical
Business Solution
Conclusion Conclusion
121
Approaches to Problem Solving. Continued
Data is collected on the Average Handling Time

(AHT) for two teams within a call centre:
• Team 1 average AHT = 4.12 minutes

• Team 2 average AHT = 5.12 minutes
What conclusion(s) would you make from this data?
122
61
Approaches to Problem Solving, Continued
Uncertainty must be factored into every decision

making process:
• Average values alone, when generated from sample data, do not take
into account uncertainty
• Uncertainty for a given sample average is a function of:
• The variation that exists in the data
• The sample size
• Measurement system variation
123
What is a confidence interval?

• The estimated range, which is likely to include an unknown
population parameter, with a specified level of confidence.
• A 95% level of confidence is used for most Green Belt projects
• Confidence intervals get:
• Smaller as the sample size increases and/or the sample
variation decreases
• Larger as the sample size decreases and/or the sample variation
increases
124
62
The 95% confidence intervals for both

call centre teams:
5.8
5.5
5.1
AHT (minutes)
Team 1 AHT = 3.4 to 4.8

4.8 minutes
Intervals ‘overlap’
overlap’
4.5 4.4
4.1
Team 2 AHT = 4.4 to 5.8
minutes
3.5
3.4
Team 1 Team 2
Team
125
The concept of hypothesis testing
What is hypothesis testing?
– Hypothesis testing is a process that enables you to

statistically evaluate differences by:
• Converting a business problem into a statistical

problem
• Performing the appropriate test to generate the
statistical conclusion.
– This provides us with process focused and data driven

business conclusions.
126
63
The concept of hypothesis testing, Continued
Hypothesis testing is used to confirm the significance of discrete x’s:
Discrete Hypothesis Tests:

 Test Probabilities (1 Proportion) Potential Vital Few Input (x)
• The Y can be  Contingency (2 or more (from the Measure phase)
continuous or Proportions)
Discrete Continuous
discrete (but the
x is discrete)
Discrete
Discrete Logistic
Hypothesis Regression
Testing
Project Y
• Different Continuous
Hypothesis Tests:
Continuous
statistical tests
 Unequal Variances Continuous
are used based  Test Mean Hypothesis
Correlation and
on what you are Regression
 Two-Sample t Testing
testing  Means/Anova/Pooled t
 Means/Anova
 Median Test
127
The prerequisites to conduct a hypothesis test
A well-defined business problem is required to conduct a

hypothesis test:
The components of a business problem
• The purpose of the test Problem Problem
• The Y for the test
• The x for the test
Conclusion Conclusion
128
64
The prerequisites to conduct a hypothesis test, continued
Assume that a Lean Six Sigma team is working to reduce the

percentage of calls that were answered after 30 seconds:
• The C&E matrix identified “call centre

location” as a potential vital input (x)
• How the team defined the business problem:
– The purpose of the test: To determine if
the call centre location is a vital input (x)
for the Y
– The Y measurement: The proportion of
calls that are not answered within 30
seconds
– The x measurement: The call centre
location
129
The concept of null and alternative hypotheses
Statistical problems are divided into two opposing statements,

theories or claims:
The… States…
Null Hypothesis Changes in the (x) produce no difference in the Y

(H0) • There is no change.
• This is often the opposite of what you hope to
prove through hypothesis testing
Alternative Changes in the (x) make a difference on the Y

Hypothesis (HA) • There is a change.
• An “alternative” to null
• The alternative hypothesis is not unequivocally
proven
130
65
How null and alternative hypotheses are recorded
Hypothesis statements are recorded in a mathematical format.

They include:
A notation of the population A mathematical symbol for the

parameter being tested: claim to be tested:
Population Claim of Hypothesis Symbol Used

Parameter Notation
No difference =
Mean  (mu)
Difference with 1 or 2 ≠, <, or >
groups
Variance 2 (sigma squared)
Difference with more than At least 1
Median  (eta) 2 groups group is
different
Proportion p
131
How null and alternative hypotheses are recorded, Continued
The null and alternative hypothesis statements for

the call centre example:
H0: PGlasgow = PEdinburgh = PDundee = PAberdeen = PInverness
HA: At least one location is different
132
66
Stating Null and Alternative Hypotheses
Assume your team is trying to reduce the time to complete orders.

Two potential vital few inputs were identified:
1. The customer entry method to the process.
– Internet
– Call centre
– Store
2. Order approval type.
– Pre-approved order
– Standard processing order Time - 20 mins
Record the statistical problem for each x through

statements of the null and alternative hypotheses.
133
Example Hypotheses
Hypotheses for Means (CT = cycletime) Hypotheses for Medians

Ho: CTA = CTB Ho: MDayshift = MNightshift
Ho: CTA = Target
Ho: MDayshift = Target
Ha: CTA  CTB
Ha: MDayshift  MNightshift
Ha: CTA > CTB
Ha: CTA < CTB Ha: MDayshift > MNightshift
Ha: MDayshift < Mnightshift
Hypotheses for Variance Hypotheses for Proportions (e.g.
Ho: 2After = 2Before defect rates)
Ho: 2After = Target Ho: pMon = pTues = pWed= pThurs= pFri
Ha: 2After  2Before Ha: At least one px is different.
Ha: 2After > 2Before
Ha: 2After < 2Before
134
67
State the decision criteria
The Two Risks of Hypothesis Testing

Alpha () and Beta () risk explained:
Conclusion
H0 HA
 risk
Correct
H0 Type I Error
Decision
Truth
 risk Correct
HA Type II Error Decision
135
 Risk
The  risk or risk of making a Type I Error, is the probability that
we could be wrong in saying that two or more things are different,
when in fact, they are the same
It is an assessment of the likelihood that the observed difference
could have occurred by random chance.
 is the primary decision-making benchmark of most statistical
tests.
In the our legal system, the

defendant is assumed to be
innocent until proven guilty.
In the statistical system, two
populations are assumed to be
the same until proven different.
The “legal  risk” is the risk that
an innocent person could have
been convicted of a crime.
136
68
 Risk
Examples of Type 1 errors (alpha risk):

• Concluding Distribution Centre A had longer average cycle times
than Distribution B, when in reality they have the same average
cycle time
• Concluding that a Black Belt project successfully reduced cycle
time variance, when in reality it did not
• Fire alarm going off when there is no fire – finding a difference
when none exists
Typically  = 0.05 or 5%
137
 Risk
The  risk or risk of making a Type II Error, is the probability that
we could be wrong in saying that two or more things are the same
when, in fact, they are different
 risks are not typically computed directly by statistical tests
The “legal  risk” is the

risk that a guilty person
could have been acquitted
of a crime and set free.
138
69
 Risk - continued
Examples of Type II errors:

• Concluding no difference exists between dayshift and nightshift
median call response time, when in reality a significant difference
does exist
• Concluding error rates are same every day of the week, when in
reality error rates on Mondays are less than the other four days of
the week
• Fire alarm doesn’t go off when it should
Typically  = 0.10 or 10%

Power = 1 - 
139
Stating the decision criteria
As a table team, discuss the impact of  and  errors on

the successful completion of your project. Answer the
questions below:
What problems, costs or frustrations would your

team experience if it made:
• an  error when confirming the vital few
inputs?
Time - 20 mins • a  error when confirming the vital few inputs?
Brainstorm the potential impact of the two risks and record them on flip chart paper.
Each group are to present their results to the class.
140
70
Establish d
Delta (d) also known as Precision, is the practical

significant difference that you want to detect in your
hypothesis test.
Examples:
• Detecting average cycle time differences of 10 minutes has been
determined to be of practical significance
• Detecting an error rate difference of 5% has been determined to be
of practical significance
d
141
Data collection plan guidelines for hypothesis testing
The data collection plan for your hypothesis test should define:
• What data will be collected and when

it will be collected
• How much data should be collected
• Where the data should be collected
• Who will collect the data
142
71
Data collection plan guidelines for hypothesis
testing
Unlike sampling for estimation (covered in the Measure module),

sampling for hypothesis testing includes consideration of  risk
Sample size will depend on:
• Hypothesis test selected
• How Ha is defined
• Choice of  risk (Typically 5%)
• Choice of  risk (statistical software will often ask for power)
• Power = 1-  ( is typically 10%)
• Choice of d (known as “difference” in Minitab)
• Other parameters (such as the sample standard deviation for a t-
test)
143
The role of the p-value in reaching a statistical

conclusion
The p-value and the stated decision criteria form the basis for the
statistical conclusion
• The p-value represents the probability that the null hypothesis is true
(that is, there is no change in the Y as we observe change in the x)
• Since the p-value is a probability, values will be between 0 and 1
• p-values and statistical conclusions:
If the p-value is … Then the statistical conclusion is to …
Less than  Reject the null hypothesis and accept the
alternative. You are concluding a statistically
(Typically  = 0.05) significant difference. Hence, the x being tested is a
vital input.
Greater than  Fail to reject the null hypothesis. You are

concluding that there is no statistically significant
difference. Hence, the x being tested is not a vital
input.
If the p-value is low, the null must go. If the p-value is high, the null must apply.
144
72
Hypothesis testing roadmap
Let’s explore the ‘Hypothesis Testing Roadmap’ in the

‘Student Data Files’ package
145
Conducting statistical tests
Review underlying assumptions of a given hypothesis

test; verify all are satisfied
• For example, before performing a test for equal variance, first
carry out normality tests on all samples
Calculate test statistic and critical value with Minitab
If Minitab is not used:
• Calculate the appropriate test statistic.
(e.g., t-value, F-value or Chi-Square value)
• Determine the critical value for the test statistic.
(e.g., based on sample size and  risk look up tcritical in a t-table)
146
73
When statistical software is used to apply the hypothesis
test
The sample data for the call centre example:

• The x – Call centre location
• Y – Proportion of calls not answered in 30 seconds
• The test selected – Chi-Square
• The sample data:
147
Using statistical software to apply the hypothesis test
The p-value
indicates a
difference. But
which call centre
or centres are
different?
148
74
Develop statistical conclusions, continued
If statistical software is not used, compare the test

statistic to the appropriate value in a statistical table as
follows:
• If test statistic > critical value, reject the null hypothesis
• If test statistic < critical value, fail to reject the null hypothesis
149
Develop statistical conclusions, continued
Examples:
The p-value for the 2-sample t-test of average cycle time at
distribution centres A and B was 0.01. Since p < 0.05, we reject
the null hypothesis and conclude that average cycle times are
different at the different distribution centres
The p-value for the F-test of cycle time variance before and after
a Black Belt project was completed was 0.78. Since p > 0.05, we
fail to reject the null hypothesis and are unable to conclude that
the Black Belt project had a significant impact
150
75
The concept of an actionable x
Your confirmed vital few inputs (x’s) must be recorded at an

actionable level of detail:
• Knowing that Glasgow appears to be better and Aberdeen

appears to be worse than others is valuable but inadequate
information to define a solution to the problem.
• Call centre location is not an actionable input.
– Actionable x’s are at a level of detail to define an
appropriate control or solution for the x
• For the x to be actionable, you need to know:
– What does Glasgow do (or not do) to influence their
performance?
– What does Aberdeen do (or not do) to influence their
performance?
151
Every hypothesis test involves a comparison and conclusion

to be drawn
Null and alternative statements are used to create the
hypothesis
There are risks associated with statistical tests which need
to be understood
The selection of hypothesis test to be used requires a full
understanding of the question to be answered and nature
of data available
An alpha risk of 0.05 and beta risk of 0.10 are normally
used in hypothesis tests
Statistical software is normally used to calculate the
hypothesis test statistics 152
76
When reviewing your planned

testing strategy
Determining confidence level
and appropriate sample size for
the test
Running the test using
statistical software
Confirming the statistical and
business conclusion from your
analysis
153
77
IMPROVE
Lean Six Sigma
Green Belt
At the end of the module, you will be able to:

Use techniques to generate solutions to
business problems
Test solutions
Prepare an initial control plan
Prepare the implementation and
communication plans
2
LSS Green Belt IMPROVE v2.6
1
4 step approach
1 2 3 4
Detailed
Create initial Implementation
Brainstorming physical
control plan plan
layout
Mistake Communication
Benchmarking Piloting
proofing plan
Visual
DoE
Management
Assumption
5S
busting
Force field
analysis
Pugh matrix
3
IMPROVE
Generate and select solutions
2
Generate and test solutions
Detailed
control plan plan
layout
proofing plan
Visual
DoE
Management
Assumption
5S
busting
Force field
analysis
Pugh matrix
5

Explain how Brainstorming, Benchmarking, Famous
Faces, Assumption Busting and Force Field analysis
are used to generate solutions
Select and apply the appropriate method to
generate one or more potential solutions for your
project
Describe the three primary factors to evaluate the
feasibility of a potential solution: Technical
Impact, Cost and Organisational Acceptance
Construct a Pugh Matrix to select the best solution
to test
6
3
Solution generation begins with the right people
You’ve confirmed your root causes in the Analyse

phase and are ready to identify potential solutions.
What people would you include in your

solution-generation activities and why?
Be sure to:
1. Identify and include the appropriate SMEs for
the high-level process
2. Avoid psychological inertia among the
participants when generating solutions
7
Overview of Brainstorming
Brainstorming is a collaborative approach for

generating a large number of potential solutions:
• Benefits
– Allows open and creative thinking
– Encourages team member participation
– Generates solutions quickly
• Requirements
– A skilled facilitator
– Works best in groups of 8 to 12 people who have diverse
perspectives
8
4
Brainstorming
Capture In scope/
Ask Loosely Clarify
initial out of
Question Affinitise ideas
ideas scope
For Brainstorming Let people Read out the When it looks Finally the ideas
to be effective – have time to ideas and start like there are need to be
a frame of generate some to make some no more ideas, sorted into
reference is ideas on their form of read through those that are
needed. This can own. When it grouping. You the ideas and to be taken
often be a simple looks like could let the clarify – forward or
question; things have team do this. practice those that may
slowed down, Let any “appreciative be eliminated
Eg. How can we enquiry” – do
gather the additional ideas
grow existing not give your
ideas on a flip be forthcoming
revenues? How own opinion –
chart
can we reduce ask how it
staff costs? relates to the
initial question
9
Overview of Benchmarking
Benchmarking is a process for:

• Identifying best-in-class practices/
standards
• Within or outside your organisation
• Comparing your internal practices
• Improving the performance of a
similar activity
10
5
Overview of Benchmarking, Continued
The two types of benchmarking:

• Performance Benchmarking Types of
Benchmarking
– A search to determine what are the “best
performance results”
Performance Process
• Process Benchmarking: Benchmarking Benchmarking
– A search to determine how “best practice”

organisations provide their products and
services
11
How Benchmarking works
Benchmarking occurs in three phases:
Preparing Conducting Learning from

for the the the Study
Study Study Findings
• Defining scope of • Developing the • Interpreting the
the study interview guide information
• Planning your • Collecting acquired
approach information using • Applying what
• Locating and the interview was learned into
securing partners guide a solution
12
6
Famous faces
Start with a clearly defined issue or problem to

generate ideas against
Provide any supporting facts and figures as
necessary
Write the problems statement on a flipchart
Place some face cards facedown then ask the team
to select one
Ask how would “.........” think about this problem?
When ideas dry up move onto the next card
Generate as many possible solutions as possible
Have fun!
13
Assumption Busting
1. List assumptions - Look at the situation you are in.

• Think about what assumptions you are making?
• What seems obvious, so much so that you would not think about
challenging it?
• Think about cost & time, and identify where it is assumed
impossible to do something because of these constraints
• Think about where something works as a result of certain rules
or conditions
• Think about the things that people perceive, believe, think or
need
• Think about whether it is a need or a want
14
7
Assumption Busting, Continued
2. Challenge assumptions – take an approach where all

assumptions can probably be challenged and
overcome at some time.
• Ask how could this be flipped so that it is no longer true or a
constraint?
• Ask what would it be like if we could do it significantly better
– or if it took half the time?
3. Find ways of making the challenge the new reality

• Take each assumption where it was identified that something
could be done, and find ways to make it happen.
15
Force Field Analysis

List out the Forces for Change List out the Forces against Change
Saves cost
Redundancy expenses
Less training required
Employee satisfaction Redundancies
Quicker cycle times Limited number of queries
can be raised
More responsive
No human contact
Increases auditing capability
Limited audit scope
1. First list out the Forces for Change on post its on the left hand side of a flip chart / white board /
wall
2. List out all the Forces against Change
3. Align the For vs Against
4. Look for any gaps, especially where there is no “for” where an “against” exists and identify where
what could be done to counter that force.
5. When communicating the results it can be useful to show the impact / strength of a force (for or
against) by varying the size of the arrow. Relative positioning (horizontal) can also be used to show
the probability of the force being realised.
16
8
What separates the best from the rest?
You’ve confirmed your x’s in the Analyse phase and

identified multiple solution alternatives so far
in the Improve phase.
• What factors would you use to separate the “best” solutions from
the others?
17
Factors to consider when evaluating solutions
The technical impact of a solution:
• Defines the extent to which a solution

produces a positive change in the
project’s metric(s):
And therefore achieves the

chartered goal for the metric(s)
18
9
The cost of a solution:

• The true cost of a solution will include its:
• Design costs
• Development costs
• Implementation costs
• Operating/sustaining costs
• Sun-setting costs
19
The level of organisational

acceptance for a solution:
• Describes the extent to which the project’s
key stakeholders will adopt or support the
solution:
• Focuses on the stakeholders who are
identified in the Project Risk Assessment
• Considers how the solution relates to
other projects/initiatives
20
10
Overview of the Pugh Matrix
The Pugh matrix is a tool to identify which potential

solutions are more important or “better” than others.
How it works:
• Compares all identified solutions to the
weighted criteria to identify the best solution
Benefits:
• Facilitates the comparison of multiple
solutions
• Helps you arrive at an optimum solution
that could be a hybrid of the other solutions
21
How to construct the Pugh Matrix
Step 1: Confirm the evaluation criteria to use for the

solutions.
• At a minimum, use Technical

Impact, Cost and
Organisational Acceptance.
• Add other criteria, if
necessary.
• Record the evaluation criteria
in the first column of each
row.
22
11
Step 2: Assign an “Importance Rating” to each

criterion.
• Rate the importance of each

criterion using a scale from 1
to 5, with 5 having the
greatest importance
23
Step 3: Select the initial “best solution” as the datum.
• Record a description in the

“datum” column
24
12
Step 4: Use the evaluation criteria to compare each
solution against the datum. Assign the appropriate
comparison rating.
• Record the alternative solutions
in the first row of the remaining
columns
• Compare each solution against
the datum using the evaluation
criteria.
• Rate the solution as:
– Better than (+) the datum
– The same (S) as the datum
– Worse than (-) the datum
• Rate the datum with an (S) for
each criterion. LSS Green Belt IMPROVE v2.6
25

The Pugh Matrix template will
Step 5: Score each solution. automatically score the solutions based
on your input
• Calculate the “Sum of” ratings
– Record the number of (+) ratings
for in the “Sum of +’s” row
– Record the number of (-) ratings
in the “Sum of –‘s” row
– Record the number of (S) ratings
in the “Sum of Sames” Row
• Calculate the “Weighted Sum of”
ratings
– Add the “importance ratings” for
all (+) ratings assigned to the
solution.
– Add the “importance ratings” for
all (-) ratings assigned to the
solution.
26
13
Step 6: Select the best solution.
• The “best” solution from the

Pugh matrix is the one with:
– The most pluses and the
least minuses
– The highest weighted sum
of pluses and the lowest
weighted sum of minuses
27

Step 7: Improve the best solution by incorporating the
superior features or characteristics of the remaining
solutions.
• Your "best" solution may have (-)
ratings for some criterion.
– (-) ratings indicate the potential to
make the "best" solution even
better.
• Examine other solutions in the
matrix with (+) ratings for these
criterion
• Review the features of these
solutions that led to its (+) rating
• Synthesising these features with
your best solution.
28
14
An example of the Pugh Matrix
Evaluating potential solutions for a family holiday
29
Constructing a Pugh Matrix
Pugh Matrix
Objective: To practise use of the Pugh matrix in selecting
solutions
As a table team, use the Pugh Matrix to identify the best solution to deliver Lean
Six Sigma training to Green Belt candidates
Use the potential solutions and evaluation criteria below as a starting point
Feel free to add other criteria as needed
Time - 20 minutes Be prepared to discuss your results
You have identified the following potential Your team also has identified the following criteria to
solutions: evaluate the solutions:
• Web-based training • The cost of the training
• Printed, self-instructional materials • The time to complete the training
• Instructor-led training using Web meeting • Each participant’s ability to apply the training content to
• Videotaped sessions his or her project
• Each participant’s ability to apply successfully the
content to his or her daily work activities
• Each participant’s ability to ask questions
30
15
There are multiple different ways to

generate solutions, the key is to
generate as many as possible and
throw the bad ones away
A Pugh matrix can be used to choose
between alternative options
31
Be sure to engage your coach when

Planning any type idea generation session
Analysing the results of a Pugh Matrix
32
16
IMPROVE
Test selected solutions
Test selected solutions
Detailed
control plan plan
layout
proofing plan
Visual
DoE
Management
Assumption
5S
busting
Force field
analysis
Pugh matrix
34
17
At the end of this lesson you will be able to:
• Find sources of help to prepare a detailed

physical layout of a process
• Construct a detailed layout for any process steps
where the product flows physically
• Evaluate your detailed physical layout
• Identify the stakeholders to review your detailed physical layout
• Conduct a stakeholder review of your detailed physical layout
• Pilot and validate the new solutions as part of the physical layout
35
Importance of the lesson
A well-designed, detailed physical layout of the process

is necessary for two reasons:
• The layout impacts the flow of
the product and the amount of
waste in a process.
• The detailed layout provides the
information to transition to the
new layout for the pilot test.
36
18
When the step is performed
A detailed physical layout is necessary when:
Your future-state process Your conceptual layout from

design: the Analyse phase:
Requires a change in the and Lacks the information or detail

physical arrangement of staff, to guide the appropriate staff
equipment or materials at one in deploying the change.
or more process steps.
The layout can be limited to the process steps where

changes are required.
37
The key inputs to begin the step
Any site drawings or

blueprints for the facility or
work area
Any governmental
regulations/requirements that
must be considered in
developing the layout
38
19
The software to use when preparing the layout
Several software applications are available to speed the

process of a generating a detailed layout.
Microsoft Visio
Computer-Aided Design
(CAD) software such as
AutoCAD
Microsoft PowerPoint or
Excel
39
The indicators of a good physical layout
A well-designed, physical layout will:

• Promote a continuous product flow
• Promote the visual management systems
• Minimise waste
• Protect the health and safety of staff
• Maintain the security of information used
• Optimise the use of the workspace
• Be flexible to address changing demand levels and product/process
changes
• Accommodate the processing of rework without impacting the flow of the
product
• Incorporate the principles of 5S
40
20
The indicators of a good physical layout, continued
The 5S Audit Worksheet lists a series of assessment criteria of each of

the 5S categories.
41
The 5S Audit Worksheet lists a series of assessment criteria

of each of the 5S categories.
• The worksheet is used primarily in an
audit context for an existing physical
workspace.
• The criteria can be used to evaluate
and improve your layout.
• Review each criteria. Ask:
• Does the layout align to this criteria?
• How can the layout align to the criteria?
42
21
The indicators of a good physical layout
The 5S Audit Worksheet for a Physical Environment
43
44
22
Secure stakeholder feedback on the proposed layout.
Changes to the physical work environment are often

resisted by staff because:
• They represents a change to the status quo.
• They directly impact their personal needs.
Resistance increases when employees have

no opportunity to provide feedback on the
change prior to its implementation.
• Good ideas can be rejected, not because they lack value,
but as a means of retaliation.
45
Validating selected solutions
Pilot Testing
What is a pilot test?
• A limited-scale, live, controlled deployment
of your selected solution to validate the impact
on the project’s metric(s)
Key points to remember:

• You are piloting the improved process and
initial control plan
• The objective is to:
– Acquire statistical proof of an improvement to the Y with the new process
– Verify the effectiveness of the initial control plan
46
23
How pilot tests are conducted
2. The
1. The pilot test improved process
is planned and control plan
is deployed
Pilot Testing
4. The data (if stable) 3. The stability of the Y

is analysed is monitored
with control charts
• Hypothesis testing
determines if the Y
was improved
• Sigma Levels define
the capability of the Y
47
How control charts are used

Control charts are used to assess the stability of the Y
during the pilot:
• Data on the Y is collected

• The data is plotted on the
control chart (you can add to
the initial control charts for
your project metrics created
in the Measure phase)
• The control chart helps
determine whether the
improved process
(during the pilot run)
performs in a stable manner
48
24
How hypothesis testing is used
Hypothesis tests can determine whether

improvement has occurred with the Y:
Has the goal for the Is the Y for the improved
project metric been process different from
achieved? its baseline measure?
H0: YImproved = Goal H0: YImproved = YOld
HA: YImproved < Goal HA: YImproved < YOld
49
Why a sigma level is calculated for the Y
The sigma level defines the capability of the new process

to meet the requirement for the Y over time:
• The Sigma Level for the Y will
estimate the number/percentage
of defectives for the Y.
• Use the customer specification to
calculate the Sigma Level.
50
25
The duration of pilot test
The duration of the test is driven by the data requirements

to:
• Assess the stability of the Y
• Conduct the appropriate hypothesis test
• Calculate the Sigma Level
51
Pilot tests require careful planning
Careful planning of your pilot test is

critical when:
• Your solution may present significant risk
to the business
• The scope of the project may be large
• The cost of the solution (acquisition and
deployment) may be high
• The solution might have long term effects and
consequences
52
26
The elements of the pilot test plan
Elements of a good pilot plan:

• The validation methods to be used
• How the test will be conducted
• The project risk management approach
(including stakeholders)
• The measurement systems used
• The data collection plan
• The evaluation criteria and strategies
for the pilot
53
Tips for planning the pilot test
Improve the value of your pilot test by:

• Applying the principle of “stress testing”
• Collecting data on other factors (process or external) that are
likely to influence the results
• Observing as much of the pilot test as possible
54
27
Other benefits of pilot testing
In addition to validating the effectiveness of your solution,

the pilot test:
• Evaluates the effectiveness of your initial control plan
• Reveals tips and traps for the full deployment
• Simplifies the effort to estimate the cost savings for the solution
• Minimises long-term resistance to the solution among stakeholders
55
The importance of checking your measurement

systems
Consider the quality of the measurement systems

before conducting the pilot.
• Review the key measurement systems
(defined in the control plan) that will be
used during the pilot test.
• Determine if there are likely precision
and accuracy issues for:
• The measurement of the project metrics
• The measurement of the vital few inputs (x’s)
Do not start the pilot until you have a high level of confidence
in your measurement systems.
56
28
The importance of educating impacted staff
Staff involved/impacted in the pilot test need to
understand:
• The purpose and duration of the pilot
test
• The requirements to operate the
process using the control plan
• Any additional data collection
requirements for the pilot
(not specified in the control plan)
57
Team roles during the pilot
Clarify and communicate what the team should

look out for during the test:
What is observed (and learned) by team members

can be applied when:
• Conducting any follow-on pilot tests
if the improved process shows minimal
or no improvement to the Y
• Developing the full-scale implementation plan
for the solutions (if the goal for the Y is achieved).
• Developing the final control plan
for any affected process steps
58
29
Evaluate the results of the pilot tests -
graphically
59
Evaluate the results of the pilot tests -

numerically
Calculating Sigma Level for Discrete Data
Enter Number of Items Evaluated 16,000

Enter Number of Defective Items 8,902

This is your DPM 556,375
Directions for Template: Provide the required data in the gray boxes for
either continuous or discrete data types and the blue boxes will be calculated
automatically. Calculating Sigma Level for Discrete Data
Enter Number of Items Evaluated 1,500

Enter Number of Defective Items 460

This is your DPM 306,667
60
30
Determine the appropriate follow-up actions
Outcome Action
The pilot results • Develop a plan for the full-scale implementation of
are acceptable the process
• Finalise control plan
Improvement • Determine if the test was conducted properly.
occurs with the • Verify the vital few inputs (x’s)
project metric
(but not to the goal • Review solution to determine if changes are
level) necessary
• Modify solution and the control plan as necessary
• Repeat the pilot test
No change or • Determine if the test was conducted properly
improvement • Verify the vital few inputs (x’s)
occurs with the
project metric • Repeat the solution generation and selection activity
• Update the control plan for the new solutions
• Repeat the pilot testing process
61
Pilot the new process to refine the design
Testing of selected solutions should include:

- Layout of the workspace where the new process will be and
- Testing of the new process itself
Assess the pilot results for improvement in process
performance and stability
62
31
Reviewing your detailed floor layout

Determining how to engage
stakeholders in a review of the
proposed layout
Reviewing your pilot test plan
When developing the strategy to
rollout the pilot test.
When evaluating the pilot test results
63
IMPROVE
Initial control planning
32
Initial control planning
Detailed
control plan plan
layout
proofing plan
Assumption Visual
busting Management
Force field
5S
analysis
Pugh matrix
65

At the end of this lesson, you will be able to:
• Describe the purpose, benefits, focus and structure of a control
plan
• Explain where and how control plans are applied in the DMAIC
methodology
• Construct an initial control plan for the vital few inputs (x’s) with
their solution in the improved process
• Select and apply the appropriate:
• Method to determine the out-of-control conditions for your vital
few inputs (x’s)
• Control method for the vital few inputs (x’s)
• Construct a reaction plan to use when out-of-control conditions are
detected
66
33
The purpose of a control plan
Control plans are based on the fundamental concept of Y=f(x):
If… Then…
You can control the vital You will have an
few inputs (x’s) that affect improvement that lasts
your metric (Y)
67
Control plans are a key management tool to sustain your project

improvements:
• Managing the process to assure its long-term performance involves:
Ongoing Process Management
Defining How the

Improved Process
will Work
Controlling the Vital Control Plans are a

Few Inputs (x’s) Critical Management
within the Process Tool for Each Activity
Monitoring the Ys for

the Process
68
34
What is a control plan?

• The blueprint for managing the improved process.
• It is a one-stop reference view of:
– The vital contributors to the success of a process
– The required data-driven actions to manage the process
• It includes documentation to effectively manage the
process such as:
– Detailed process maps
– FMEAs for process steps
– Standard Operating Procedures (SOPs)
69
Benefits of a control plan
A well-designed control plan offers significant benefits:
• Institutionalises the improvements

• Defines what needs to happen for the process to operate
consistently on target with minimum variation
• Minimises process tampering (over-adjusting the process)
• Defines the key training needs for the daily operation and
management of the process
Control plans must be reviewed and updated regularly
70
35
The focus of a control plan
Control plans are built for processes, not projects
• Control plans are not isolated, project-specific documents

completed to pass a tollgate.
• Control plans are process management tools.
• The findings from your DMAIC project are incorporated into an
existing control plan for the relevant business process.
If the process … Then you …

Has an existing Update the existing control plan with
control plan the findings for your project.
Does not have an Populate a new control plan for the
existing control plan process with your project information.
71
The structure of a control plan
Control plan elements:

1. Detailed process maps 7. Training requirements for the
2. FMEA’s for the process process
steps 8. Review/maintenance schedule
3. SOPs for the control plan
4. Customer Critical-to- 9. Capability studies for the
Quality (CTQ) process metrics
Requirements and 10. Measurement Systems
Measures Analysis (MSA) results for the
5. Critical information vital few inputs (x’s) and key
technologies used metrics (Y’s)
6. Reaction plans for the x’s 11. A control plan summary sheet
and Y’s
72
36
How control plans are used in DMAIC
Control plans are developed/updated at two points in the DMAIC

process:
• An initial control plan is developed in the
Improve phase.
– Focuses on the vital few inputs (x’s) to which the
solution is applied.
– Ensures that your solution is implemented
properly for the pilot test
• The final control plan is implemented in the
Control phase.
– Updates the initial control plan based on the pilot
test results
– Incorporates all findings and lessons learned from
the project to the process
73
Identifying out of control conditions
Before identifying these conditions, we need to have an updated

process map to develop the initial control plan:
• Applying your solution for the vital few
inputs (x’s) will require changes to the
process
– New steps might be added
– Existing steps might be removed
– Requirements or standards might have
changed for existing steps
• You should update the detailed process map
from the Measure phase to reflect the process
as it “should operate” with the solution
74
37
Methods to identify out-of-control conditions:
Specification Limits
• The solutions Values outside specification limits define an
for some x’s out-of-control condition:
might have
been limited to
establishing
acceptable
levels or
settings for the
x.
• For example:
• The number
of call
centre
staff
available
(20 to 25)
An out-of-control condition for these x’s occurs

when a value for the x falls outside a specification limit.
75

FMEA for the Improved Process
An FMEA of your improved process identifies out-of-control

conditions:
# Process Function Potential Failure Modes Potential Failure S Potential Causes of O Current Process D R
(Step) (process defects) Effects (Y's) E Failure (X's) C Controls E P
V C T N
1 Customer initiates -VRU drops call Customer must redial 3 Phone system overload 2 Alert once overload has 4 24
telephone inquiry occurred
7 Utilize resources to Wrong solution given Customer leaves bank 5 Lack of experience with 3 QA listens to random 4 60
resolve inquiry resources (computer sytems) calls and provides
coaching
7 Utilize resources to Wrong solution given Customer has to call back 3 Lack of experience with 3 QA listens to random 4 36
resolve inquiry resources (computer sytems) calls and provides
coaching
7 Utilize resources to Wrong solution given Bank must absorb an 1 Lack of experience with 3 QA listens to random 4 12
resolve inquiry additional cost resources (computer sytems) calls and provides
coaching
8 Process request -Customer action gets Customer must call again 3 Technical glitch 1 No 5 15
dropped when escalation or and sit through the wait
referral is required times
9 Prepare package for Information inaccurate Customer problem is not 4 Poor typing skills 2 None 5 40
processing center resolved or wrong action
taken 76
9 Prepare package for Information inaccurate Customer problem is not 4 no clear pollicies or 4 QA listens to random 4 64
processing center resolved or wrong action procedures for what to calls and provides
taken include coaching
9 Prepare package for -Information omitted Customer problem is not 4 Fields missing in database 3 No 4 48
processing center resolved or wrong action entry screen
taken
38
FMEA for the Improved Process
An FMEA of your improved process identifies out-of-control

conditions:
• FMEA evaluates/prioritises the potential
failures of a process to prevent them
from occurring.
• The FMEA in the Improve phase differs
from the Measure phase FMEA
• The FMEA in the Measure phase focused on the existing process.
• The FMEA in the Improve phase focuses on the improved
process, which incorporates your solution.
77
Identifying out-of-control conditions from the FMEA
Out-of-control conditions are the causes (and their

respective failure modes) with one or more of the following
indicators:
1. A high risk-priority number (RPN) relative to all
RPNs generated
2. High occurrence and high detection ratings
(regardless of the Severity rating) -- customer nuisance
3. High occurrence and low detection ratings (hidden factory)
4. A high severity rating
78
39
Follow-up actions from the FMEA
The next steps for the out-of-control conditions identified by the

FMEA:
• Determine if we can eliminate (or mistake-proof) the causes
• Determine how to identify these out-of-control causes or failure modes
in the process
• Create reaction plans for these out-of-control conditions once they are
identified
These should be documented on the FMEA as Recommended Actions.
79
Follow-up actions from the FMEA
The next steps for the out-of-control conditions identified by the

FMEA:
Process Potential Failure S Potential O Current D R Responsible S O D R

Potential Failure Recommend Taken
# Function Modes (process E Causes of C Process E P Person & E C E P
Effects (Y's) Actions Actions
(Step) defects) V Failure (X's) C Controls T N Target Date V C T N
80
40
Control Charts
Control charts can signal an out-of-control condition for your x’s:
• Control limits on a control chart
signal the presence of
a special cause of variation
within the process.
• We should consider control charting
the vital few inputs, especially the
controllable and noise variables
81
The two types of control methods
Control methods can have a prevention or detection orientation:
Prevention-Oriented Control Methods

· Prevent the cause
Failure
Cause Effect
Mode
Controls
Detection-Oriented Control Methods

· Detect the cause
· Detect the failure mode
82
41
Mistake-proofing as a control method
Mistake-proofing is an action taken to:
•Remove the opportunity for an error

in a process, or
•Make the error so obvious that it

cannot reach the customer
•Also called Poka-Yoke, a Japanese
term that means to avoid (yokeru)
inadvertent errors (poka)
83
Mistake-proofing as a control method
Prevention
Mistake proofing activities at
focus on:
Making Errors
Preventing Errors
Impossible
Failure
Cause Effect
Mode
Detection
Mistake proofing activities at
this point focus on:
· Detecting errors
· Making sure errors do not
turn into defects
84
42
The three mistake-proofing techniques
Mistake-proofing prevents or detects errors through three

techniques:
Technique Prevention Detection
Shutdown When a mistake is When a mistake or defect has
about to be made been made
Control Errors are impossible Defective items can’t move on to

the next step
Warning That something is Immediately when something

about to go wrong does go wrong
85
Table exercise: Mistake proofing

There are many ways to mistake-proof a
process. Review each of the following
examples and determine:
• What is the defect to be prevented?
• Will the method described prevent or detect the defect?
1: Some cameras will not function when there isn’t enough

light to take a picture.
Time 10 mins 2: Some clothes dryers have a device that shuts them
down when overheating is detected.
3: A fruit orchard takes great pride in the size of its apples.
All of its apples must pass through a sizer. The small apples
are sent to a discount outlet. The large apples are sent to
customers.
4. Smoke detectors provide a warning that smoke has been
detected and that there’s a possible fire.
86
43
Visual management as a control method
What is Visual Management?

• A control method using visual tools that enables your process to talk or
tell you:
• When everything is OK
• More importantly, when everything is not OK
• The visual tools enable “Management at a Glance”:
• You can easily recognise the difference between normal and abnormal
conditions as they occur.
• The abnormalities are made obvious to completion.
Technique Prevention Detection
Shutdown When a mistake is When a mistake or defect has
about to be made been made
Control Errors are impossible Defective items can’t move on to

the next step
Warning That something is Immediately when something

about to go wrong does go wrong
87
Visual displays and controls
Visual Displays:
Communicate important information, but do not influence
or control what people or equipment do.
Visual Controls:
Communicate information so that activities are performed
according to standard procedures.
Visual displays and controls

create a common visual language in the workplace.
88
44
The foundation of visual management:
The 5S Approach
• Methodology for creating and maintaining an organised, clean, high

performance workplace
• Target areas:
• People, materials, equipment methods and information
The 5S Approach
Sort • Get rid of what is not needed

Store • Arrange and identify for ease of use
Shine • Clean daily. Clean up what’s left
Standardise • Create standards and standard methods
Sustain • Set discipline, plan and schedule
89
FindThe Foundation
1 to 49 of Visual Management:
in sequence
The 5S Approach
Step 2: Determine the Appropriate Control Methods
90
45
1. Sort – Decide on what is needed
Sort
• To sort out necessary and unnecessary items

Definition • Clear Workplace and remove all un-needed items such
as files, binders, electronics, and excess materials
• Removes waste
• Creates a safer work area
Benefits
• Increases available workspace
• Simplifies the visualisation of the process
• Start in one area, then sort through everything

Tips • Discuss removal of items with all persons involved
• Do it safely and recycle as appropriate
91
Now do it again
27
30
23
47
40
10
13
92
46
2. Store – arrangement of items needed
Store
• To arrange all necessary items - have a designated place

Definition for everything and to put everything in its place
• Visually shows what is required or is out of place

• Reduces time to locate items/documents
Benefits
• Saves time, not having to search for items
• Things used together should be kept together

• Use labels, tape, floor markings, and signs to label items
Tips
• Keep items that are shared in a central location (eliminate
excess, equal access)
93
Try it again!!
5 10
9
19
13
23 30
36
40
47
94
47
3. Shine – sweep and cleanliness
Shine
• To keep areas clean on a continuing basis; while

Definition continuously raising the standards
• A clean workplace is indicative of a quality product and

process
Benefits • A clean workplace helps to identify abnormal conditions
and improve morale
• Storing “everything in its place” makes time available for

Tips cleaning
• Identify individual responsibilities for cleaning
95
4. Standardise – create a common language
Standardise
• To maintain the sorting, storage and shining activities into

Definition the workplace at a consistent level
• Helps uncover problems and reveal their magnitude

Benefits • Without standardisation of sorting, storage and shining,
the improvements from the first 3S’s will disappear
• Keep the work place neat and clean for visual identifiers
Tips to be effective in uncovering problems
• Develop a system that will enable anyone in the
workplace to see problems as they occur
96
48
5. Sustain – training & disciplined culture
Sustain
• Practicing and repeating the 4S’s regularly until they

Definition become a way of life in the workplace
• Assessing the current practices and developing
appropriate corrective actions
• Sustain the 5S’s into our everyday process as a habit.

Benefits • Commitment and discipline toward housekeeping is
essential in taking the first step in being World Class
Tips • Develop schedules and checklists for the 4S actions
97
What 2 numbers are missing between 1 and 49?

27
30 51
62
23
47
40
10
58 13
79
98
49
So easy!!!!
1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 19 20
21 22 23 24 25 26 27 28 29 30
32 33 34 35 36 37 38 39 40
41 43 44 45 46 47 48 49
99
Examples of Visual Management
Dashboards are a common visual management tool:

• A dashboard is a collection of
relevant visual displays and controls.
• Dashboards are readily visible.
• Other visual management tools
include:
• Pacing devices
• Real-time visual / audible feedback
(such things as counters, signals,
color-coding, speakers, alarms and
downtime clocks)
100
50
Control charts as a control method
Control charts can be applied to the vital few inputs (x’s)

to provide a statistical signal for an out-of-control
condition
• Apply control charts to x’s classified as
noise (N) or controllable (C) variables
on your detailed process flow.
• When using control charts:
• Prepare instructions and incorporate
them into the SOP.
• Define and train appropriate employees.
• Prepare a reaction plan for
out-of-control signals.
101
Process performance measurement as a control method
Process performance measurement monitors the key Y’s for

the process:
• Designing, evaluating and monitoring
the measurement systems for the Y’s
• Applying statistical process capability
and control methods to the Y’s
102
51
Standard Operating Procedures as a control method
SOPs are an agreed-upon set of procedures for the new,

improved process.
• Prepared to establish the best (and most reliable)
sequence/methods
for the process.
• Incorporated into the control plan.
• Benefits of Standard Work:
• Defines a consistent, repeatable process
• Defines a process to produce high-quality, low-cost products and
services
• Defines a process that minimises waste
• Facilitates learning and consistent performance among
employees
103
Tips and guidelines for applying standard work
Follow these guidelines when applying standard work:

• Specify requirements for inputs classified as “SOP” in your detailed process
map.
• SOPs document how you will control the process. They are a mandatory
deliverable for the improved process.
• Incorporate the use of job aids into the SOP
• Design steps that a typical employee can perform.
• Use a team approach:
• Schedule regular and frequent idea meetings.
• Revisit periodically. Standard work is not a one-time method.
• Make information readily available and easy to access for staff.
• Standard work must be sold to staff:
• The importance of standardised jobs
104
• Adhering to the SOPs and job aids LSS Green Belt IMPROVE v2.6
52
Steps to apply the standard work method
Steps to apply the standard work method.
Step Action
1 Lists the steps in the procedure in the order to be

performed. Define the position that should complete the
step.
2 Estimate the time to complete each step.
3 Document the expected outcome for each step as well as
potential out-of-control conditions (what they are and how to
identify them).
4 Specify reaction plans for these out-of-control conditions.

5 Document the process and its control checkpoints.
105
Inspection as a control method
Inspection is the least effective of the control methods:

• Most inspection activities focus upon
the Y’s (the process metrics or outputs).
• Inspection can be used for your x’s.
• It should not be the major (or only)
control method for the process. Why not?
– Inspection is not 100% effective.
– Inspection is costly and time consuming.
– Inspection, as the sole control, has a marginal impact on
the detection rating.
106
53
Control method summary
Detect
Prevent Detect Failure
Method Cause Cause Mode Notes
Mistake-Proofing    Prevents controllable x’s or
compensates for noise x’s
Visual   Helpful to identify out-of-control
Management conditions
Control Charts   Apply control charts to all vital few
inputs (x’s) designated as controllable
and noise
Process  Use to monitor the Y and ensure
Performance improvements are sustained. Identifies
Management future improvements
SOPs    Must have in all control plans
Inspection  Should not be the only control method
in the control plan
107
The concept of a Reaction Plan
Reaction plans should be prepared for each control method

• Define the probable causes and
appropriate corrective actions
whenever an out-of-control condition
is detected by the control method
• Serve as a troubleshooting guide
for staff
108
54
Components of a Reaction Plan
Reaction Plan components:

• A description of the out-of-control condition
• The probable causes for the condition and their corrective
actions
• The actions to diagnose/confirm the probable causes
• Who is responsible for the diagnosis and action
109
Document the initial control plan
The Control Plan Summary template is a spreadsheet that:
• Profiles the key process

steps (that is the step,
its inputs, outputs and
owner)
• Summarises:
– The performance characteristics for the steps
– The control methods for the steps
– The reaction plan for the steps
110
55
The Control Plan Summary Template
The Elements of the Control Plan Summary :
Ongoing Process Management
Defining How the

Improved Process
will Work
Controlling the Vital Control Plans are a

Few Inputs (x’s) Critical Management
within the Process Tool for Each Activity
Monitoring the Ys for

the Process 111
The Control Plan Summary template
The Elements of the Control Plan Summary :
Defining How the Controlling the Vital

Improved Process Monitoring the Ys for
Few Inputs ( x’s)
will Work the Process
within the Process
112
56
The initial draft of the control plan

should be drafted and tested within
the Improve phase
There are preventative, detective
and inspection methods to control
the vital x’s
Have a reaction plan to deal with
out of control conditions
113
Developing a FMEA for the new,

improved process to identify out-
of-control conditions
Selecting and configuring the
appropriate control methods for
your vital few input(s)
Reviewing the initial control plan
in its entirety
Be sure to involve the Process Owner in the development

of the Control Plan
114
57
IMPROVE
Implementation Plan
Implementation plan
Detailed
control plan plan
layout
proofing plan
Assumption Visual
busting Management
Force field
5S
analysis
Pugh matrix
116
58

• Describe how the Implementation Plan Checklist is used when
preparing an Implementation Plan for your project
• Explain why stakeholders often resist change
• Develop a proactive Communication Plan for the implementation of
your solution
• Determine the training requirements associated with the
implementation and adoption of the new solution
• Identify and evaluate the key risk items that require mitigation
strategies in your Implementation Plan
• Construct strategies to transfer the Control Plan to the Process
Owner
117
Guidelines and tools for preparing an implementation plan
A good Implementation Plan will include activities to:

• Procure the solution, if appropriate
• Introduce the solution to:
• The staff who will apply the solution
• The managers who oversee the
process where the solution is deployed
• Equip people with the knowledge and skills to introduce, use, and
manage the solution
• Monitor and evaluate the success of the transition to the new
solution
The content of the Implementation Plan will vary

from solution to solution 118
59
Overview of the implementation plan checklist
The Implementation Plan checklist defines the critical
elements to address in your Implementation Plan:
1. Key Assumptions
for the
Implementation
2. Process Change
Considerations
119
Overview of the implementation plan checklist, Continued
The Implementation Plan Checklist defines the critical

3. Improved Process
and Implementation
Risk
4. Control Plan
120
60
Overview of the implementation plan checklist, Continued
The Implementation Plan Checklist defines the critical

5. Verification of
Results
6. Implementation
Approval
7. Sign-Off
121
Document key implementation assumptions
Typical areas of assumptions

People Staff or customers impacted by the solution, vendors,
other stakeholders of the project or process
Operational The upstream or downstream impact of the process
Impacts improvement
Technology Whether existing technology is capable of handling the
impact of the change
Facilities/ Whether facilities, locations or workspace where the
Locations solution will be implemented are adequate and prepared
Communication Assumptions regarding planned communication of the
change
Training Assumptions regarding the ability to deliver or receive
training in the solution
122
61
Identify process change considerations
Use the Implementation Plan Checklist to document the improved

process considerations.
123
Change Adoption - Why solutions are resisted

Reasons why your solution might be resisted:
• It is viewed as a threat to a personal or
organisational need
• It is viewed as unnecessary
• It is perceived to do more harm than good
• The people introducing the solution are not
respected
• No opportunities were given to provide feedback
on the solution
• It is perceived as a personal criticism of the work
performed
• It is perceived to add more work or effort than to
keep the status quo
124
• Information on the change is heard secondhand
62
The importance of proactive communication
The minimum you can do to reduce resistance to change is

proactive, effective communication.
• Communicate the right information to the
key stakeholders:
• The need for the project
• The vital few inputs confirmed for the Y
• The rationale and benefits of the solution
• How the standard work process will change
(or remain the same)
• How the implementation will occur
• Their roles in implementing/supporting the change
• Have the right people communicate the change:
• Project Champion, Process Owner or both
• Solicit input from the stakeholders
125
Overview of a communication plan template
126
63
The implementation plan checklist summarises key
communication activities
Use the Communication Summary section of the

Implementation Plan Checklist to highlight the key
communication activities.
127
The three types of training requirements for your solution
Training is required whenever people lack the knowledge or

skills to:
• Roll out your solution to the targeted user community
• Who: The people who will introduce the solution.
Normally the Process Owner or Managers of the staff who are
adopting the solution
• Adopt the solution
• Who: The staff who will use/apply the solution on a daily basis in
the process
• Monitor/manage the solution
• Who: Managers of the staff who are adopting the solution
128
64
Common training methods
A range of training methods can be used to address the

training requirements for your solution.
Consider … When your training need …
Job aids or self- Is knowledge-based and does not require
study materials memorisation or quick access of information.
E-learning Is knowledge-based and occurs regularly among a
large and/or geographically dispersed audience.
Instructor-led, Is lengthy and involves complex skills requiring
classroom-based practice and instructor feedback.
training
Telephone/Web- Is short, involves simple skills requiring
Based demonstrations and simple practices and occurs
Conferences among a geographically dispersed audience.
129
The implementation plan checklist summarises your key

training activities
Use the Training Summary section of the Implementation
Plan Checklist to highlight your key training activities.
130
65
Identify and evaluate the risks associated with the
deployment of the solution
Mitigation strategies should be incorporated into your

Implementation Plan for the following risk items:
Risk Type Description

Project Risks “Red Status” risk items from the recent Project Risk
Assessment that do not have a mitigation plan in place
Process Risk items from the FMEA of the improved process that
Risks have either:
• A Risk Priority Number (RPN) greater than 30
• A Severity Rating of 5 and no mitigation plan in place
Other Risk items identified during the pilot test or when reviewing
Risks the process steps that will change for the solution.
131
The implementation plan checklist summarises the high-risk

items for the deployment
Record your high-risk items on the implementation plan

checklist:
132
66
Your plan should include activities to update and transfer
the control plan to the Process Owner
You will need to incorporate the following Control Plan activities

into your overall Implementation Plan:
• Activities to update the initial
Control Plan, if necessary,
based on the pilot
test results
• Activities to finalise the
Control Plan
– Develop and implement an audit schedule for the Control Plan.
– Transition ownership of the newer Control Plan to the Process Owner.
133
Your plan should include activities to update and transfer

the control plan to the Process Owner (cont’d)
The checklist includes questions to verify that you have

planned adequately for the transfer of the Control Plan.
134
67
Supporting the transition
The Project Lead:
• Assumes a consultative role, assisting the Process Owner in the
activities to:
– Control the vital few x’s and monitor the Y
– Verify sustainable improvement to the Y
The Implementation Plan Checklist identifies how long the Project
Lead will work with the Process Owner:
135
Approving the transition
Include the signatures of the Project Champion, Coach,

Process Owner and Finance Certifier on the Checklist:
136
68
A clear implementation plan will ensure

that solutions are deployed in the most
effective manner
The plan should not only cover technical
implementation activities but address some
of the people issues associated with change
Ensure that you have full approval on your
plan before deploying your solution within
your business
137

There are recommended support points with
your Coach, Project Champion or Process
Owner.
Be sure to engage your Coach in a review of:
• Your Communication and Training Plans for the
deployment
• The high risk items for the deployment and your
risk mitigation strategies
• Your approach to transfer the Control Plan to the
Process Owner
Be sure to engage your Project Champion

and Process Owner in a review of:
• The overall Implementation Plan
138
69
Key points of the Improve phase
Lean Six Sigma is all about implementing

sustainable successful solutions for
business.
Generate high volumes of ideas through
creative thinking techniques
Use a Pugh matrix to select between and
synthesise different solution options
Assess solutions within the physical work
environment and as part of a pilot
Have a detailed plan to implement
successfully addressing the people aspects
of change
139
Lean Six Sigma simulation: Final Round
The coin simulation

Objective: Use the tools covered in this and the Analyse module
to prepare:
The future state value stream map

Implement any changes you feel will improve performance
NOTE THERE IS A NEW COMPETITOR IN THE MARKET
60 mins
140
70
CONTROL
Lean Six Sigma
Green Belt
Use statistical process control

Deploy your solutions successfully
Close your Lean Six Sigma project
2
LSS Green Belt CONTROL v2.6
10/8/2014
1
Control Phase Overview
Sampling
Deployment
strategy
Choosing the
Lessons
appropriate
Learned
control chart
Signals of Project
special cause closure
3
10/8/2014
CONTROL
Statistical Process Control
2
Statistical Process Control
Sampling
Deployment
strategy
Choosing the
Lessons
appropriate
Learned
control chart
Signals of Project
5
10/8/2014
Create the sampling strategy for the

Control Chart and document it in the
Control Plan
Choose the appropriate Control Chart

for the variables
Identify signals of special cause when

looking at a Control Chart
6
10/8/2014
3
Review of Measure material
What are Control Charts?
Upper Control Limit
Center Line
Lower Control Limit
0 10 20 30
7
10/8/2014
Use of Control Charts to identify type of variation
Two types of variation:

• Common cause variation
• Special cause variation
UpperUCL
Control Limit
Center Line
LCL
Lower Control Limit
0 10 2 0
0 10 20 30
8
10/8/2014
4
Error in control charting
9
10/8/2014
Signals of unusual variation
Variation in call handle time:

•What are some examples of common cause
variation in the handle time for a call?
•What are some examples of special cause
variation in the handle time for a call?
10
10/8/2014
5
Control Chart variables to evaluate
Variables typically include:

•Project Y’s: monitor the primary and secondary metrics
•Vital Few Inputs (x’s): manage and control
11
10/8/2014
Control chart sampling strategy

The Control Plan will specify:
• Type of data
• Sampling strategy
• Designated control method
PROCESS PERFORMANCE CHARACTERISTICS CONTROL METHODS

REACTION PLAN
Process Process Step Process Step Process Step Measurement
SAMPLE (Refer to Process
Step Input (s) Output (s) Owner Customer Out of Control System for CONTROL
LSL USL Target FMEA)
CTQ Conditions Process SIZE FREQ. METHOD
Evaluation
Call Center Call Type Customer 0.95 0.97 1st Call 1 point outside Q&P randomly 100 Weekly np chart Investigate out of
System inquiry Resolution of control limits, records calls and control conditions
resolved trend reviews to ensure
resolution
12
10/8/2014
6
Control Plan example
CESS PERFORMANCE CHARACTERISTICS CONTROL METHODS

REACTION PLAN
Measurement (Refer to Process
SAMPLE
Customer Out of Control System for CONTROL FMEA)
L USL Target
Evaluation
95 0.97 1st Call 1 point outside Q&P randomly 100 Weekly np chart Investigate out of
Resolution of control limits, records calls and control conditions
trend reviews to ensure
resolution
13
10/8/2014
What type of data are we collecting?
PROCESS PERFORMANCE CHARACTERISTICS CONTROL METHODS

REACTION PLAN
p Measurement
SAMPLE (Refer to Process
Customer Out of Control System for CONTROL FMEA)
LSL USL Target
Evaluation
0.95 0.97 1st Call 1 point outside Q&P randomly 100 Weekly np chart Investigate out of
Resolution of control limits, records calls and control conditions
trend reviews to ensure
resolution
Our data for control charts can be:

• Continuous
• Discrete
• Defects
• Defectives
14
10/8/2014
7
Sampling strategy
Determining the sampling strategy:

• The four sampling strategies
Sampling Sampling
from a from a
Population Process
Random Sampling X
Stratified Random Sampling X
Systematic Sampling X X
Rational Subgrouping X
• The sampling strategy used by Control Charts is rational

subgrouping.
15
10/8/2014
The challenge of Rational Subgrouping

Challenge:
Construct a sampling plan that focuses on detecting between-
subgroup variation by identifying subgroups that minimise within-
subgroup variation.
Considerations:
• Size of the subgroup
• Frequency
• Composition of the samples for the subgroup 16
10/8/2014
8
Minimising within subgroup variation
Subgroup composition:
• The within-subgroup variation
influences the control limits.
• The greater the within-subgroup
variation, the wider will be the
control limits.
• Wide control limits may mask
special cause variation that
occurs between subgroups.
17
10/8/2014
Determining sampling frequency
Factors to consider when determining frequency:

• The sensitivity of the Control Chart refers to the ability of the
Control Chart to detect special cause (or out-of-control conditions)
between subgroups.
• The frequency of sampling should be often enough for special
cause variation to be detected when it occurs.
• Try to maximize the variability between subgroups.
18
10/8/2014
9
Selecting the correct control chart –
Control Chart Decision Tree
Question 1: What is the data type for the variable?
Question 2: What is the sampling strategy? 19

10/8/2014
Subgroup sizes
Sampling Take this path when…

Strategy
Equal You collect the same number of samples for each
subgroup subgroup
For example, you might collect, 150 forms per day,
sizes 100 calls per hour
Unequal You collect different numbers of samples for each
subgroup subgroup.
For example, you might collect all forms received in a
sizes day (this will change day-to-day), all calls in an hour
20
10/8/2014
20
10
Table activity: selecting the appropriate control chart
Objective: Use of Control Charts

In your table team, read each scenario on the following
slide and answer the questions below to determine which
Control Chart is most appropriate:
What variable is identified?
Is this continuous or discrete? If discrete, is the data
defects or defectives?
20 mins
How is data collected (as an individual value or in
subgroups)?
Is the same amount of data collected each time
(fixed or variable subgroups)?
How much data is collected each time (this is your
subgroup size)?
Prepare to report your team’s findings to the group 21
10/8/2014
Table activity: selecting the appropriate control chart
The billing department in a gas supply company has found a

relationship between customers receiving estimated bills and the
number of credits issued by customer service. The department has
started to monitor the number of bills that are based on actual meter
readings. It will take a sample of 100 statements every day.
Incorrectly lodged application forms are a problem for the company.
Too many errors on each form cause major delays for new customers
and drive significant additional cost into the business. The manager is
monitoring the errors daily based on all the forms filed each day.
The time to close mortgages is related to the time it takes to submit
the application. The mortgage processing manager feels there may
be a lot of variation in the time to submit the application and wants
to start monitoring it. The manager plans to collect the time to
complete every 50th mortgage application.
22
10/8/2014
11
What are signals of special cause?
Signals of special cause:

Signals of special cause are
•One point outside the control limits
•Non-random patterns in the data plotted over time that are
unlikely to occur solely due to natural variation.
What is a non-random pattern?
23
10/8/2014
Overview of non-random patterns
There are four different categories of non-random patterns:

• Cycling Pattern
• Mixture Pattern
• Shift Pattern
• Trend Pattern
24
10/8/2014
12
Cycling Pattern
Cycling pattern example:

UCL
LCL
Possible causes:
• Seasonal or periodic influences
25
10/8/2014
Mixture Pattern
Mixture pattern example:
UCL
LCL
0 10 20
Possible causes:
• Two or more overlapping process output distributions
• Tampering
26
10/8/2014
13
Shift in process level
Shift in process example:

UCL
LCL
0 10 2 0
Possible causes:
• Introduction of new inputs
• A change in inspection methods or standards
• A process improvement
27
10/8/2014
Trend pattern
Trend pattern example:

UCL
LCL
0 5 10 15
Possible causes:
• A cumulative effect over time, such as an increase in staff knowledge with
more experience
• A gradual effect due to noise variables, such as an increase in the cycle time to
process loans when volume increases
28
10/8/2014
14
Control charts for continuous data
29
10/8/2014
Overview of XBar R Charts
XBar R Chart example:
30
10/8/2014
15
Control charts for discrete data
Discrete
Defects Defectives
Equal Unequal Equal Unequal

subgroup size subgroup size subgroup size subgroup size
C Chart U Chart NP Chart P Chart

Number of Defects per Number of Proportion
Defects unit Defectives defective
31
10/8/2014
Characteristics of the U Chart
Review the chart below:
How would you interpret this?

What do you notice about the control limits? 32
10/8/2014
16
Characteristics of the P chart control limits
Look at the chart below:
What do you notice about the control limits?

Why might this happen? 33
10/8/2014
The key points to remember when using

control charts to monitor and manage a
process are: UCL
Identify the variables to monitor

Establish the sampling strategy.
Select the appropriate Control Chart.
Create and analyse the Control Chart. LCL
34
10/8/2014
17
When identifying variables to monitor using

Control Charts
The Data Collection Plan for the Control Charts
For rational sub-grouping, the rationale for
the subgroups, how your sampling plan will
show less variation than expected between
the subgroups
35
10/8/2014
CONTROL
Deploying the solutions and
closing your project
18
Deploy your solutions and close your project
Sampling
Deployment
strategy
Choosing the
Lessons
appropriate
Learned
control chart
Signals of Project
37
10/8/2014
After completing this lesson, you will be

able to:
• Describe the indicators of a successful
deployment of your solution and control plan
• Deploy the solutions and Control plan on a
full-scale basis
• Complete the required actions to close your
Lean Six Sigma project
• Explain the responsibilities of the Coach and
Finance Certifier in closing your project
38
10/8/2014
19
Deploying the solution
A successful deployment has three indicators:

The Process Owner “owns” the solution
and Control Plan
Training for the solution and
Control Plan is delivered
The deployment of the solution is verified
39
10/8/2014
Deploying the solution

Most organisations have their own implementation methodologies
and standards. At a minimum, after a successful period of systems,
integration and user acceptance testing a detailed task schedule
and/or Gantt chart will be prepared documenting:
Detailed implementation task list

Implementation task ownership
Timing
Task dependencies
Escalation
Notifications of task start/completion times
The task schedule may by controlled by the Project Manager or be

managed centrally by a Control Room
40
10/8/2014
20
Document the key lessons learned from your
project
What are lessons learned?
• Additional pieces of information discovered
during a work effort. They can be:
• Positive lessons (possible best practices)
• Negative lessons
• Lessons learned and best practices
are the basis for continuous improvement
of your Lean Six Sigma program.
• Lessons learned are captured:
• Informally throughout the project
• Formally at the close of the project
41
10/8/2014
Document the key lessons learned from your project
Sources of Lessons Learned:

• Lessons Learned can be identified from:
• The results of the project
• The process improved
• The use of the DMAIC methodology
42
10/8/2014
21
Other project closure activities
Your Finance Certifier should:

• Approve the Predicted benefits for your project
• Agree how the benefits will be recognised
• Mandate reporting frequency and responsibility
43
10/8/2014
The control phase tollgate must be completed
A project must pass the Control Phase tollgate review to

meet the criteria for closure.
• The Project Lead, Coach and Project Champion
must agree that the chartered project goals
were met.
• The Project Lead is responsible for completing
the tollgate review.
44
10/8/2014
22
Poor deployment can

undermine all the good work
completed on your project to date
Ensure you have a detailed plan for
implementation
Document and share any lessons
learned and best practise
Agree project benefits and recognition approach
Complete Control tollgate and gain formal approval to
close your project
45
10/8/2014

When deploying the solution/control plan
• The final Control Plan
• The strategy to transfer the Control Plan
to the Process Owner.
• The validation strategy for the Y after the
solution/Control Plan has been deployed
When closing your project

• The information required to complete the
Control phase tollgate review.
• Your lessons learned from the project.
46
10/8/2014
23

Sixgreen Courseware PDF

Uploaded by

Copyright:

Available Formats

Sixgreen Courseware PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sixgreen Courseware PDF

Uploaded by

Copyright:

Available Formats

Lean Six Sigma

Copyright © Property of Firebrand Training Ltd www.firebrandtraining.com

Be prepared to share with the class your:

FIREBRAND LEAN SIX SIGMA GREEN BELT

Number of Questions & Format

Objectives of this module

At the end of this module, you will be able to

Lean emerged from post-WWII Japanese

Ford Rouge Plant

Lean Thinking, continued

Lean Thinking, continued

Lean thinking is the dynamic, knowledge-

8 Forms of Waste - Continued

Over-processing Effort that adds no value to the service from the

Skills Not utilising people’s experience, skills, knowledge,

Reducing Waste & Ancillary Activities

Jaguar focussed on removal of the waste (braking)

What is Six Sigma?

Six Sigma is a 5-phase problem solving methodology that

Six Sigma is a data driven philosophy & process resulting

Definition of Six Sigma

What is Six Sigma?

What is Six Sigma? … and what is a standard deviation?

Variation exists in all processes.

The principal aim of Six Sigma is to reduce variation to

Definition of Sigma Level

Sigma Level states how many Standard Deviations lie between

How well can the pilot achieve the target performance?

Why Use Six Sigma?

Variation Errors & Poor Dissatisfied Business

Is 99% Good Enough?

3.8-Sigma Level 6-Sigma Level

• 5,000 incorrect surgical • 1.7 incorrect operations per

• 11.8 Million shares incorrectly • 4,021 shares incorrectly traded

• 10,700 defects per million • 3.4 defects per million

So what are the Factors for a x1 – coffee beans

Six Sigma Overview Summary

Six Sigma v Lean in a nutshell

Imagine a leaking pipe and consider how

Six Sigma tackles defects/rework, one of the

Lean Six Sigma Simulation

The coin simulation

Your instructor will read the instructions of the simulation

Deploying Lean Six Sigma in the business

Lean Six Sigma is customer centric. It strives to delight

The satisfaction of customer needs (internal

Why should leaders lend their support to your LSS

What should Lean Six Sigma leaders do?

Too many marginal-value projects being worked = Low ROI

Consistently use prioritisation and selection

Actively manage projects-in-process

Right Projects – Identify Value Levers

 Client Needs provides focus on  Financial analysis drives to

Revenue Growth Economic Profit

Strategic Focus ROIC Opportunity Areas Project Ideas

Reduce delivery time for…

Improve Initial Gathering of Requirements

Too broad to address Objectives that may be

Right Projects – Screen Initial List of