You are on page 1of 536

Workshop on Structured Problem

Solving Using DMAIC Methodology


Agenda

Day 1: Day 3:
TQM FMEA
Introduction to Six Sigma Process Capability
Control Charts
Define Phase
Day 4:
Project Charter
Lean Enterprise
Day 2: Day 5:
Seven QC Tools VSM Exercise
Process Mapping NTPC Case Study
Exam & Wrap Up
The Evolution of Quality

• Provides a framework for understanding


the history of the quality movement.
• Expands the definition of quality.
Juran’s Definition of Quality

Defined as “fitness for use” based on:


• Customer’s perceptions of product design.
• Degree to which a product conforms to
design.
• Product’s availability, reliability, and
maintainability.
• Available customer service.
ISO Definition of Quality

• Degree to which a set of characteristics


fulfills requirements

Requirements:
Product: Characteristic:
Convenience
Telephone Speed dial
and speed
Crosby’s Definition of Quality

• Quality is conformance to requirements.


• Requirements are answers to key
organizational questions:
– How quickly will orders ship?
– What is our return policy?
– What forms of payment are
acceptable?
Quality Evolution: Medieval Guilds

Year and Period


1200-1799 1800-1899 1900-1940 1941-1945 1946-Present
Guilds of Product Process Quality during Birth of
medieval Europe orientation orientation World War II total quality

Guilds:
 Developed strict rules for
products and services.
 Used stamps to identify
flawless goods.
Quality Evolution: Product Orientation

Year and Period


1200-1799 1800-1899 1900-1940 1941-1945 1946-Present
Guilds of Product Process Quality during Birth of
medieval Europe orientation orientation World War II total quality

 Master craftsmen trained


apprentices.
 Industrial Revolution divided
trades into specialized tasks;
inspectors guaranteed quality.
 Taylor system increased
productivity; inspection
departments found defects.
Quality Evolution: Process Orientation

Year and Period


1200-1799 1800-1899 1900-1940 1941-1945 1946-Present
Guilds of Product Process Quality during Birth of
medieval Europe orientation orientation World War II total quality

 Processes became critical.


 Shewhart identified statistical
quality control.
 Developed strict rules for
products and services.
 Quality became relevant for
process, not just product.
Quality Evolution: Wartime

Year and Period


1200-1799 1800-1899 1900-1940 1941-1945 1946-Present
Guilds of Product Process Quality during Birth of
medieval Europe orientation orientation World War II total quality

 Quality became a safety


issue.
 The military developed a
sampling inspection system
and trained suppliers.
Quality Evolution: Total Quality Movement

Year and Period


1200-1799 1800-1899 1900-1940 1941-1945 1946-Present
Guilds of Product Process Quality during Birth of
medieval Europe orientation orientation World War II total quality

 Developed in response to
Japanese quality movement.
 Focused on improving all
processes through people
who used them.
W. Edwards Deming

Quality keys:
• Understanding
customer needs
• Process
improvement
• Statistical analysis
• Expertise of workers
• PDCA cycle
Joseph M. Juran

Quality keys:
• Features that
satisfy customers
• Freedom from
deficiencies
• Juran Trilogy®
– Quality planning
– Quality control
– Quality
improvement
Kaoru Ishikawa

Quality keys:
• Company-wide
participation
• Quality control circles
• Advanced statistical
methods and tools
• Nationwide quality
control promotion
Armand V. Feigenbaum

Quality keys:
• Total quality control
• Integration of quality
development,
maintenance, and
improvement
• Focus on internal
and external
customers
Genichi Taguchi

Quality keys:
• Quality should be
designed in.
• Quality should
minimize
deviations from a
target.
• DOE optimizes
performance.
Philip Crosby

Quality keys:
• Conformance to
requirements
• Prevention
• Zero Defects
• Price of
nonconformance
Total Quality Management

Total quality management (TQM):

• A management approach

• Centered on quality

• Based on company-wide participation

• Aimed at long-term success

• Through customer satisfaction


3 Cs of TQM

1 Customer relationships

2 Continuous improvement

3 Company-wide participation
Customer Definitions

Pucker Up Lemonade, Inc. Consumers

Internal Customers External


Loading Dock Sorting Department Customers
Sorting Department Juicing Department
Juicing Department Mixing Department
Mixing Department Bottling Department
Levels of Customer Satisfaction
1

Noriaki Kano identified three


levels:
• Expected quality
• Desired quality
• Excited quality
Customer Feedback
1

Has two parts: Partnering with


 Efforts to capture customer:
what customers  Extension of listening
say about to customer feedback
company’s  Most direct route to
products/services customer satisfaction
 Efforts to drive
feedback back into
organization
PDCA
2
Act • Analyze reason
Plan
for not making
• A well-known desired results.
• Determine what • What to do.
model for changes to make to • How to accomplish it.
better achieve desired
continuous results.
• Standardize if desired
process results achieved.
improvement
is the Plan- • See if the desired results • Carry out the plan.
Do-Check- were obtained.

Act cycle.
Check Do
Company-Wide Participation
3

• Leadership must come from management.


• All employees must be involved.
• Employee involvement usually requires
employees to work in cross-functional
teams.
Employee Involvement
3

Benefits Barriers
• Improved • “It Won’t Work Here”
productivity and • Perception of loss of
cost reduction management
• Increased authority
participation and • Employees feeling
job satisfaction “used”
• Opportunities for • “Flavor of the
professional month”
development
Quality Benefits

• Tangible • Intangible
– Increase in – Customer goodwill
earnings – Alignment
– Decrease in waste between business
– Increase in activities
productivity
W. Edwards Deming on Quality

• Meeting customer needs + wants =


quality.
• Quality improves products/services and
processes.
• Improved products/services and
processes = profitability.
A Quality Approach Benefits . . .

Employees Organizations

Customers

Suppliers Society
Benefits to Employees
Pride in products and
 Product quality services
Job satisfaction

 Process quality Improved communications


Streamlined work
processes
 Customer satisfaction Happier customers
Strong customer
relationships
 Profit Greater job
security/benefits
Benefits to Organizations

Market
Share
Quality Profit

Cost

Organizations
Quality Studies and Standards

Released the Profit


Strategic Planning
Impact of Market
Institute
Strategy (PIMS) study.

National Institute of Partnered with the


Standards and Baldrige recognition
Technology program.

Both organizations support the link between


quality and profitability.
External and Internal Customers

Publication Sales
Department Department Customer
Benefits to Customers

Quality results in:


• Increased choices.
• Improved goods
and services.
• Expectations met or
exceeded.
Benefits to Suppliers

• Achievement of performance requirements

• Streamlined processes

• Efficient communication

• Increased customer satisfaction


Benefits to Society

Economic growth and stability

Increased employment opportunities

Product safety
Process Management

Quality improvements are


Before TQM applied to single processes
within manufacturing.

Quality improvements are


After TQM applied to all organizational
activities through process
management.
Organizational Process

Resources
• People
• Equipment
• Material
• Money
• Time

Process A
Input C P Output
D

Preceding Subsequent
Process Process

Customer Influence
Introduction to Six Sigma
What is Lean Six Sigma?
Introduction

• Six Sigma goal is process perfection through


defect reduction.
• Lean goal is cycle time reduction through
elimination of waste.

“Only those companies that eliminate their


defects will have what it takes to win.”

“Breakthrough companies strive for 100


percent DEFECT-FREE products and
services.”
Larry Bossidy
CEO of AlliedSignal
What is Six Sigma?
Methodology and Improvement Strategy

• Six Sigma is an overall strategy to accelerate improvements in


processes, products, and services–create breakthrough.
• Six Sigma measures how effective strategies are in eliminating
defects and variations from processes, products, and services.

= + + + ...

Y f(x1, x2, x3 …)
Process output (Y) is a function of (f) the inputs (Xs).
Understanding and controlling this relationship is a major
aspect of Six Sigma projects.
Focus on Variation
• Sigma (σ) refers to standard deviation, a measure of process
variation (smaller is better).
• Process Sigma is the number of units of standard deviations
between the process center and the closest specification limit
(larger is better).
• A Six Sigma process has six standard deviations (short term)
between the target and the closest specification limit.
Lower Target Upper
Spec Spec

Lower Process Upper


Control Center Control
6 Standard Deviations

6 Sigma
Sources of Variation

Desig
n

Materi Proces
al
$ s
Capabil
ity

Measurem
ent
System
Harvesting the Fruit of Six Sigma
Sweet Fruit
Design for Manufacturability

5 s Wall - Must Address Designs

ta
da
us
Bulk of Fruit

uo
Process Characterization
and Optimization

in
nt
Co
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 4- s Wall - Must Improve Internally
Discrete
data

Low Hanging Fruit


Seven Basic Tools

----------------------------------
3 s Wall - Demand improvement

Ground Fruit
Logic and Intuition

The walls crumble faster when


working WITH suppliers and
CONCURRENTLY addressing design
and process issues
The Focus of Six Sigma

f (X) Y
 Y  X1 . . . XN
 Dependent  Independent
 Output  Input-Process

 Effect  Cause

 Symptom  Problem
 Control
 Monitor
Would you control shooter or target to get the Gold
Medal at Olympics
Six Sigma Defines Problems Statistically

Y= f (X) Should we focus on


the process outputs
(Y) or inputs (X)?

The product is The process is


used to Y - Outputs X1 . . . XN - used to
evaluate the Inputs control the
process.  Dependent product.
 Independent
 Output
 Inputs,
 Effect
process
 Symptom
 Cause
 Able to
 Problem
Monitor
 Controllable
99% Good (3.8 Sigma) 99.99966% Good (6 Sigma)
20,000 lost articles of mail per hour Seven articles lost per hour

Unsafe drinking water for almost One unsafe minute every seven
15 minutes each day months

5,000 incorrect surgical operations per


week 1.7 incorrect operations per week

Two short or long landings at most


major airports each day One short or long landing every
five years
200,000 wrong drug prescriptions each
year 68 wrong prescriptions per year

No electricity for almost seven


hours each month One hour without electricity every 34
years
Define Phase
The Define Phase
The define phase module provides an
overview of the following tools:
1. Project selection
2. Project charter
3. Supplier, Input, Output, Customer (SIPOC)
diagram
4. Collecting Voice of the Customer (VOC)
Projects Must be …
Sorting Projects from Messes
• A mess is a morass of unsettling
symptoms, causes, data, pressures,
shortfalls, opportunities, etc.
• A problem is a well-defined situation that is
capable of resolution
• Identifying a problem from within the mess
is frequently the first step in the process of
project definition
Project Qualifications

• There is a gap between current and desired

performance.

• The cause of the problem is not clearly

understood.

• The solution is not predetermined.


Project Selection is Critical
• High leverage projects lead to largest $$ Savings
• Large returns justify the investment in time and
effort
• Developing a Lean Six Sigma culture depends
upon successful projects having significant
business impact
• Black Belt training depends on completion of a
meaningful project within ~ 6 months
• Future Projects are frequently identified through
initial projects
Lean Six Sigma Project Criteria
• Aligned with business objectives and plans
– Voice of Customer/ Critical to Satisfaction (CTS)
– Quality (CTQ) / Cost (CTC) / Delivery (CTD)
• Consistent with principles of Six Sigma
– Elimination of process defects
– Reduction of variation
• Concentrates on significant issues/opportunities
..... not “problem of the day”
• Justify the investment
Project Selection

“The best Six Sigma projects begin not inside


the business but outside it, focused on the
question — how can we make the customer
more competitive? What is critical to the
customer’s success? … One thing we have
discovered is that anything we do to make the
customer more successful inevitably results in a
financial return to us.”
Jack Welch
Address to the General Electric
annual meeting April 23, 1997
What is Customer Satisfaction?

• A comparison of expectation to
experience
• A matter of degree
• A result of a good match between
supply and demand
• A predictor of repeat business
Customer Needs Translated into
“Critical To” (CT) Characteristics
Customer needs are translated into product, service or deliverable
requirements in terms of quality, delivery and cost. The “Y” of Y=f(X).
Typical Critical to Characteristics

CTQ Critical to Quality

Y CTD Critical to Delivery

CTC Critical to Cost

Breakdown of• Projects are identified by the relationship between


the processes product, service, or deliverable requirements and processes.
required to
produce the • The process parameters that affect the requirements
product, are later identified (X , X , … X )
1 2 n
service, or
deliverable. • Leverage processes are identified.
Project Selection Criteria Aligned with
Pain area core objectives

Faster
Deployment
Stakeholder
Satisfaction

Repeatable
High probability
of success
S. Criteria Weight
No.
Higher
1 Aligned with core objectives 10
returns
2 High probability of success 10
3 Data Availability 8
4 Pain area 8
Process 5 Process Improvement 7
Improvement Ease of 6 Higher returns 8
implementation
7 Repeatable 6
Data Availability
8 Faster Deployment 5
9 Stakeholder Satisfaction 7
10 Ease of implementation 6
Project Selection Matrix Diagram

Project 1 Project 2 Project 3 Project 4 Project 5


S. Criteria
APC Reduction Boiler (Stage II) DM (Stage I) Make ESP (Stage I) Inlet Soot Blowing
No. in stage I Optimization up Optimization Duct Replacement Sy.Optimization
1 Aligned with core
objectives
2 High probability of
success
3 Data Availability

4 Pain area

5 Process Improvement

6 Higher returns

7 Repeatable

8 Faster Deployment

9 Stakeholder Satisfaction

10 Ease of implementation

Strong Relationship Moderate Relationship Weak Relationship


Factors that Improve Project Success

• Dedicated Black Belt


• Champion phase reviews
• Black Belt has knowledge of
process/product to be improved
• Historic data
• Clearly defined deliverables
• Committed process owners (skin in the
game) with the authority to modify the
process
Project Metrics – Success Criteria
Primary Metric
• Used to measure project success
• Consistent with the problem description and objective
• Plotted on a time series graph and shows the goal and actual
performance lines

Secondary Metric(s)
• Control unintended negative consequence (assures the Primary
Metric is not achieved artificially)
• May be used to measure project progress when the Primary Metric
responds slowly
• More than one may be required
• Plotted on a time series graph and shows the goal and actual
performance lines
8
Project Charter
Traditionally created by LSS Champion
Specifies details of a project including:
1. Scope
2. Responsibilities
3. Benefits
4. Schedule
5. Success criteria
A project charter template that may be adopted to fit
your organization is: Project Charter Template.
SIPOC
Supplier, Input, Process, Output, Customer diagrams
are used to:
1. Define the scope of the project
2. Identify key stakeholders
3. Gain a “30,000 foot” view of the process targeted
by the project
Tips:
1. The SIPOC is the first of several ways the
process will be documented. Therefore, it should
be at a relatively high level of abstraction.
2. It is a good way to assure agreement on the
scope of the project.
SIPOC Diagram – Questions to Answer
• Who are the customers of this process?
• What are their requirements?
• How are those requirements reflected in the
process parameters (output measures)?
• What are the process outputs?
• What are the process inputs that cause the
outputs?
• What controls are in place for the inputs?
• Who are the suppliers of the material for this
process?
• What are their requirements?
Purpose of Various Diagrams / Maps
• SIPOC diagrams provide a “forest view” of the
process – maintains focus on the customer’s
requirements.
• Value stream “trees level” describes the time, effort,
resources, and information used in the process – a
frequent source of early wins.
• Lay-out diagram (spaghetti diagram) documents
the distance an item travels during its production –
illustrates unnecessary movement.
• Process map “ground level” documents the inputs
and outputs of each step in the process – provides
the raw material for building a model of the
process.
The SIPOC Diagram – The Forest Level
• Start with the end in mind.
• Who are the customers of this process?
• What are their requirements?
• How are those requirements reflected in the
process parameters (output measures)?
• What are the process outputs?
• What are the process inputs that cause the
outputs?
• What controls are in place for the inputs?
• Who are the suppliers of the material for this
process?
• What are their requirements?
Sample SIPOC

Supplier(s) Inputs/Req'ts Process Output(s)/Req'ts Customer(s)

Grocery store Coffee machine Install filter 1800 Husband


Utility Measuring cup Measure coffee Specified Wife
Company Electricity Add coffee number of cups Guest
Appliance Qualified Add Water No grounds
store operator Turn on < one hour old
Water machine Dark
Filters Hot
Cream/milk Aromatic
Sweetener Fresh
Ground coffee Strong
SIPOC Validation

Review the SIPOC with your: team,


Champion and process owner(s) to assure
agreement on the SIPOC content as well
as the project scope and success criteria
among all stakeholders.
S I P O C CTQs

Plant Availability
Suppliers Inputs Process Outputs Customers UI Earned
Coal Internal Plant Load factor
Coal Field Power
Water -Plant Mgmt Heat Rate
Rihand Dam Effluent
Suppliers
Air
Process Ash
-Corp Mgmt AUXILLIARY POWER
-
Power -Employee CONSUMPTION
Steam
Manpower Preferred Employer
CO2 External
Lubricants R&R
-Cent.Gov.
Maintenance Man : MW Ratio
-Ministry
Services Lead Time
-State Gov.
Spares and Manpower Utilization
-Shareholders
Consumables Quality System
-Contractors
Social Responsibility
-Suppliers
Maintenance Cost
-PAPs
Overhead Expenses
-Labour
Environment ENERGY EFFICIENCY
Boiler
DM Water Used
Turbine Safety aspect
COAL Generator Training of manpower
Cycle Time
AIR STEAM GRID In process idle stock
Air Emission quality
WATER Noise level
Collecting VoC
• Who is a customer?
• What does the customer need?
– Gathering the Voice of the Customer
– Define customer requirements
• How do the customers prioritize their needs?
• How are customer needs translated into CTQs?
 Generallya less intense exercise for DMAIC projects than for
DFSS projects
 More likely to be based on information that is already
available internally for DMAIC projects than for DFSS projects
Who Is a Customer?
Process

Voice of Step Step Step Product/


Customer Service

Customer

A customer sets/affects requirements for your product


or service.
• External
• Buying customers
• End-users
• Regulatory agencies
A customer is one who receives your output.
• Internal
Customer Segmentation
• Generally, external customer needs are more important
than internal customer needs.
• Are all customers equally important?
– External vs. internal
– Customer segments
• Regions
• Type of business
• Volumes
• Profitability
• Strategic market
• Future potential
Voice of the Customer
• The Voice of the Customer (VOC) is the
starting point of any project and data
collection plan.

• The Voice of the Customer includes:


– Expectations

– Requirements

– Opinions
Questions to Find Voice of the Customer
• What are the elements of your business that are
the most critical, from the perspective of your top
or best customers? What are their relevant
needs?
• What data has been collected to understand the
customer requirements?
• How do you operationally define the defect from
the perspective of the customer? Under what
conditions does it occur?
Voice of the Customer Concerns
• Real vs. stated needs

• Perceived needs

• Intended vs. actual usage

• Internal customers vs. external customers

• Effectiveness vs. efficiency needs

• Change over time


Determining Customer Requirements
• Use verbatim comments from customers to
help determine the key customer
requirements
– We often receive many verbal comments from
our customers.
– We need to look for ways to probe for deeper
meaning behind the comments in order to
translate these comments into what the
customer actually requires.
Tools for Gathering Voice of the Customer

Unsolicited data from customers


– Complaints
– Field reports
– Trade journals
– Benchmarking
– Internal research
• Requirements documents
• Contracts
• Customer observation
• Be a customer
Tools for Gathering VOC
Solicited data from customers

– Interviews

– Focus groups

– Surveys

– Informal customer discussions

– Market research
Determining Customer Requirements
• Review customer verbatim comments and
comparative data
• If possible, probe for deeper understanding
• Convert into terms of process performance
• Describe the actual customer requirement
– Write the requirement, not the solution
– Use measurable terms
– Identify performance targets
– Be concise
Define Phase Review
• The purpose of the define phase is to identify and
launch a project.
• VOC should be reflected in the project charter
especially in the project success criteria.
• The Black Belt and Champion should review and
agree on the details of the project charter and
SIPOC.
• A common cause of project failure is poorly
defined or projects with excessive scope. This
review is an opportunity to mitigate that risk.
VOC ( Voice of Customer ) D M A I C

HOW ?

Replace with eff. pro


Arrest Pro.Deviation
Improve Processes
Competitor

Better Systems
CustomerNeed
Customer Needs
WHAT ?

Aux Power Consumption 5 5 5 4 4


Generation & Dispatch 5 4 5 4 2
Environmental Pollution 4 4 5 4 4
Conservation of Coal 4 5 5 2 5
RATING 81 90 64 66
D M A I C
Project Charter
D M A I C

Resource Plan General Information Project Review Dates


Champion: Mr. P.K.Mohapatra Tollgate Date Signoff
(xx/xx/xx)
Process Owner: Mr. N.N.Mishra
Process Owner: Mr. N.K.Sinha Define 05/02/07 Yes
Quality Leader Mr. S. Mathew Big Y:
Coach (BB): Mr. S. Banerjee Measure 05/04/07 Yes
Green Belt: Mohit Yadav Analyze Yes
15/04/07
S. Sinha TMD Iswar BMD Reduction in APC of Improve Yes
V. Agarwal EEMG Ashish Jain EMD 31/05/07
Stage-I units Control
Green Belt Contact Information : 31/08/07 Yes
Mohit Yadav 9425823299
1550

Project Team Rhythm and Review Rhythm and Review:


On-Track
Meeting Frequency: 15 Days Off-Track
Need Attention
Mandatory Attendees: Mohit Yadav V. Agarwal Iswar S. Sinha A. Jain

Optional Attendees: Mr. P.K.Mohapatra Mr. N.N.Mishra Mr. N.K.Sinha Mr. S. Banerjee
VOC
( Voice of Customer )
The Seven QC Tools
 Check sheet
 Stratification
 Pareto diagram
 C &E Diagram
 Histogram
 Scatter diagram
 Graphs and Charts
Area of Uses

Problem Solving
Continuous Improvement-- Kaizen
Dispersion Control-- Six Sigma
Waste Elimination--Lean
QMS, EMS, TS 16949, OHSAS-Process
control, C&P
Supplier Development
Project Management--Team working
What is their role ?

In problem solving
Role

Tool Role they Play

Data gathering Quantify current status


or magnitude of the
Problem
Check sheet Facilitate data gathering

Stratification Identify and segregate


different sources of the
problem
Role

Tool Role they Play

• Pareto diagram • Prioritize the problem

• Generate many ideas for


• Brain storming
solving a specific problem
• Cause & effect • Identify possible causes of a
diagram problem in a structured way
Role

Tool Role they Play

• Histogram • Study pattern of variation in


a set of data.
• Study relationship between
• Scatter diagram 2 types of variable
• Visual display of data
• Graph & chart
DATA GATHERING

Data Collection
What is Data ?

Data is a numerical expression of


an activity
Conclusions based on facts and data
are necessary for any improvement.
K. Ishikawa

If you are not able to express a


phenomenon in numbers, you do not
know about it adequately
Lord Kelvin
Types of Data

Quantitative Qualitative
•Measurable •Subjective assessment
 e.g. :Length,  e.g. :Score in a
Temperature beauty contest
•Countable
 e.g. :Number of
defects
Population, Sample and Data

Action

Random
X
Measurement
Population Sample Data
Sampling / Observation

Action
A Saying
 When you see the data, doubt it
 When you see the measuring instrument, doubt
it.
 When you see the chemical analysis, doubt it.
 Three Categories
1 False Data
2 Mistaken Data
3 No Data available
How to Collect Data?
 Define the purpose (Follow 5W 1H approach).
 Define the period for data collection.
 Define the Stratification.
 Design the check sheet and assess Measurement
System Capability.
Purpose
Be Clear on What, Where, When, Why, Who and How
the data should be generated
STRATIFICATION
Stratification
 Method of grouping data by Common
points or characteristics

 A Filtration Process for isolating the


cause of a problem.

 Prevent mix up and helps in easy &


faster identification-
Basis for Stratification

Workers Material
Machines Time
Defect Environment
Product Region
 Folder Files
Machine A Machine B
Machine A Machine B
N = 450
N = 450 N 450
= 450
N=
D = 12D = 12 D =D 1118
= 118
P% = 2.7 P% = 26.2
P% = 2.7 P% = 26.2

COMBINED

N = 900
D = 130
P% = 14.4
Blister Defect

UCL

CL

Pinhole Defect
UCL

CL

All Defects
UCL

CL
CHECKSHEET
Check sheet

 A convenient and compact format


for data collection

 A Simple rule– Maximum


information with minimum writing
efforts and easy to fill
Check Sheet For Machining Operation

Location Group Date Shift


M/C Comp. DRG. Token Op. Qt. Material Defect No. M/c Defects Total Remark
No. No No. Prod. Insp.

123456 A B C D………P Q R

Material Defect Machining Defect

1: Blow holes A: Dia.+ G: Length+ M: Oblong


2: Cracks B: Dia.- H: Length - N: Taper
3: Hard Metal C: Ch+ I: Sp+ O: Hole Shifted
4: Eccentric D: Ch- J: Sp- P: PCDV
5: Others E: CDV+ K: D& T Size+ Q: Poor Finish
6: Total F: CDV- L: D & T Size- R:105
Others
Graphs & Charts

Graphs represent data pictorially. A picture can


see what 1000 words can not tell.

Line chart Pie chart


Bar chart Gantt chart
Multiple bar Chart Pareto diagram
Component bar Scatter diagram
Chart Control chart
Radar chart
Bar Chart

Bar graphs are parallel bars of identical width but differing


length to compare size of different quantities / things.

Line Chart

Line graphs manifest the overall trend in time series data


by direction of their lines.

Pie Chart

Pie charts makes it easy to grasp the breakdown of the


components of a quantity over a certain period.
Comparison of Machines A & B
for weekly Rejection

25
21
20
20 20
Multiple bar chart
% Rejection

15 15
15 14
13 13
11 11 11
10 10
10 8
6
5
5 4
3 3
2

0
1 2 3 4 5 6 7 8 9 10
Week Number

Comparison of Machines A & B


for Units Produced 900
800
Units Produced

700
600 422
257
500 358

Component bar chart 400 244


336 281 317 299

300 221
275
200 435
375 321 348
263 285 307 294
100 201
133
0
1 2 3 4 5 6 7 8 9 10
Week Number
Pie Chart for
Customer returned watches

G
F 5%
E 3%
• A – Glass Broken 4%
D
• B - Stop 6%

• C - Mvt. Trouble A
43%
C
• D - Defective Dial 12%

• E - Regulation
• F - Stem Loose
• G - Others B
27%
Control Chart
X- Bar and R Chart

Xbar/R Chart for C1

2854
UCL=2853
Sample Mean

2849 Mean=2849

2844 LCL=2844

Subgroup 0 5 10 15 20 25

UCL=16.41
15
Sample Range

10
R=7.76
5

0 LCL=0
Radar Chart on ISO 9001-1994 Implementation

1
20 80 2

19 70 3
60
18 50 4
40
30
17 5
20
10
16 0 6 Series1

15 7

14 8

13 9

12 10
11
Gantt Chart for Construction Activity
Weeks
Type of work 1 2 3 4 5 6 7 8 9 10 11 12

Foundation work
Frame work
Dry-walling
Exterior touch up
Sheetrock work
Plumbing
Electrical wiring
Fit Fixtures
Paint interior wall
Interior touch up
Inspection delivery

Gantt Charts makes it easy to understand


the details of a plan and progress in its
implementation schedule.
Pareto Diagram

Vital few from Trivial many


94.6 100 100
87.5
50 90
76.8 80
40 70
60.7
No. of complaints

Cum Percentage
Pareto Analysis of 30
41.7
60

50

Customer Complaints 20
40

30

10 20

10

0 0
Fish not Vegetable Bread Cashier Meat not Eggs rotten
fresh wilted Stale Rude Fresh
Pareto Principle

 80% of problems are caused by less


than 20% of probable causes
 Establishes proof of the need
 Identifies vital few
PARETO ANALYSIS: Outstanding branch wise

BRANCH BRANCH Rs. DUE RATIO TO CUM. %


TOTAL

A= 5.0 E 45.0 51.7 51.7


B= 2.5 D 20.0 23.0 74.7
C=10.0 C 10.0 11.5 86.2
D=20.0 A 5.0 5.7 91.9
E=45.0 B 2.5 2.9 94.8
F=1.5 F 1.5 1.7 96.5
G=1.0 H 1.5 1.7 98.2
H=1.5 G 1.0 1.2 99.4
Oth.=0.5 OTHERS 0.5 0.6 100.0
87.0 TOTAL 87.0 100.0
Pareto Analysis on Outstanding

94.8 96.5 98.2 99.4 100 100


91.9
50
86.2 90

74.7 80
40 70
Outstanding Value

Cum Percentage
60
51.7
30
50

40
20
30

20
10
10

0 0
E D C A B F H G O

Branches
Pareto Diagram for Production Stoppage
100
A. M/C quality change
45 90
B. Intermediate conveyor
C. Power failure 40 80
D. Hopper/duct line jamming
E. Dryer drum coupling pin 35 70
F. B P full press problem

No. of stoppages
30

Cumulative %
G. Dryer preventive 60
H. Nip roller
25 50
I. Rotary comb tripped
J. Comber jamming 20 40
K. Al conveyor idle roller
L. Fire 15 30
M. Accumulation
10 20
N. Drum seal changing
O. Fan tripping 5 10
P. Chain problem
Q. Fiber jamming 0 0
R. Zone gear box A B C D E F G H I J K L MN O P Q R S

S. Zone conveyor
Pareto Analysis of Complaints at a Laundry

200 93 98 100 100


180 85 90
160 75 80
140 70
60
No. of complaints

120 60
100 50
80 35 40
60 30
40 20
20 10
0 0
Late Missing or Fading Stains Creased Buttons Stretched
delivery wrong colours Missing or torn
items
Brain Storming

generating large number of ideas


by a group of people
Basic Rules

 Defer evaluation
 Fantasize freely
 Generate quantity
 Build on ideas
Defer Evaluation

Put critical faculties in cold storage


- even constructive criticism.

Ensure a proper climate for acceptance of


all sorts of ideas.

No idea should be treated as stupid.


Fantasize Freely

Don’t operate with your brakes on.


Participants are encouraged to generate
ideas, no matter how fanciful they are.
Generate Quantity

Generate as many ideas as possible.


A pearl diver will be more successful in
finding pearls, when he brings up 200
oysters than when he surfaces only 15-20
oysters.
Build on ideas

Idea of one participant is more


effectively built up by another
participant.
Steps in Brainstorming

Select the topic

Each member, in rotation gives ideas

Member offers only one idea per turn,


regardless of how many he or she has

Continue till all ideas are exhausted

Ideas are recorded and displayed


Benefits

 Individual is limited in generating ideas and group


produces more ideas

 Ideas are improved upon by members

 Presence of others increases creativity

 Pooling of ideas and resources is made possible by


coming together as a group
CAUSE AND EFFECT
DIAGRAM
C & E Diagram
Graphic tool to represent relationship
between an effect and influencing
causes

There can not be an effect without a


cause.

Reduce incidence of subjective


decision making.

Identify main causes X’s influencing Y


Construction of C & E Diagram

Define problem
Gather members for discussion
Conduct Brainstorming
Group causes into 4M’s
Man, Material, Machine, Method
For each cause, ask, “What goes wrong
that produces the effect”.
Identify major causes
Cause and Effect Diagram for high petrol consumption

Procedure Driver Vehicle

Spark plugs
Impatience Heavy
Poor Bad Contacts
Craze Life
anticipation attitude
Body Technical
Wrong Poor details
skill Shape
Always gears Fuel mix
late Lack of Inexperience High H.P Carburetor
awareness Wrong
Riding on culture Engine
clutch
Cylinders High Petrol
Spurious Consumption
Crossings
Spares Impurities
Restrictions Traffic Incorrect
One way
No turn Tyres Inferior Octane no.
Frequent Petrol
Faulty
Circuitous stops Negligence
Speed Breakers pressure Additives
Road
Ignorance
Potholes Irregular Incorrect viscosity
Low pressure
Poor servicing
Clogged
condition Oil
False filters
Steep economy Not changed
Low level
Road Maintenance Materials
Cause & Effect Diagram

PROCESS MATERIALS

COOKING QTY OF WOOD QLTY


TIME COOKING WATER (OLD/FRESH) VARIATION
TEMP. IN
PENTOSANS
IN
FINAL PULP
INSTRUMENT
TRAINED UNTRAINED ACCURACY

PERSONNEL EQUIPMENTS
Uses of C & E Diagram

Trace out real root cause


Help evolve countermeasures
Making C & E an education in itself
Everyone participating, learn more
about their work.
Is a focus for discussion.
Shows level of expertise available.
Can be used for any problem
30

25 24

20 19
17
Frequency

15
12
11
10 9

6
5 3
2

0
1.776 1.868 1.96 2.052 2.144 2.236 2.328 2.42 2.512
Consumption(KWh)

HISTOGRAM
Histogram

Method of analyzing data


Data is condensed in a table
Tabulation is known as frequency
distribution.
Presented by a Graph displaying
distribution of data
Histogram

Graph is Characterized by 3 constituents


• centre ( mean)
• width (spread-variation)
• over all shape 30

25 24

20 19
17

Frequency
15
12
11
10 9
6
5 3
2

0
1.776 1.868 1.96 2.052 2.144 2.236 2.328 2.42 2.512
Consumption(KWh)
Histogram Construction
 Select a sample of min. 50
 Record the measurements.
 Determine the range.
 Decide the no. of classes.
 Divide range into no. of classes
 Determine boundary or class limits.
 Prepare frequency distribution.
 Construct histogram (GRAPH).
Data on Metal Block thickness (in mm)
3.56 3.46 3.48 3.50 3.42 3.43 3.52 3.49 3.44 3.50
3.48 3.56 3.50 3.52 3.47 3.48 3.46 3.50 3.56 3.38
3.41 3.37 3.47 3.49 3.45 3.44 3.50 3.49 3.46 3.46
3.55 3.52 3.44 3.50 3.45 3.44 3.48 3.46 3.52 3.46
3.48 3.48 3.32 3.40 3.52 3.34 3.46 3.43 3.30 3.46
3.59 3.63 3.59 3.47 3.38 3.52 3.45 3.48 3.31 3.46
3.40 3.54 3.46 3.51 3.48 3.50 3.68 3.60 3.46 3.52
3.48 3.50 3.56 3.50 3.52 3.46 3.48 3.46 3.52 3.56
3.52 3.48 3.46 3.45 3.46 3.54 3.54 3.48 3.49 3.41
3.41 3.45 3.34 3.44 3.47 3.47 3.41 3.38 3.54 3.47

Range= Max. – Min.=3.68-3.30=0.38


N=100
No. of classes= 9 Class width= 0.5
Frequency Table
Class no. Class Boundaries Mid-value Frequency

1 3.275 – 3.325 3.30 3


2 3.325 – 3.375 3.35 3
3 3.375 – 3.425 3.40 9
4 3.425 – 3.475 3.45 33
5 3.475 – 3.525 3.50 37
6 3.525 – 3.575 3.55 10
7 3.575 – 3.625 3.60 3
8 3.625 – 3.675 3.65 1
9 3.675 – 3.725 3.70 1
Histogram for Metal Block Thickness

45

40 37
35 33

30
Frequency

25

20

15
9 10
10

5 3 3 3
1 1
0
3.3 3.35 3.4 3.45 3.5 3.55 3.6 3.65 3.7
Thickness (in mm)
Histogram for Bearing Thickness

45
41
40

35
31
29
30
Frequency

25 22
20 18 17
15 12
9
10
5
5 3

0
5.24 5.28 5.32 5.36 5.4 5.44 5.48 5.52 5.56 5.6
Thickness (in mm)
Histogram for Energy Consumption

30

25 24

20 19
17
Frequency

15
12
11
10 9

6
5 3
2

0
1.776 1.868 1.96 2.052 2.144 2.236 2.328 2.42 2.512
Consumption(KWh)
Types of Histograms

Bell shaped
Symmetrical shape with a peak in middle
representing a normal histogram
30

25 24

20 19
17
Frequency

15
12
11
10 9
6
5 3
2

0
1.776 1.868 1.96 2.052 2.144 2.236 2.328 2.42 2.512
Consumption(KWh)
Skewed to Left & Right

Skewed to Left
Caused by centering the process toward high end of the
tolerance
Skewed to Right
Caused by centering the process toward low end of the
tolerance..
Bimodal & Truncated

Bimodal : Two combined populations-- two shifts,


operators, inspectors, suppliers, machine settings,
gages, tools, machines, measurement locations, etc.

Truncated: This can happen when a process is not


capable of meeting the specifications, parts are sorted
from both ends, or too few classes are chosen.
Missing Centre Spike's at Tail (s)

Missing Centre : Centre of the distribution has been


sorted from the rest. Portion may have been delivered
to a customer with tighter specifications.

Spike's) at the Tail (s) : Parts in outer ends of


distribution are probably being reworked to bring
characteristic just within specifications.
The Need for Transformations--Prediction from a ND is
possible
… Skew distribution Transformation

Before XTrans = 1/XRaw

After
Histogram Uses

To know--whether variation in data


is due to chance or assignable causes.

To tell about Process Behavior

--about its capability to produce defect


free output
Sources Of Variation

 Common Causes
---Chance Causes Of Variation
 Special Causes
--- Assignable Cause Of Variation
Common Cause
 Consists of combined effect of several sources of
uncontrollable variation inherent to a process.

 Collective influence of common cause variation


defines natural process fluctuation and is known
as Chance causes of variation.

 Process output is predictable

 Process is said to be in Statistical Control


Special Cause
 Variation has a large impact on performance.

 Determination of source of impact makes


cause "assignable." and is termed as

assignable cause of variation.

 If they exist, process or key characteristic is


said to be "out-of-control".
 Out-of-control process is not predictable
Process Behavior
 It tells whether process is under control?
 Is it producing defect free output?
Process out
----process under Of control
control or
LSL USL LSL USL
30

25 24

20 19
17
Frequency

15
12
11
10 9

6
5 3
2

0
1.776 1.868 1.96 2.052 2.144 2.236 2.328 2.42 2.512
Consumption(KWh)
process variability process variability
Good Process Behavior

Shape close to normal curve


Mean at target value
Spread within Specification limits
Cp is greater than 1.67 and Cpk is
greater than 1.33
SCATTER DIAGRAM
What: To study the possible relationship
between two variables.
Why: Diagram make it clear whether a
relationship exists, and shows
the strength of relationship.
When: To test a theory that the 2
variables are related.
Examples:

Cutting speed and tool life


Breakdown and equipment age
Temperature & lipstick hardness
Temperature and percent foam
in soft drinks
Hardness and tensile strength
Different Scatter diagram Patterns
Scatter Diagram on Conveyor Speed vs. Severed Length

1050
1045
1040
Severed Length (mm)

1035
1030
1025
1020
1015
1010
1005
1000
5 5.5 6 6.5 7 7.5 8 8.5 9
Conveyor Speed (cm/sec)
Uses:

Control Purpose
Replacing a destructive test by a
non-destructive test
Study of Cause & Effect
relationship
Process Optimization
Process Mapping
About This Module…

Process Mapping is a tool used to:

 Clearly define processes

 Identify areas where data collection should take


place

 Visualize activities involved in a process at the early


stages of project development

\DataFile\ProcessT.ppt
Six Sigma, A Quest for Process Perfection
Attack Variation and Meet Goals
What We Will Learn …
1. The importance of process maps and the character of
the product and process parameters.
2. The when, why and where to use process maps.

3. The x’s & y’s and X’s & Y’s and Y=f(x,X).

4. What to measure and control.

5. The need for process maps prior to FMEA’s, Gage


Studies, DOE’s and SPC.

6. When the process map is completed.

7. How to use the tool for your process.

Extending Flow Charts to Process Mapping


Process Management

Basic
Process
Model Resources
• People
• Equipment
System • Material
• Money
• Time

Input Process A Output


Input Process Output
C P

Feedback
Cycle Time

• Time it takes to complete a process from


beginning to end
Question

“How does a reduction


in cycle time benefit an
organization?”
Fundamentals of Process Mapping

A process map should be


A process map should describe: reviewed frequently and is
 Major activities/tasks never done.
 Sub processes A process map should
 Process boundaries document how the process
actually operates, not how it
 Inputs
is supposed to operate.
 Outputs (“As is,” not “To Be”)
 Process & Product Parameters
 Customers & Suppliers A process map will identify
opportunities for quality
 Process owners improvements.
Principles of Process Management

• Establish ownership.
• Verify and describe the purpose of the
process.
• Define the process, boundaries, and
interfaces.
• Organize and train the process
improvement team.
Principles of Process Management

• Define and document the process.


• Define points of control.
• Establish process measurements.
• Improve process.
Process Mapping Steps
1. IDENTIFY INPUTS AND OUTPUTS
– Identify Inputs (raw material, equipment, energy, 6M’s, etc.)

– Identify Outputs (measurable/assessable end product parameters)


2. SHOW ALL STEPS
– Value adding steps have the following characteristics:
 Something the customer would be willing to pay for
 Transforms the product or service
 Done right the first time
– Non value-added steps in the process are presented graphically:
 Evaluation points
 Rework points
 Scrap points
 Inventory
Steps (cont.)
3. SHOW OUTPUTS OF EACH STEP
– Show after each process step the characteristics
that can impact the following step(s).

4. SHOW ALL PROCESS PARAMETERS AT EACH


STEP
– List under each step the parameters that can
change a product characteristic at that step (i.e.,
parameters that can be controlled at that step).
More Steps
5. CLASSIFY THE PARAMETERS
– Classify the process parameters identified (in #4
above) into the following categories:
N = Noise Factors - Uncontrollable - May be controllable, but
are not controlled by decision.

C = Controllable factors - Process factors that can be


changed to see the effect on product characteristics.

S = Standard Operating Procedures - A procedure is used to


define and run those factors. Tooling, Fixtures.

CR = Critical Factors - Determined through FMEA, DOE, etc.


Inputs & Outputs

High Level Process Map


X’s Y’s

INPUTS PROCESS OUTPUTS


Processes Come in Hierarchies
Process - Level #1

Step #1 Step #2 Step #3

Process - Level #2

Step #1 Step #2 Step #3

Process - Level #3

Step #1 Step #2 Step #3

Select the appropriate process level.


Product and Process Parameters

Inputs STEP OF PROCESS Outputs

Process Parameters, x’s


y = f(x)
Remember the 6 M’s
 Man (People)
 Machine (Equipment)
Product Parameters, y’s
 Method (Procedures)
KEY for (x’s)
Process Parameters
 Material
N Noise Parameters
C Controllable Process Parameters
 Measurement S SOP Parameters
CR Critical Parameters
 Mother Nature (Environment)
Why List the Parameters?

Defects
TO REDUCE DEFECTS!
 The x’s and X’s are the sources of variation in your
process.
 Variation causes defects.
 The x’s must be under control to prevent defects.
 The root cause of a defect is variation of the x’s!
 The y’s and Y’s are the measured results of the process
and include the failure modes of the process.
 Defects are also outputs of a process step.
What Are We Measuring?
Measure the x’s, not the Y’s !

X’s _________
Inputs ?
x’s _________
Process Parameters
?
Process Step Outputs
y’s _________ ?
Y’s Process Outputs
_________ ?

We cannot control what we don’t measure!


Is Workmanship an x?
X’s Y’s

INPUTS PROCESS OUTPUTS


Process Parameters, x’s
WORKMANSHIP ??
y = f(x)
OPERATOR ??

6 M’s reminders:
 Man (People)
 Machine (Equipment) Product Parameters, y’s
 Method (Procedures)  DEFECT FREE OUTPUT
 Material
 DEFECTS IN PROCESS OUTPUT
 Measurement
 Mother Nature (Environment)
What Are the x’s and y’s?
X’s Y’s
Inputs Outputs
6 M’s reminders:
Process Parameters, x’s


Man (People)
Machine (Equipment)
N y = f(x)
 Method (Procedures) N
 Material
N
 Measurement
N Product Parameters, y’s
 Mother Nature (Environment)
C
C
KEY for (x’s)
Process Parameters
C
N Noise Parameters
C Controllable Process Parameters C
S SOP Parameters
CR Critical Parameters
S
S
Identifying Product & Process Parameters
The following may be helpful to identify process
parameters that have a potential effect on the product
parameters and process output:
 Brainstorming Remember the 6 M’s
 Man (People)
 Literature review
 Machine (Equipment)
 Operators manuals  Method (Procedures)
 Material
 Work Instructions
 Measurement
 Operator experience  Mother Nature (Environment)

 Customer / supplier input


 Engineering knowledge
 Scientific theory
Completeness Checks
Inputs Step of the Process Outputs
Process Parameters, x’s y = f(x)
Are there y’s for every x in this step?
Product Parameters , y’s
Is there a “good” type of y for every x ?
 Good
Is there a “bad” type of y for every x ?
 Bad = Defects
Are x’s here that impact downstream y’s?  Good Rejected = Defect
Does the map have input of extended team?  Bad Accepted = Defect
 Horror Stories: What has
happened in the past that
Remember the 6 M’s caused disasters?
 Man (People)  Success Stories: What
KEY for (x’s)
Process Parameters  Machine (Equipment)
outputs of this step thrilled
N Noise Parameters
the customer(s)?
C Controllable Process Parameters  Method (Procedures)
S SOP Parameters  Are there x’s for each y in
CR Critical Parameters  Material this step?
 Measurement
 Are upstream x’s changing
 Mother Nature (Environment) y’s of this step? How?
Wave Solder Process
INPUT LOAD IN LOAD ON FLUX
FLUX TO PREHEAT
FIXTURE CONVEYOR AIR KNIFE

X’s  Masking  Conveyor speed Pressure Pressure of air knife


Boards with Operator Rail position over pot Flux Brand Orientation
Board thickness Flux Type Distance to board
components Conveyor angle
4-Corner support  Thinner Air Knife used
 Solder Time req’d in Solder
Bd. Spacing in Fixture Cleanliness of flux Angle of air knife
 Flux Bd quantity in fixture 4-Corner Fixture support
 Electricity Tritration level Temperature
Bd. position within fixture Conveyor drive
 Machine Setup Front or rear Stone Type
smoothness
Side Cleanliness of Stone
Finger condition Height over Stone
Fixtures/conveyor width
Bd Orientation Rail Straightness Temperature
Solder in all holes
Frame size/Board size Humidity
Solder Bridging
Fixture mass. Ambient Temp Solder Insufficients
Fixture dimensions
Immersion depth Fire
Vert. location bd.in fix.
Raised Carrier Frame corner Qty of Excess Flux
Component Orientation Lead Solderability
Poor wetting
Component Density Amount on Bd.
Partial filled holes
Flux through holes to top side
Skip Soldering
Distribution on Bd.
Bd Orientation Bridging KEY for (x’s)
 Excessive Solder
Bd Spacing Excessive solder speed Process Parameters
 Bridging
Bd Alignment Insufficient solder  Noise Parameters
 Icicles
Bd to Rail Position Bridging  Controllable Process Parameters
 Partially filled Holes  SOP Parameters
Critical areas masked Cycle Time  Critical Parameters
Cleanliness of Board
Current Setup: Specifications
Flux: Hot Air
Kester Pre Heat #1 Pre Heat #2 Pre Heat #3 Solder Emmersion Conveyer Knife
2% Solid FACTOR Temperature Temperature Temperature Temperature Depth Speed Angle Angle Pressure
S.G.: NA RANGE - - - - - 2.75-4.25 - 45-65 12-20
NOMINAL 400 445 470 490 1/4" 3.75 N/A 60 17PSI
Wave Solder Process (cont.)
PREHEAT PREHEAT PREHEAT SOLDER HOT EXIT OUTPUT
#1 #2 #3 POT AIR KNIFE WAVE SOLDER INSPECTION
Y’s
Solder Temp  Workmanship
 Temp - Zone 2  Temp - Zone 3 Position Boards:
 Temp -Zone 1 Ht of Pot/Position Stds REWORK

Temp - Zone 1 Temp - Zone 1 Dist to board •Accepted


Temp - Zone 2 Solder Pot Angle  Samples
Temp -Zone 3 Temp - Zone 2 Temp •Rejected
Temp - Zone 3 Exit Point  Lighting
Time in Preheat 2 Time in Preheat 3 Angle  Magnification
•Scrap
Time in Preheat 1 Amount of dross SCRAP

 Resp time of heater  Response time of Air Pressure  Prod/Dmnd/Sch


Resp time of heater heater Solder Pump Speed
Stability of temp Temp Dist  Census/Inspection
Stability of temp Stability of temp Solder Height Staffing
 Distance to board  Dist to board  Distance to board Solder Pump Pressure .  Inspection
. Sample
Temp distribution Temp distribution Temp. Distribution Choke bar setting/adj (%)
Fixture warp Fixture warp Fixture warp Solder contact  Eyesight
Rail warp Rail warp Rail warp Solder type  Inspector Variation
Solder tin content Solder Distribution
Solder Contam Coverage on board
Baffle qual, hole wear, Solder Appearance
bent condition, clean Solder joint Quality
Board Deflection y’s
Board Temp Board Temp Board Temp
Product
Flux Condition Flux Condition Flux Condition
Paramenters
Flux Activated Flux Activated Flux Activated
Flux Solvent drive off Flux Solvent drive off Flux Solvent drive off Excess Pressure Solder Appearance
Thermal Shock Thermal Shock Thermal Shock  Solder Bridging Solder Joint
 Board Warping  Board Warping  Board Warping KEY for (x’s)  Insufficient solder/Opens Quality/Defects
Process
Excess Heat Excess Heat Excess Heat Parameters
Insufficient Pressure Open Solder Joints
 Excess Solder  Excess Solder  Excess Solder  Solder Bridging Good Boards
 Noise Parameters
 Poor Fillets  Poor Fillets  Poor Fillets  Controllable Temp Too Low Dirty Boards
 Excess flow thru  Excess flow thru  Excess flow thru Process  Solder Bridging Scrap Boards
Inadequate Heat Inadequate Heat Inadequate Heat Parameters Temp Too High Hot Boards
 SOP Parameters
 Solder splatter  Solder splatter  Solder splatter  Insufficient Solder/Opens Damaged Components
 Critical Parameters
 Trapped Gas  Trapped Gas  Trapped Gas  Damage Board Lifted Components
 Solder Bridges  Solder Bridges  Solder Bridges Angle Too Steep
 Poor flow thru  Poor flow thru  Poor flow thru  Insufficient Solder/Opens
(plated thru holes) (plated thru holes) (plated thru holes) Angle Too Shallow
 Solder Bridging
The Importance of Questions
Noise Parameters:
 What are they?
 Are they impossible or impractical to control?
 How robust is the system to the noise?

Controllable Parameters:
 How are they monitored?
 How often are they verified?
 Are optimum target values known?
 How much variation is there around the target values?
 How consistent are they?

Standard Operating Procedures:


 Do they exist?
 Are they understood?
 Are they being followed?
 Are they current?
 Is operator certification performed?
 Is there an audit schedule?
Process Parameter Questions
Process Parameters:
 What causes variation of the process parameter?
 How is the process parameter controlled?
 How often is the parameter out of control?
 Is there data on the parameter?
 Which of your process parameters should have
control charts on them?
 When should you place a control chart on a process
parameter?
 Which of your process parameters have control charts
on them?
 How are the control charts used?
 How do you know which process parameters to monitor?
 Should we focus on parameters of non value-added
steps?
Product Parameter Questions
Product Parameters:
 What is the goal of the improvement effort?
 Is the product parameter qualitative or quantitative?
– An attribute or a variable?
 For the product parameters:
– Is larger better?
– Is nominal best?
– Is smaller better?
– Is it dynamic in nature?
 Is the concern for...
– Process centering?
– Process variation?
– Both?
 What is the process baseline for the product parameter?
– What is the mean and sigma?
More Product Parameter Questions
Product Parameters:
 Is the product parameter currently in statistical control?
 Is the product parameter affected by time?
 How much of a change in the product parameter do you
need/wish to detect?
 Do you know the expected distribution of the product
parameter?
 Is the measurement system adequate?
 Are there multiple responses of concern? What are the
priorities for optimization?
 What measurements are taken on product parameters?
 How do you know which product parameters to monitor?
 Which product parameters need control charts on them?
 Which product parameters have control charts on them?
 How are the control charts used?
Process Map-Flow Charting (Step-by-Step)
Identify all value Identify/classify
Identify the inputs Document entire
and the scope of the
and outputs of flow of the
Non value-added process
the process process selected
operations

Identify/classify
measurements
taken on product
& process
parameters
Continue to
update &
classify Classify/ Develop initial list Identify/classify
process map! characterize of process parameters upstream
process parameters along with current in-process
into 3 main factors operating conditions product parameters

 Noise factors
 Standard operating procedures
 Controllable process parameters
Consider Every Process Step
1. What Process
feedback 5. What Customer
 All work is a process
Requirements
 All
processes have 2. What Output
owners  Specifications
 Allprocesses can be 3. What Input 4. Who are Customers – Function
described as a verb – Reliability
and noun – Format
6. Who are  Cost
 Allprocesses can be
analyzed and Suppliers  Schedule
improved
 Process owner’s
feedback 7. What
requirements on the
supplier Supplier
 Specifications
Requirements
 Cost
 Schedule
8. Process Controls/Dependencies
Procedures/Policies Training/Education Equipment/Facilities Quality Attitudes
Any written  Performance  Space required Personal attitude
document that skills  Processing which is less than
controls/impacts a  Certifications equipment the requirement
process
Causes of Process Map Failure

CONCERN RESPONSE
Requires extra effort - “We know the Initial payoff is team understands
process - lets just move forward” process -- team members are not
working on different set of
assumptions. Use experts to help
you through the mapping process

The process appears straight If necessary, adjust process


forward - then becomes difficult as boundaries -- initiate another
you realize you do not understand improvement team
process as well as you thought

The process map just seems to Think about approaching the


grow and grow and grow problem hierarchically
Process Mapping Summary
What is the tool? When do you apply this tool?
 Graphical method to illustrate the  Always: to fully understand
details of a process process & process flow
 Find where/when/how defects
What will the tool identify/show? are being created
 All process steps, value-added &  Define elements of cycle time
non value-added
What results can you expect?
 Input parameters (Xi,in)
 Systems needing MSE’s
 End product parameters (Yi)
 List of Factors for DOE’s
 In-process parameters (x’s & y’s)
 Find the hidden factory
 Characterization of all
 Opportunities for process step
parameters
elimination (i.e. flow
 Defect/data collection points
improvement)
 Steps needing FMEA’s
 Ways to re-layout the process
 Sources of variation identified
 Sources of variation reduced
Map Your Process
For the next session:
• Map and characterize a critical part of your process.
• Identify:
– The Inputs (X’s)
– The Outputs (Y’s)
– The Process Parameters (x’s)
– The Product Parameters (y’s)
– The process owner, supplier, & customer
– Classify the parameters at each step
– The next step to reduce defects in your process
What We Have Learned …
1. The importance of process maps and the character of
the product and process parameters.
2. The when, why and where to use process maps.

3. The x’s & y’s and X’s & Y’s and Y=f(x,X).

4. What to measure and control.

5. The need for process maps prior to FMEA’s, Gage


Studies, DOE’s and SPC.

6. When the process map is completed.

7. How to use the tool for your process.


Extending Flow Charts to Process Mapping
Failure Modes and
Effects Analysis
FMEA
About This Module…
Failure Modes and Effects Analysis

An FMEA is a systematic method for identifying,


analyzing, prioritizing and documenting potential
failure modes, their effects on system, product,
process performance and the possible causes of
failure.

Six Sigma, A Quest for Process Perfection


\DataFile\FMEAform.xls
Meet Goals and Attack Variation
\DataFile\CopyFMEA.xls
|Datafile|causeeffecte.igx
\Datafile\catapultflow.igx
\Datafile\catapultC&E.igx
What We Will Learn…
Failure Modes and Effects Analysis

1. As a Team, how to construct an


FMEA and associated Action Plan

2. How the FMEA process ties to


process mapping

3. The relationship between Failure


Mode, Cause and Effect

4. The different types of FMEAs


Sample FMEA
p R
S O D R Actual r i Risk
Failure E C E P Resp. Schedule Compl. p p p p s X
Effects V Causes C Controls T N Action Recommended Person Date Action Taken Date S O D n k prpn
Must PM Schedule
redo Periodic created and
copy 6 Paper Jam 7 Maint. 7 294
Periodic preventive maintence Key Opr 3/1 implemented 2/15 6 3 7 126 3 378
Place sign over copier outlining
Must Existing standard size enlarge/reduce or
redo User notes on reliable mach to clearly indicate Place sign over
copy 6 misset size 6 copier 5 180 standard reduce/enlarge Key Opr 2/20 mach 2/15 6 3 2 36 1 36
Must User Existing
redo misset notes on Place sign to encourage user to Place sign over
copy 6 control 5 copier 4 120 utilize auto settings Key Opr 2/20 mach 2/15 6 2 2 24 1 24
Used
landscape
Must instead of
redo portrait or Tray Place note on ruler re tray
copy 6 vice versa 7 Selection 2 84 selection Key Opr 1/20 Placed Note 1/15 6 2 1 12 1 12

Must align Use Auto


redo marking Feeder / Enlarge marks for 8.5 " paper on
copy 6 not clear 4 align ruler 3 72 ruler Key Opr 1/15 Enlarged marks 1/14 6 2 1 12 1 12
Doc
Must moved Use Auto Place sign over copier re
redo when lid Feeder / "Ensure align prior to copying or
copy 6 closed 5 align ruler 2 60 use auto Feeder" Key Opr 1/15 Displayed Sign 1/14 6 1 1 6 1 6

Must User Auto Place sign over copier to


redo selected select encourage user to use auto tray Place sign over
copy 6 wrong tray 3function 3 54 select Key Opr 2/25 mach 2/20 6 2 3 36 1 36
Must Periodic
redo Cleaning Place cleaning material near Placed Cleaning
copy 6 Dirty Glass 6 SOP 1 36 copier Maint. 1/15 Matl 1/15 6 1 1 6 1 6

Datafile/CopyFMEA.xls
Why Use FMEAs?
What is an FMEA?

Identify
critical product characteristics and process variables

Prioritize
product and process deficiencies in support of
downstream improvement actions

Help focus
on prevention of product and process problems
Benefits of FMEA’s
What is an FMEA?

 Improves the quality, reliability and safety of products.


 Helps increase customer satisfaction.
 Reduces product development timing and cost.
 Reduces the amount of rework, repair and scrap.
 Documents and tracks actions taken.
 Prioritizes deficiencies to focus improvement efforts.
Process and FMEA Hierarchies
What is an FMEA?
Process - Level #1

Step #1 Step #2 Step #3 FMEA - Level #1

Process - Level #2

Step #1 Step #2 Step #3 FMEA - Level #2

Process - Level #3

Step #1 Step #2 Step #3 FMEA - Level #3


Process FMEA Steps
What is an FMEA?
 Steps Completed Prior to FMEA:
– Charter Team
– Develop and Characterize Process Map
 FMEA Steps:
1. Identify “Heavy Hitter” Process Step
1. Identify Associated y’s (Product Parameters)
1. Identify Failure Mode
1. Identify Failure Effects/Rate Severity
1. Identify Causes/Rate Occurrence
1. Identify Controls (if any)/Rate Detection
1. Calculate RPN
1. Prioritize by RPN Order
1. Determine Actions/Plan
1. Recalculate RPN Based on Plan
1. Take Action
FMEA Form Header Accessible from
View Header/Footer
in Excel
What is an FMEA?

Workbook in Excel

\DataFile\FMEAForm.xls
Cause -Failure Mode -Effect Continuum
What is an FMEA?

Effect (y’s)
Cause Failure
(x’s) Mode
The Cause and Effect Diagram Example
What is an FMEA?
Admin/Service Example First produced in 1950 by
Measurements Materials Manpower Professor Kaoru Ishikawa -
Also called the:
 Ishikawa Diagram
 Fish Bone Diagram

Developed to represent the


Failure Mode relationship between some
(Defect) “effect and all possible
“causes” influencing it.

Failure Create using Igrafx:


Effect
Mother Methods Machines
Nature
The Cause and Effect Diagram Example
Cause and Effects Diagram

Measurement
Manpower Material
Reproducibility Repeatability
Inadequate Late
training Linearity
Wrong quantity
Stability
Lack of
experience Defective Calibration

Distractions

Defects
Too hot
Vague
Too humid
Not maintained Out of date

Too cold Inadequate capability Complex


Mother nature
Machines Methods

Datafile/Causeeffecte.igx
Copy Machine Example
What is an FMEA?
• Our process is copying
documents on a Xerox model
XC1045 copy machine.
• First we will construct a
process map
• Then we will construct a cause
and effect diagram
• Finally we will complete an
FMEA
Process: Making A Copy
What is an FMEA?

Place Set Set Select


Enter size Press Retrieve
Document number of light/dark paper
required button copies
in Copier copies settings source

N Hinges C Copies C Size desired C Darkness C Size desired Cr Button


N Glass clean required Cr Size button desired x Paper
Cr Number
Make Copies

button

Document set Number of Size selected Darkness set Correct Copies Copies
correctly copies selected correctly directly paper tray Right number
Glass clean correctly selected Right contrast
Right
orientation
Right size
Right paper

Legend
C Controllable
Cr Critical
N Noise
P Procedure
x Input
Step 1: Identify “Heavy Hitter” Process Step
The FMEA Process
 From the Process Map, identify the process step with
the most likelihood of having failure modes with
significant effects
 Use defect data and/or team knowledge about failure
modes when selecting process steps
 Significant impact to the business? (COPQ, cycle time,
fill rate, ...)
 Use a Cause and Effect Diagram to capture
brainstorming results.
 After completing FMEA Steps #2-7 for all failure modes
associated with this process step, return to this step and
select the next most likely “Heavy Hitter” process step
 Not all process steps will need to be analyzed by the
FMEA
Step 2: Identify Associated y’s
The FMEA
Process

 From the Process Map, identify the y’s that are


associated with the process step being investigated

 As the y’s are the indications of a successful completion


of the process step, they are crucial as a basis for
determining failure modes
Step 3: Identify Failure Mode
The FMEA
Process
• Brainstorm failure modes for the selected process step :
– Identify the ways in which the process could fail to
generate
each of the expected “y’s”
• Eliminate “duplicates” from brainstorm list
• Are the failure modes from the same level of the process?
• Are the failure modes specific?
• Are the failure modes the most likely?
• Do the failure modes provide good coverage of the
process step?
• Have all y’s been considered?
Step 4: Identify Failure Effect/Rate Severity
The FMEA Process

 Pick the most likely failure mode and brainstorm the


most important Effects:

– FAILURE EFFECTS are the outcome of the


occurrence of the failure mode on the process. The
impact on the customer --- What does the customer
experience as a result of the Failure Mode?

 Identify each effect as being “Attribute” or “Variable”

 Severity doesn’t change unless the design changes.


Step 5: Identify Causes/Rate Occurrence
The FMEA
Process
 Identify the most likely causes for each failure mode
using a Cause and Effect Diagram:

– CAUSES are the conditions that bring about the


Failure Mode

 Transfer the resulting information to the FMEA form

 Assign an occurrence value (1-10) to the likelihood that


each particular cause will happen and result in the
failure mode

 The occurrence score for each cause should be related


to the likelihood of that cause resulting in the failure
mode and producing the specific associated effect
Organize Brainstorming Ideas
The FMEA Process

Materials Manpower
Measurement
Wrong Size Selected wrong
Wrong Paper orientation
Selected Size

Copy Misaligned
Too Humid Document Moved
When Lid was Alignment Marking
Closed Unclear

Mother Method
Nature Machine

What would you add?


Step 6: Identify Controls/Rate Detection
The FMEA Process

 Identify the current mechanisms in place which prevent


the cause from occurring, or detect it before the product
reaches the customer. Some examples of controls are
SPC, training, maintenance, inspection, SOP etc.

 Assign a detection value (1-10) based on an


assessment of the likelihood that the current control
mechanisms will detect the cause of the failure mode
before it reaches the customer.

 Don’t agonize over detectability.


Step 7: Calculate Risk Priority Number
The FMEA
Process
The product of the estimates of severity

occurrence and

detection.

The RPN provides a relative priority for taking action


the bigger the RPN, the more important to address.

RPN = SEVERITY x OCCURRENCE x DETECTION


Steps 8 and 9
The FMEA Process

8: Prioritize by RPN Order


Use the “Sort” command in Excel to order the
spreadsheet in descending order of Risk Priority
Number (RPN).

9: Determine Actions/Plan
Based on the causes found, determine actions that
will minimize the effect of each cause, in priority
order.
Steps 10 and 11
The FMEA Process
10: Recalculate RPN Based on Plan
 Assuming the actions are carried out successfully, reassign
severity, occurrence and detectability.

 Place these new ratings in the “predicted” columns (ps, po & pd).

 Assign a rating from 1 to 5 for each action that will show the “risk”
associated with each action (5 being the greatest risk). Place the
rating in the “risk” column.

11: Take Action


 Based on the risk mitigation column (Risk * prpn), take the actions
indicated or reassign actions. Then….

 Complete the actions indicated by the times stated!


The FMEA Process
Steps 1-11: Step 2
Step 1
ID y’s \DataFile\CopyFMEA.xls
Step 6 Step 9
ID Process Steps
Step 4
Step 11
p R
Actual i Risk
S
Failure E
O
C
D R
E P FMEA Resp. Schedule Compl. p p p
r
p s X
Effects V Causes C Controls T N Action Recommended Person Date Action Taken Date S O D n k prpn
Must PM Schedule
redo Periodic created and
copy 6 Paper Jam 7 Maint. 7 294Periodic preventive maintence Key Opr 3/1 implemented 2/15 6 3 7 126 3 378
Place sign over copier outlining
Must Existing standard size enlarge/reduce or
redo User notes on reliable mach to clearly indicate Place sign over
copy 6 misset size 6 copier 5 180 standard reduce/enlarge Key Opr 2/20 mach 2/15 6 3 2 36 1 36
Must User Existing
redo misset notes on Place sign to encourage user to Place sign over
copy 6 control 5 copier 4 120 utilize auto settings Key Opr 2/20 mach 2/15 6 2 2 24 1 24
Used
landscape
Must instead of
redo portrait or Tray Place note on ruler re tray
copy 6 vice versa 7 Selection 2 84 selection Key Opr 1/20 Placed Note 1/15 6 2 1 12 1 12

Must align Use Auto


redo marking Feeder / Enlarge marks for 8.5 " paper on
copy 6 not clear 4 align ruler 3 72 ruler Key Opr 1/15 Enlarged marks 1/14 6 2 1 12 1 12
Doc
Must moved Use Auto Place sign over copier re Step 10
Step
redo 3 when lid Feeder /
Step 5 "Ensure align prior to copying or
ID Failure
copy 6 Modes 5
closed align ruler 2 60 Step
use auto 7 &8
Feeder" Key Opr 1/15 Displayed Sign 1/14 6 1 1 6 1 6
Process FMEA
Types of FMEA

• Helps analyze manufacturing and assembly


processes to reduce the occurrence and
improve detection of defects.
• Assists in the development of process control
plans.
• Establishes a priority for improvement activities.
• Documents the rationale behind process
changes and helps guide future process
improvement plans.
• IS PROACTIVE! Should be started when new
processes are designed or when old processes
are changed.
Process FMEA Scoring Definition
Types of
FMEA
Score SEVERITY CRITERIA OCCURRENCE DETECTION
10 Hazardous Without Warning 1 in 2 Very High Absolute Uncertainty
9 Hazardous With Warning 1 in 3 Very High Very Remote
8 Very High 1 in 8 High Remote
7 High 1 in 20 High Very Low
6 Moderate 1 in 80 Moderate Low
5 Low 1
1 in
in 400 Moderate Moderate
4 Very Low 2,000
1 in Moderate Moderately High
3 Minor 15,000
1 in Low High
2 Very Minor 150,000
£1 in Low Very High
1 None 1,500,00 Remote Almost Certain

Note: When completing a Process FMEA, first assume the material is good and
the process is bad. Then assume that the process is good and the material is
bad. Lastly, review the process for safety considerations.
Design/Product FMEA
Types of FMEA

• Helps to identify potential product failure


modes early in the product development cycle.
• Increases the likelihood that all potential failure
modes and their effects on assemblies will be
considered.
• Assists in evaluating product design
requirements and test methods.
• Establishes a priority for design improvement.
• Documents the rationale behind design
changes and helps guide future development
projects.
• IS PROACTIVE! Should be done when new
products are designed or existing products are
changed.
Defect FMEA
Types of FMEA

• Helps identify the root causes of defects.


• Establishes a priority for improvement activities.
• Documents plan of action.
• Provides methodology to battle initial ground swell
of defects.
• Focuses effort on defects with highest $ impact.
• IS NOT PROACTIVE!
Scoring Criteria
Use actual
Types of defect quantities
FMEA
Score SEVERITY CRITERIA OCCURRENCE DETECTION
10 Hazardous Without Warning Very High Absolute Uncertainty
9 Hazardous With Warning Very High Very Remote
8 Very High High Remote
7 High High Very Low
6 Moderate Moderate Low

5 Low Moderate Moderate

4 Very Low Moderate Moderately High


3 Minor Low High
2 Very Minor Low Very High
1 None Remote Almost Certain

Note: To change header information, click on "View" then "Header".


RISK: Optional field used to reflect the probability of completing actions.
The Catapult
FMEA
Exercise
Analyze the Catapult process using the FMEA tool.
(Remember we want to get the “most bang for the buck”.)
• Break into the Catapult teams
• We have already constructed a process map
• First, we will construct a cause and effect
diagram
• Then we will complete at least two failure
modes for the most critical step(s) of our
process 25 minutes!
• Appoint a spokesman for your team to debrief
the class on your progress, questions, etc.
• Complete the FMEA (FMEAform.xls) for the
Catapult process before the third session (We
will use this information for our DOE
competition)
Catapult Process Map
FMEA Exercise
Assemble Select Catapult Set Catapult
Start Secure to table
Catapult Settings Pins

Pins (2) Clamp Plan or Positions


Arm Tape Measure Prediction Designated
Rubber Band Tape equation
Ball Computer

Arm Aligned Feasible Correct


moves with tape settings settings
smoothly ± 3 inches

Pull Arm to Measure Record


Shoot Stop
Proper Angle distance distance

Operator Tape Measure


Consistency Operator Recorder
Lateral Observers Computer
No Parallax positioned
movement
properly

Correct
angle Ball flys Accurate Correct
straight measurement distance
± 2 inches recorded

Datafile/Catapultflow.igx
Complete the Diagram Below
FMEA
Exercise
Men
Method Material
Rubber Band Release consistency
Calculation procedure Angle measurement
Ball

Distance

Air Conditioner Arm moves freely Repeatability

Reproducibility
Mother nature
Machine Measure

Datafile/CatapultC&E.IGX
When To Update an FMEA?
FMEA Summary
An FMEA should be updated whenever a change is being
considered to a product’s:
 design
 application
 environment
 material
 product’s manufacturing or assembly process
Summary of Product/Process FMEA’s
FMEA
Summary
 What is the tool?  When do you apply this tool?
– Spreadsheet – When evaluating product for
robustness (functionality,
 What will the tool produceability, reliability)
identify/show? – During early stages of defect reduction
efforts to identify causes
– All product/process failure
modes, related effects, – When identifying key process/product
causes, & methods of parameters and evaluating methods
controlling them for controlling them
– Risk Priority Number (RPN)
for action based on failure  What results can you expect?
severity, probability of – Learn to identify critical product/
occurrence and detection process parameters
capability – Achieve consensus on solutions and
– Actions/plans to reduce methods of implementation
elements of RPN – Detailed product/process
understanding
Keys to Success
FMEA
Summary
 Identify purpose...BE SPECIFIC!

 Understand effects...INVOLVE CUSTOMERS & SUPPLIERS!

 Link to the process map.

 Use to prioritize efforts, allocate resources.

 Use as a risk assessment/prioritization tool based on predicted impact.

 Use to build consensus on prioritization.

 Encourage creativity...TEAMWORK!

 PLAN!

 ASK QUESTIONS!
Process FMEA Steps
What is an FMEA?
 Steps Completed Prior to FMEA:
– Charter Team
– Develop and Characterize Process Map
 FMEA Steps:
1. Identify “Heavy Hitter” Process Step
1. Identify Associated y’s (Product Parameters)
1. Identify Failure Mode
1. Identify Failure Effects/Rate Severity
1. Identify Causes/Rate Occurrence
1. Identify Controls (if any)/Rate Detection
1. Calculate RPN
1. Prioritize by RPN Order
1. Determine Actions/Plan
1. Recalculate RPN Based on Plan
1. Take Action
FMEA Appendix
Key Definitions for FMEA
Severity is an assessment of how serious the effect of the potential failure mode is on the customer.
The customer in this case could be the next operation, subsequent operations, or the end user.

Occurrence is an assessment of the likelihood that a particular cause will happen and result in the
failure mode.
Detection is an assessment of the likelihood that the current controls (design and process) will detect
the cause of the failure mode, should it occur, thus preventing it from reaching your customer.
The customer in this case could be the next operation, subsequent operations, or the end user.

Current Controls (for both design and process) are the mechanisms which prevent the cause of the
failure mode from occurring, or detect the failure mode, should it occur, before the product
reaches your “customer.” For example, current controls include SPC, inspections, written
procedures, training, preventive maintenance and all other activities that ensure a smooth
running process.
Critical Characteristics are those items which affect customer safety and/or could result in non-
compliance to regulations and thus require controls to ensure 100% compliance. These are
usually process“settings” such as temperature, time, speed, etc.

Significant Characteristics are those items which require SPC and quality planning to ensure
acceptable levels of capability.
FMEA Appendix
Terminology
A. Process or Product Name – Description of Process or Product being analyzed.
B. Responsible – Name of Process Owner.
C. Prepared By - Name of Agent coordinating FMEA study.
D. FMEA Date – Dates of Initial and subsequent FMEA Revisions.
E. Process Step/Part Number – Description of individual item being analyzed.
F. Potential Failure Mode – Description of how the process could potentially fail to meet the
process requirements and/or design intent, i.e. a description of a non-conformance at that
G. Potential Failure Effects – Description of the effects of the Failure Mode upon the customer,
i.e. what the next user of the process or product would experience or notice.
H. SEV (Severity) – An assessment of the seriousness of the effect of the potential failure mode
I. Potential Causes – Description of how the failure could occur, described in terms of something
J. OCC (Occurrence) – Description of how frequently the specific failure cause is expected to
K. Current Controls – Description of process controls that either prevent, to the extent possible,
L. DET (Detection) – An assessment of the probability that the current controls will detect the
potential cause, or the subsequent failure mode.
M. RPN (Risk Priority Number) – The product of the Severity, Occurrence, and Detection
Rankings i.e., RPN = SEV * OCC * DET.
N. Actions Recommended – Actions to reduce any or all of the Occurrence, Severity or Detection
rankings.
O. Responsibility – Person or group responsible for the Recommended Action.
P. Actions Taken – Brief description of actual action and effective date.
Q. New SEVERITY Rating after corrective action.
R. New OCCURENCE Rating after corrective action.
S. New DETECTION Rating after corrective action.
T. Resulting new RPN after corrective action.
Introduction to Process Capability
About this Module
• Process capability enables the prediction of the
ability of any process to produce products and
services that meet their desired specifications.
• This module focuses on typical manufacturing
processes. Transactional and other
manufacturing processes are not discussed here.
• The principles of process capability will be
introduced and Minitab will be used to calculate
process capabilities.
Learning Objectives
At the conclusion of this module participants
will be able to:
1. Recognize the value of and uses for process
capability.
2. Calculate and explain the capability of
processes whose output is normally
distributed.
3. Predict the probability that the output of a
process will be within its specification limits.
We Live in a Statistical World
Basic Statistics
• Statistics have a pervasive influence on
our lives
– Every day there is another poll
– Sampling is being used to perform many
aspects of the census
– All major economic indicators are based on
samples
– TV ratings are based on samples
– Statistics determine insurance rates
• Quantum physics has demonstrated that
Types of Statistics
Definitions
Descriptive statistics
is the process of describing the information we have.
We summarize information from a sample or
population give a clear understanding, or description,
of the data.

Inferential statistics
is the process of using information from a smaller set
of data (sample) to reach conclusions or inferences
about a larger group (population). Usually, we have
only sample information, not the entire population, and
must infer understanding of the population based on
our sample. We want these conclusions to be
mathematically correct.
Data Types
Attribute
Definitions
Yes - no
Good - bad
Accept - reject

Discrete
Multiples of whole units
Can not be meaningfully divided
Count or classification

Continuous
Can be meaningfully divided into finer and finer
increments of precision weight, length, voltage,
time
Measures of Central Tendency - Location
Definitions

Mean - the sum of all members divided by the

m = Population mean = 1 X1 , X2 , X3 ... X j


population size (average) j

X = Sample mean N

Mode - the most frequently occurring or most likely value


Median - the fiftieth percentile
(half the values are above and half below the
median)
Population Versus Sample
Definitions

Population Samples
Size N n
Location Average (Mean) m x

Dispersion: Variation
Variance s2 s2
Std dev s s
Range R = XHi-XLo

Statistics infer information about the parameters of the


population.
Quantifying Dispersion - Spread
Definitions

x1 xn
X
x2

We could add the differences between each value


x and the average of the values x however that
would always yield zero. Therefore we square the
difference between each x and x, to eliminate the
negatives and emphasize the s 2 outliers, then take
the average of the results. This is defined as the
variance or s2. Obviously, s=
Attributes of the Histogram - Location &
Spread
Definitions n
Number of Cases = 50
Mean & Median = 75 s=s
ˆ=  i
(
i=1
X  X )2

Standard Deviation = 8.3299


Range = 40
Variance = 69.388
n1
14
Minimum = 55 n

ˆ= X
12 Maximum = 95

X=m
F
r 10 i
e i=1
q 8
u n
e 6
n Variable X measurements:
c 4
y 75 80 75 65 70
2 85 70 70 85 70
60 80 80 80 65
0
50 60 70 80 90 100 110
80 75 75 70 85
70 75 75 75 85
Values of X 80 55 70 70 85
65 70 80 75 65
75 85 90 80 65
70 75 75 80 80
75 95 90 60 65
Measures of Variability - Variance =
Sigma Squared
Definitions
 Sigma Squared is a measure of dispersion of the
population about the mean
 Variances are not in the units of interest; standard
deviations are in the units of interest
 Variances are additive; standard deviations are not
additive...
…so s12 + s22 + s32 is OK,

but, s1 + s2 + s3 is NOT OK
Measures of Variability – S. D. = Sigma
Definitions
Standard Deviation is a measure of dispersion of the
population about the mean
s = the units of interest and is population standard
deviation
m = population mean
N = total population

(
s = estimate of standard
X1  m
2
) (
deviation2
+ X2  m + X3 m) (2
... X N  m ) ( ) 2

s (X )=
n = sample size
N

(X X ) + (X
2
X ) + (X
2
X ) ...(X
2
X ) 2

ŝ = s = 1 2

n-1
3 n
Normal Distribution
Each curve shown here has: Therefore, the same % of
 An area of one the population is under
 A mean of zero each of the curves for n s
 A standard deviation of s about the mean.
Quantifying the Normal Distribution
Definitions

m3sm2sm1smm+1sm+2sm+3s
68.26%
95.46%
99.73%
Probability Density Function –
Standard Normal Distribution
Definitions

Area under the curve = 1 The formula for the


s=1 probability density
m=0 function is  z2
1
f ( z) = e 2
Any normal distribution
can be converted to a
2
standard normal
distribution
The Standard Normal Transform
Definitions

X m
Z=
s
Permits conversion of any data point (X) into a Z
value. This value allows us to look up the percentage
of the population that is above and below the data
point.
Sample Questions………
Q. A new iron ore mine is discovered. 10 Kg ore is
collected from each of 20 spots.
a. This procedure is called as __________ .
b. Average is calculated from iron content of each of 20
spots. This value is called as ________________
c. If we calculate average, range and standard deviation
from the iron content values of each lot, this data is
called as ________________ .
d. Predictions are made regarding average iron content
of the mine, total iron that can be extracted, impurities
present etc. the analysis is called as ___________ .
e. The above calculations will give answers which will be
100% correct. True / False
Answers
a. The procedure is called as SAMPLING.

b. The value is called as STATISTIC.

c. The values are called as STATISTICS.

d. The analysis is called as STATISTICAL


ANALYSIS.

e. The Statement is FALSE.


Basic Principle
• All measures of process capability are based
on the concept of calculating the number of
standard deviations between the process
center and the specification limits.
• A Six Sigma process has six units of standard
deviation between the process center and
both specification limits.

LSL USL
Visualizing Process Capability

Process width

Specification
width
Quantifying Process Capability
Yield of a one sigma process 0.683 If we assume the
process is centered
on the target and
does not shift or drift
the yields would be.

Process Sigma Yield


1 0.6826895
2 0.9544997
3 0.9973002
4 0.9999367
5 0.9999994
6 1
LS US
L L
The Standard Deviation

m s= S (X – X)2
1 Sigma - 68%
2 Sigma - 95% n
3 Sigma - 99.73 %

1s
p(d)

Upper Specification Limit (USL)


Target Specification (T) T USL
Lower Specification Limit (LSL)
Mean of the distribution (m)
Standard Deviation of the distribution (s)
3s
Calculating Yield
We know that Z is the number of units of standard
deviation on a standard normal curve which has a mean
of zero and a standard deviation of one. We also know
any normal distribution can be converted to the standard
normal using the Z equation.
If the specification limits are substituted for x
x m we can determine the number of units of
Z =
s standard deviation between the process
center and the specification limits on the
X  X standard normal curve. Then we can use the
Z =
sˆ tables to look up the probability a value will be
less than that number.
In most cases we do not know m or s so we
substitute the sample statistics for the
population parameters as shown.
Calculating Yield
The mean time taken for completing an operation is 500 hrs.
and this is normally distributed with a standard deviation of
100 hrs.

1. What is the probability that an operator taken at random


will take between 500 to 650 hrs to complete the
operation?

2. What is the probability that he will take > 700 hrs?

Express in graphical form also.


Calculating Yield – Example

1. What is the probability that an operator taken at random

will take between 500 to 650 hrs to complete the

operation?
x m
Z = = (650 – 500) / 100 = 1.5
s

Looking up z table, the corresponding value under the

Standard Normal Distribution is 0.4332. i.e. 43%.


Calculating Yield – Example
2. What is the probability that he will take > 700 hrs?

x m
Z = = (700 – 500) / 100 = 2
s
Looking up z table, the corresponding value under the

Standard Normal Distribution is 0.4772.

Thus the probability of an operator taking more than 700 hrs

is (0.5 – 0.4772) = 0.0228, i.e. slightly over 2%.


Calculating Yield Example
Consider a process that has the following specification
limits: Lower Specification Limit (LSL) of -1 and a Upper
Specification Limit of 1. Data indicates the process is
centered on 0 with a standard deviation of 1. What is the
yield? Using tables or software to look up the
USL-X area under the curve when Z=1 we
ZUSL = find .8413. This means that 84.13% of
σ̂
1-0 the product has a value less than the
ZUSL = =1 upper specification limit.
1 Again using tables or software to look
LSL-X up the area under the curve when Z=-1
ZLSL =
σ̂ we find .1586. This means that
-1-0 15.86% of the product has a value less
ZLSL = = 1
1 than the lower specification limit.
What if the Process Shifts?
Generally speaking, processes have been
observed to shift and/or drift 1.5 standard
deviations over time. How would that effect the
yield of a one sigma process?
LSL USL

-4 -3 -2 -1 0 1
Calculating Shifted Process Yield
USL-X Using tables or software to look up the
ZUSL =
σ̂ area under the curve when Z=-.5 we
1-0-1.5 find .3085
ZUSL = = .5
1
LSL-X Again using tables or software to look
ZLSL =
σ̂ up the area under the curve when Z=-
-1-0-1.5 2.5 we find .0062.
ZLSL = = 2.5
1
Subtracting the two we obtain .3023
Using These Principles

• Process capability and process capability

indices are unambiguous and will be

addressed first.

• Process sigma is somewhat ambiguous

and will be addressed second.


Process Capability Terms
Measures of the ability of a process to produce compliant
products/services:
Cp - Short-term process capability
 For a limited period of time (not including shifts and drifts)
 Does not consider process centering
 Also known as process entitlement
Cpk - Short-term process capability index
 For a limited period of time (not including shifts and drifts)
 Does consider process centering
Pp - Long-term process capability
 For an extended period of time (including shifts and drifts)
 Does not consider process centering
Ppk - Long-term process capability index
 For an extended period of time (including shifts and drifts)
 Does consider process centering See formulae on next page.
Capability Formulae

Specification Width (s) USL-LSL


Cp = =
Short-Term Process Width 6σ ST
Specification Width (s) USL-LSL
Pp = =
Long-Term Process Width 6σ LT
USL-X X-LSL
Cpk= Lesser of: or
3σ ST 3σ ST

Ppk= Lesser of: USL-X or X-LSL


3σ LT 3σ LT
Using Minitab
The data is continuous so test for normality
Stat>Basic Statistics>Normality Test
The Normality Test
Probability Plot of Caps
Normal
99.99
Mean 1.000
StDev 0.001986
N 750
99 AD 0.619
95 P-Value 0.107

80
Percent

50 The P value is >


20 .05 therefore do
5 not reject the
1 assumption of
normality.
0.01
0.992 0.994 0.996 0.998 1.000 1.002 1.004 1.006 1.008
Caps
Worksheet: Bottle Caps.MTW
Using Minitab to Calculate Process Capability
Minitab Results
Process Capability of Caps

LSL Target USL

LSL
Process Data
0.995
Within
Target 1 Overall
USL 1.005
Sample Mean 1.00006 Potential (Within) Capability
Sample N 750 Z.Bench 2.26
StDev(Within) 0.00198431 Z.LSL 2.55
StDev(Overall) 0.00198636 Z.USL 2.49
Cpk 0.83
Overall Capability
Z.Bench 2.26
Z.LSL 2.55
Z.USL 2.48
Ppk 0.83
Cpm 0.84

945 960 975 990 005 020 035 050


9 9 9 9 0 0 0 0
0. 0. 0. 0. 1. 1. 1. 1.
Observed Performance Exp. Within Performance Exp. Overall Performance
% < LSL
% > USL
0.40
0.67
% < LSL
% > USL
0.54
0.64
% < LSL
% > USL
0.54
0.65
Let’s examine this
% Total 1.07 % Total 1.18 % Total 1.19
in detail
Worksheet: Bottle Caps.MTW
Process Data
Process Capability of Caps

LSL Target USL

LSL
Process Data
0.995
Calculated Within
Target 1 Overall
USL 1.005 directly
Sample Mean 1.00006 Potential (Within) Capability
Sample N 750 from the Z.Bench 2.26
StDev(Within) 0.00198431 Z.LSL 2.55
StDev(Overall) 0.00198636 data. The Z.USL 2.49
Cpk 0.83
within Overall Capability
Z.Bench 2.26
standard Z.LSL 2.55
Z.USL 2.48
deviation is Ppk 0.83
Cpm 0.84
the pooled
45 960 975 990 005 020 035 050
99 9 9 9 0 0 0 0
0. 0. 0. standard
0. 1. 1. 1. 1.
Observed Performance
% < LSL 0.40
Exp. Within Performance
% < LSL 0.54
deviation% < LSL
of
Exp. Overall Performance
0.54
% > USL 0.67 % > USL 0.64
% Total 1.07 % Total 1.18
the %% >Total USL 0.65
1.19

subgroups.
Worksheet: Bottle Caps.MTW
Observed Performance
Process Capability of Caps

LSL TheTarget
percent USL
Process Data
Within
LSL 0.995 of product
Target 1 Overall
USL
Sample Mean
1.005
1.00006
that was Potential (Within) Capability
Sample N
StDev(Within)
750
0.00198431
outside of Z.Bench
Z.LSL
2.26
2.55
Z.USL 2.49
StDev(Overall) 0.00198636
the upper Cpk 0.83

and lower Overall Capability


Z.Bench 2.26
specification Z.LSL
Z.USL
2.55
2.48

limits in this Ppk


Cpm
0.83
0.84
5 0 5
7data90 set.
05 020 035 050
994 996 99 99 00 0 0 0
0. 0. 0. 0. 1. 1. 1. 1.
Observed Performance Exp. Within Performance Exp. Overall Performance
% < LSL 0.40 % < LSL 0.54 % < LSL 0.54
% > USL 0.67 % > USL 0.64 % > USL 0.65
% Total 1.07 % Total 1.18 % Total 1.19

Worksheet: Bottle Caps.MTW


Expected Within Performance
Process Capability of Caps
The percent of
LSL Target USL
product that is
Process Data
Within
LSL
Target
0.995
1
expected to be Overall
USL
Sample Mean
1.005
1.00006
outside of the Potential (Within) Capability
Sample N 750 upper and lower Z.Bench
Z.LSL
2.26
2.55
StDev(Within) 0.00198431
StDev(Overall) 0.00198636 specification Z.USL
Cpk
2.49
0.83
limits on an Overall Capability

short term basis. Z.Bench Z.LSL


2.26
2.55
Z.USL 2.48
This projection is Ppk 0.83

5 0 5 0
based
5 0
on 5
the0
Cpm 0.84

994 996 997 999 within 000 002 standard003 005


0. 0. 0. 0. 1. 1. 1. 1.
Observed Performance Exp. Within Performance
deviation and the
Exp. Overall Performance
% < LSL
% > USL
0.40
0.67
% < LSL
% > USL
0.54
0.64
process
% < LSL
% > USL
0.54
0.65
mean.
% Total 1.07 % Total 1.18 % Total 1.19

Worksheet: Bottle Caps.MTW


Expected Overall Performance
The percent of Process Capability of Caps
product that is
expected to be LSL Target USL

LSL
Process Data
outside 0.995 of the Within
Target 1 Overall
USL upper 1.005 and
Sample Mean 1.00006 Potential (Within) Capability
Sample Nlower 750 Z.Bench 2.26
StDev(Within) 0.00198431 Z.LSL 2.55
specification
StDev(Overall) 0.00198636 Z.USL 2.49
Cpk 0.83
limits on an Overall Capability
long term Z.Bench 2.26
Z.LSL 2.55
basis. This Z.USL 2.48
Ppk 0.83
projection is Cpm 0.84
5 0 5 0 5 0 5 0
based on the 994 996 997 999 000 002 003 005
0. 0. 0. 0. 1. 1. 1. 1.
overall
Observed Performance Exp. Within Performance Exp. Overall Performance
standard
% < LSL 0.40 % < LSL 0.54 % < LSL 0.54
% > USL 0.67 % > USL 0.64 % > USL 0.65
deviation
% Total 1.07 and
% Total 1.18 % Total 1.19

the process
Worksheet: Bottle Caps.MTW
mean.
Potential Within Capability
Process Capability of Caps
Z USL and Z LSL are
LSL Target USL
calculated using the
Process Data
Within
LSL
process
Target
0.995
1
mean, the Overall
USL 1.005
respective
Sample Mean 1.00006 Potential (Within) Capability
Sample N 750 Z.Bench 2.26
specification
StDev(Within) 0.00198431 limits and Z.LSL 2.55
StDev(Overall) 0.00198636 Z.USL 2.49
the within standard Cpk 0.83
Overall Capability
deviation. Z.Bench 2.26
Z.LSL 2.55
Z bench is calculated by Z.USL 2.48
Ppk 0.83
putting the projected Cpm 0.84

within yield .(.982) 45 60 975 990 005 020 035 050


99 .99 .9left 9 0 0 0 0
0 0 0 0. 1. 1. 1. 1.
hand tail of
Observed Performance
a standardExp. Overall Performance Note: Minitab uses
Exp. Within Performance
normal
% < LSL
% > USL
0.40
0.67
then%% <>looking
LSL
USL
0.54
0.64
up%% <> LSL
USL
0.54
0.65 within to describe short
the respective Z score.
% Total 1.07 % Total 1.18 % Total 1.19
term variation and
Cpk is calculated
Worksheet: Bottle Caps.MTW 1/3 of overall to describe long
the lesser of Z USL or Z term variation.
LSL.
Overall Capability
Process Capability of Caps
Z USL and Z LSL are
calculated using LSL the Target USL
Process Data
LSL process 0.995 mean, the Within
Target 1 Overall
USL respective 1.005
Sample Mean 1.00006 Potential (Within) Capability
specification
Sample N
StDev(Within)
750
0.00198431
limits and Z.Bench
Z.LSL
2.26
2.55
the overall
StDev(Overall) 0.00198636 standard Z.USL
Cpk
2.49
0.83

deviation. Overall Capability


Z.Bench 2.26
Z bench is calculated by Z.LSL
Z.USL
2.55
2.48
putting the projected Ppk
Cpm
0.83
0.84
within yield (.981) 945 960 9left 75 990 005 020 035 050
.9 0.9 0.9 0.9 1.0 1.0 1.0 1.0
hand tail of0 a standard
Observed Performance Exp. Within Performance Exp. Overall Performance
%normal
< LSL 0.40 then % < looking
LSL 0.54 up % < LSL 0.54
% > USL 0.67 % > USL 0.64 % > USL 0.65
the
% Total respective
1.07 % Total Z score.
1.18 % Total 1.19

Ppk is Bottle
Worksheet:
calculated
Caps.MTW
as 1/3
of the lesser of Z USL or
Z LSL.
Within - Between
Process Capability of Caps

LSL Target USL


Process Data
Within
This process is very
LSL
Target
0.995
1 Overall
USL 1.005
stable as indicated by
Sample Mean 1.00006 Potential (Within) Capability
Sample N 750 Z.Bench 2.26
the Within and
StDev(Within) 0.00198431 Z.LSL 2.55
StDev(Overall) 0.00198636 Z.USL 2.49
Between lines being Cpk 0.83
Overall Capability
very close together. Z.Bench 2.26
Z.LSL 2.55
Z.USL 2.48
Ppk 0.83
Cpm 0.84

945 960 975 990 005 020 035 050


9 9 9 9 0 0 0 0
0. 0. 0. 0. 1. 1. 1. 1.
Observed Performance Exp. Within Performance Exp. Overall Performance
% < LSL 0.40 % < LSL 0.54 % < LSL 0.54
% > USL 0.67 % > USL 0.64 % > USL 0.65
% Total 1.07 % Total 1.18 % Total 1.19

Worksheet: Bottle Caps.MTW


Calculating Cpk and Ppk
Calculating Cpk and Ppk
CPL and CPU are Process Capability of Caps
calculated using the
process mean, the
LSL Target USL
Process Data
LSL respective
Within
0.995 specification
Target 1 Overall
USLlimits 1.005 and the within
Sample Mean 1.00006 Potential (Within) Capability
standard
Sample N 750 deviation. Cp 0.84
StDev(Within) 0.00198431 CPL 0.85
Cp is calculated
StDev(Overall) 0.00198636 by CPU 0.83
Cpk 0.83
putting the projected Overall Capability

within yield (.982) left Pp


PPL
0.84
0.85
hand tail of a standard PPU
Ppk
0.83
0.83
normal then looking up Cpm 0.84

9 45 960 975 990 005 020 035 050


the respective 0.
9 Z .9score.
0 0.
9
0.
9
1.
0
1.
0
1.
0
1.
0
Then dividing
Observed Performance
by 3. Cpk Exp. Overall Performance
Exp. Within Performance
PPMis the4000.00
< LSL lesser PPMof< LSLCPL or PPM
5355.08 < LSL 5395.59
PPM > USL 6666.67 PPM > USL 6432.04 PPM > USL 6478.48
PPMCPU.
Total 10666.67 PPM Total 11787.12 PPM Total 11874.07

Worksheet: Bottle Caps.MTW


Calculating Cpk and Ppk
Process Capability of Caps
PPL and PPU are
calculated usingLSLthe Target USL
Process Data
LSLprocess 0.995mean, the
Within
Target 1 Overall
USLrespective 1.005 specification
Sample Mean 1.00006 Potential (Within) Capability
limits
Sample N and
750 the overall Cp 0.84
StDev(Within) 0.00198431 CPL 0.85
standard
StDev(Overall) deviation.
0.00198636 CPU 0.83
Cpk 0.83
Pp is calculated by putting Overall Capability

the projected within yield Pp


PPL
0.84
0.85
(.9813) left hand tail of a PPU
Ppk
0.83
0.83
standard normal then Cpm 0.84
5 60 975 990 005 020 035 050
looking up the.994respective 99 9 9 0 0 0 0
0 0. 0. 0. 1. 1. 1. 1.
Z score.
Observed Performance Exp. Within Performance Exp. Overall Performance
PPMThen
< LSL dividing
4000.00 PPM <byLSL 3. Ppk is
5355.08 PPM < LSL 5395.59
PPM > USL 6666.67 PPM > USL 6432.04 PPM > USL 6478.48
PPMthe
Total lesser
10666.67 of PPMPPL
Total or PPU.PPM Total 11874.07
11787.12

Worksheet: Bottle Caps.MTW


A Process that Drifts
Consider a similar process that does change over time.
The data is in Caps Drift.MTW.
First test for normality
Normality Test
Probability Plot of Caps Drift
Normal
99.99
Mean 1.000
StDev 0.001291
N 720
99 AD 0.326
95 P-Value 0.520

80
Percent

50

20
Do not
5
reject the
1
assumption
0.01 of
0.9950 0.9975 1.0000 1.0025 1.0050
Caps Drift
normality.
Worksheet: Caps Drift.MTW
Capability Analysis
Capability Analysis
Process Capability of Caps Drift

LSL Target USL


This process is
LSL
Process Data
0.995
Within
changing
Target
over
1
time as Overall
USL 1.005
indicated
Sample Mean by the
0.999998 Potential (Within) Capability
Sample N 720 Z.Bench 4.76
difference
StDev(Within) between
0.00102066 the Z.LSL 4.90
StDev(Overall) 0.00129055 Z.USL 4.90
Within and Between Cpk 1.63
Overall Capability
lines. Z.Bench 3.70
Z.LSL 3.87
Plot this data using a Z.USL 3.88
Ppk 1.29
control chart. Cpm 1.29

0.9960 0.9975 0.9990 1.0005 1.0020 1.0035 1.0050

Observed Performance Exp. Within Performance Exp. Overall Performance


% < LSL 0.00 % < LSL 0.00 % < LSL 0.01
% > USL 0.00 % > USL 0.00 % > USL 0.01
% Total 0.00 % Total 0.00 % Total 0.01

Worksheet: Caps Drift.MTW


Process Capability Exercise 1

1 ± .06 Inches

Data indicate the process is centered with a standard


deviation of .02.
Calculate the yield.
Exercise 1 Visualizing the Answer

-.06 -.04 -.02 +.02 +.04 +.06

Process average

Design width
(.12 Inches)

There are 3 units of standard deviation between the


process average and the specification limits therefore this
is a 3 sigma process (short term).
Exercise 1 Calculating the Yield

USL-X Using tables or software to look up the


ZUSL =
σ̂ area under the curve when Z=3 we
find .9986
1.06-1
ZUSL = =3
.02
LSL-X Again using tables or software to look
ZLSL = up the area under the curve when Z=-3
σ̂
we find .00135.
.94-1
ZLSL = = 3
.02 Subtracting the two we find a yield of
.9973
Exercise 2
Assume a process owner has asked you to analyze the
data in Process Capability Exercise 1.MTW. Parts A and
B are made on different machines in lots (subgroups) of
5.
The customer has established specification limits of 10
± .1 and requires a Ppk of 1.33.

Prepare a brief presentation to describe your analysis


and recommendations? Remember to present data
practically, graphically and analytically.
Exercise 2 Normality Tests
Probability Plot of Part A
Normal
99.99
Mean 10.00
StDev 0.05049
N 750
99 AD 0.420
95 P-Value 0.324

80
Percent

50

20

5
1

0.01
Probability Plot of Part B
9.8 9.9 10.0 10.1 10.2 Normal
Part A
99.99
Worksheet: Process Capability Exercise 1.MTW Mean 10.00
StDev 0.05647
N 750
99 AD 0.248
95 P-Value 0.752

80
No reason to reject the Percent
50

assumption of 20

normality for either 1

part. 0.01
9.8 9.9 10.0 10.1 10.2
Part B
Worksheet: Process Capability Exercise 1.MTW
Exercise 2 Process Capabilities
Exercise 2 Part A Process Capability
Process Capability of Part A

LSL Target USL

LSL
Process Data
9.9
Within
Target 10 Overall
USL 10.1
Sample Mean 10.0009 Potential (Within) Capability
Sample N 750 Cp 0.67
StDev(Within) 0.0496334 CPL 0.68
StDev(Overall) 0.0504922 CPU 0.67
Cpk 0.67
Overall Capability
Pp 0.66
PPL 0.67
PPU 0.65
Ppk 0.65
Cpm 0.66

9.88 9.92 9.96 10.00 10.04 10.08 10.12


Observed Performance Exp. Within Performance Exp. Overall Performance
% < LSL 1.87 % < LSL 2.10 % < LSL 2.28
% > USL 2.80 % > USL 2.30 % > USL 2.49
% Total 4.67 % Total 4.40 % Total 4.77

Worksheet: Process Capability Exercise 1.MTW


Exercise 2 Prepare the Control Charts
Control Chart Shows Process Stability
Xbar-R Chart of Part A
UCL=10.0675
10.05
Sample Mean

_
_
10.00 X=10.0009

9.95
LCL=9.9343
1 16 31 46 61 76 91 106 121 136
Sample

UCL=0.2441

0.2
Sample Range

4 _
R=0.1154
0.1 44
4

0.0 LCL=0
1 16 31 46 61 76 91 106 121 136
Sample

Worksheet: Process Capability Exercise 1.MTW


Process Capability Part B
Process Capability of Part B

LSL Target USL

LSL
Process Data
9.9
Within
Target 10 Overall
USL 10.1
Sample Mean 9.99961 Potential (Within) Capability
Sample N 750 Z.Bench 1.71
StDev(Within) 0.0496334 Z.LSL 2.01
StDev(Overall) 0.0564727 Z.USL 2.02
Cpk 0.67
Overall Capability
Z.Bench 1.43
Z.LSL 1.76
Z.USL 1.78
Ppk 0.59
Cpm 0.59

9.84 9.90 9.96 10.02 10.08 10.14


Observed Performance Exp. Within Performance Exp. Overall Performance
% < LSL 4.13 % < LSL 2.24 % < LSL 3.89
% > USL 4.00 % > USL 2.16 % > USL 3.77
% Total 8.13 % Total 4.39 % Total 7.66

Worksheet: Process Capability Exercise 1.MTW


Control Chart Shows Time Based Variation
Xbar-R Chart of Part B
1 1 1 11 1 1
UCL=10.0662
5
10.05
Sample Mean

_
_
10.00 X=9.9996

9.95
5 5 LCL=9.9330
1 1
9.90 1 1 1
1 16 31 46 61 76 91 106 121 136
Sample

UCL=0.2441

0.2
Sample Range

4 _
R=0.1154
0.1 44
4

0.0 LCL=0
1 16 31 46 61 76 91 106 121 136
Sample

Worksheet: Process Capability Exercise 1.MTW


Confidence Intervals
About This Module…

Confidence Intervals (CI) permit us to state that we


are X% confident that the population parameter of
interest is at most a specified interval from the
sample statistic.

Six Sigma, A Quest for Process Perfection


Meet Goals and Attack Variation

\DataFile\PurchOrd.mtw
\DataFile\PwrSuply.mtw
\DataFile\Conf-Int.mtw
\DataFile\OEack.mtw
What We Will Learn...

1. Significance of confidence intervals

2. How to calculate confidence intervals for:


– Means
– Standard deviations or Variation
Confidence in the Midst of Uncertainty?
 Mean and Standard Deviation statistics are:
– estimates of the population Mu’s (m) and
Sigma’s (s)
– based on one sample
Ninety-five Percent Certain

 Variability exists from sample to sample

 By using statistically based confidence


intervals, uncertainty can be quantified
The chances are approximately 95 out of 100
 Usually,
that 95% confidence
the calculated intervals
confidence arecontains
interval
calculated
the population parameter, or…

With 95% certainty, the population parameter is


inside the confidence interval.
Population vs. Sample
Population is the entire area of interest.

Sample is a subset of the population.

What is the relationship between the population and


the sample?

How
Population representative is
this sample?
Sample
Confidence Interval Symbols and
Definitions
Measure Population Sample
Use
Parameter Statistic
Z  n 30÷

sis known
Mean m X
t  n <30÷

Variance s2 s2 c2
Standard
s s c2
Deviation
Process ^
Capability
Cp Cp c2
^ F or
Proportion p p Z (approx)

a Typically
Alpha Risk
.05
Confidence Intervals (CI)
CI take the general form :

C.I.=Statistic +/- K * (Standard Deviation)


Statistic= Mean, Variance, CP, etc.
K = Constant based on a statistical distribution

CI reflect the sample to sample variation of our point estimates

We will look at CI for: m , sX , and C P


What is the Student t-distribution?
The Student t distribution is a family of bell shaped (Normal
like) distributions that vary by degrees of freedom (sample
size) - the fewer degrees of freedom, the wider and flatter
the distribution.
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
5

5
.5

.5

.5

.5

4
-4

-3

-2

-1

0.

1.

2.

3.
-3

-2

-1

-0

DF2 DF10 DF30


t-Distributions, Normal Approximation, Risk
To give an idea of the values of t
compared to Z for 95% (a= 0.05),
look at the table below:
Sample t-value Z-value
5 2.78 1.96
10 2.26 1.96 a = risk
20 2.09 1.96 t
a/2=0.025 a/2= 0.025
30 2.05 1.96
100 1.98 1.96
1000 1.96 1.96

We can use Z to
estimate t if
n  30
and s is known
STATISTICS FOR
EXPERIMENTERS-- BHH
Hypothesis Testing
About This Module…

Hypothesis Testing helps:


 Determine if there is a statistically significant
difference between two relatively small samples
 Quantify the risks of making an incorrect decision

Six Sigma, A Quest for Process Perfection


Meet Goals and Attack Variation
What We Will Learn

1. How to test variable data


a. Use a t-test to compare two means
b. Alpha (a) and Beta (b) Risks
c. Use a paired t-test to compare paired
treatments
d. Use test for equal variances

2. How to test discrete data


a. Compare proportions
b. Compare discrete data
Real World Scenario

Design:
Determine if two alternate design
changes are significantly different.

Manufacturing:
Determine if two different types of
material wear differently.

Administrative/Transactional / Service:
Determine if the change to a process
affected the cycle time.
Purposes of Hypothesis Testing

 Determine if there is a real difference


between ? and ? .

 Use relatively small samples to answer


questions about the population.

 Quantify the associated risks.


Example
Old Design New Design
 Hard disk transfer speed 89.7 84.7
(megabytes per second) is 81.4 86.1
marginal. A new design is 84.8 91.9
proposed. 87.3 86.3
79.7 79.3
 An Engineering Change
85.1 86.2
Notice (ECN) is
incorporated 81.7 89.1
83.7 83.7
 Is the new design better?
84.5 88.5
To answer this, we need some
fundamentals of significance testing first!
Steps in Hypothesis Testing
1. Define the problem 10. Select the samples
2. Determine the objectives 11. Conduct the test and collect data
3. Establish the Hypothesis 12. Calculate the test statistic (Z, t, or
– Write the Null Hypothesis F) from the data
(H0)
13. Determine the probability that the
– Write the Alternative calculated test statistic has
Hypothesis (Ha) occurred by chance
4. Determine the appropriate
14. If that probability is less than
statistical test (assume
distribution Z, t, F) alpha, reject H0
5. State the alpha risk (usually 5 %) 15. If that probability is greater than
alpha, do not reject H0
6. State the beta risk (usually 10-20
%) 16. Translate the statistical
7. Establish the effect size (delta) conclusion into a practical
solution
8. Compute the sample size
9. Develop a sampling plan
Significance Tests
 Used to assess evidence
provided by sample data to Typical Examples
Ho : m a = m b
reject, or fail to reject a claim
about a population parameter.
 Null hypothesis (Ho) is the
statement we assess.
Ha : m a  m b
 Ho is usually stated as, “there Ho : s a = s b
is no difference”.
 Alternate hypothesis (Ha) is
Ha : s a  s b
usually stated as “there is a Ho : pa = pb
difference”.
 We fail to reject Ho unless Ha : pa  pb
there is convincing evidence to
reject it.
What does 5% Level of Significance Mean?

This means that we will reject the null hypothesis if

the difference between the sample statistic and the

hypothesized population parameter is so large that

it would occur, on an average, only 5 or fewer

times in every 100 samples when the

hypothesized population parameter is correct.


Type I and II Errors: Associated Risks
Type I errors are made Type II errors are made
when we reject the null when we fail to reject the null
hypothesis when in fact it is hypothesis when in fact it is
true. false.

Alpha (a) risk is the Beta (b) risk is the


probability of making a type probability of making a type
I error. II error.

This risk is This risk is


typically set at 5%. typically set at 10%

Risks are set before the test or experiment is conducted


The Risk Truth Table
ais the risk of finding a difference when there really isn’t one.
b is the risk of not finding a difference when there really is one.

Action
Ho not rejected Ho rejected
State of Nature
Ho should not Type I or
Correct
be rejected producer’s risk
Decision
(Ho is true) a = P(Type I)

Ho should Type II or Correct


be rejected consumer’s risk Decision
(Ho is false) b=P(Type II)
Another Look at Risks
Remember:
ais the risk of finding a difference when there really isn’t one.
b is the risk of not finding a difference when there really is one.
Ho Ha

b a

Now we can determine if the ECN improved performance


Normal Distribution and t Distribution

When When
Population SD Population SD
is Known is Not Known
Sample size n is Normal Normal
> 30 distribution, z distribution, z
table table

Sample size < Normal t distribution, t


30, and we can distribution, z table
assume table
population is
approx normal
P Value

P value is the smallest level of significance that

would lead to rejection of the null hypothesis Ho.

eg. supposing Ho were true, what is the

probability of getting a value of x-bar this far from

the population mean? This probability is called a

prob value or p-value.


Manual Test Null Hypothesis m1  m2
 s 12 s 22 
Difference Upper bound = X 1  X 2 +  t a ,df * + 
 n n 
 1 2 
Difference Upper bound = -1.99 + (1.7459 * 1.57 )
Difference Upper bound = .756
The upper bound for the difference in the means indicates
the difference between the means of these populations
could be as great as .756 (at the 95% confidence level).
Therefore, the evidence is not statistically significant to
conclude that the difference between the new design and
the old design is less than 0. Fail to reject Ho.
Manual Test Null Hypothesis m1  m2

X1  X 2
t(a ,df ) = = -1.26
2 2
s s
+
1 2
n1 n2
P = .112
P> .05 therefore fail to reject Ho.
Hypothesis Testing Decision Tree
Hypothesis Ho: P1=P2 Ha: P1 P2
Testing Minitab: Stat-Basic Stat-1or 2 proportions

Proportions Testing (2 factors only)


Ho: 2 factors are independent
Continuous Data
Attribute Data Ha: 2 factors are dependent
(One factor only) Minitab: Stat-tables-Chi square test

Non-normal Normality test Normal Contingency Table


Ho: Data is Normal
Ha: Data is NOT Normal
2 or more samples Minitab: Stat-Basic Stat-Normality Test
Ho:s1=s2=s3…
Use Anderson-Darling
Ha: At least 1 is different 1 Sample 2 or More Samples
Levene’s Testfor Equal Variances
Minitab: Stat-ANOVA-Test
For only 2 s, this is similar to an F-test: F=(S1)2/(S2)2 Chi-Square Bartlett’s Test/F-Test
If Fcalc>Fcrit, reject null Ho:s1=s2=s3… Ha: s1 At least 2 are different
(Use Chi-Squared for 1 sample) Ho: s1=sTarget Ha: s1 sTarget Minitab: Stat-ANOVA-Test for Equal Variances
Minitab: Stat-Basic Stat-Graphical Summary For only 2 ss, this is same as F-test:
If s target falls within C1: then fail to reject Ho Stat>BasicStat>2 Variances
F=(S1)2/(S2)2
Ho: M1 = Target Ha: M1 Target 1 Sample T Test If Fcalc>Fcrit, reject Ho
Minitab: Stat-Nonparametric-1 Sample - sign (or) Ho: m1=mTarget Ha: m1 mTarget
1 Sample
Stat-Nonparametric-1 Sample Wilcoxon Minitab: Stat-Basic Stat-1 Sample-T One Way ANOVA
(Also used for paired comparisons Ho: M1-M2=0) (Also used for paired comparisons:Ho: m1=m2=0) Ho: m1=m2=m3=…
M1=Median or sample 1 1 Sample Z Test Ha: m1 at least 2 are different
M target = Target Median Minitab: Stat-ANOVA-One Way
Ho: m1=mTarget Ha: m1  mTarget
(Caution Bartlett’s p<0.05;
Minitab: Stat-Basic Stat-1 Sample-Z
assumes=variances)
Ho: M1=M2=M3… (Also used for paired comparisons: Ho:m1=m2=0)
Ha: At least 2 are different Sample Size >=30 s is known
2 or More Samples 2 Samples
Minitab: Stat-Nonparametric-Mann-Whitney (or) Ho: m1-m2=0 Ha: m1-m2  0
2 Sample T Test
Stat-Nonparametric-Kruskal-Wallis (or) Minitab: Stat-Basic Stat-Paired T
Stat-Nonparametric-Freidmans Ho: m1=m2 Ha: m1 m2 Paired T Test
(Compares (Variance
Means =)
when observations are
M1=Median sample 1, etc. Minitab: Stat-Basic Stat-2 Sample-T paired or dependent in a pairwise manner)
(Compares Means using pooled Std Dev)
Check box to assume equal variances or
Check box to assume unequal variances
Regression and Correlation
About This Module…
Correlation Analysis is used to quantify:
the degree of linear association between
variables

Regression Analysis is used to quantify:


the functional relationship between variables

Six Sigma, A Quest for Process Perfection


Meet Goals and Attack Variation
\DataFile\Correlat.mtw
\DataFile\Yarn.mtw
\DataFile\Cases.mtw
\DataFile\Water.mtw
\DataFile\RMystery.mtw
\DataFile\R-Exampl.mtw
\DataFile\Callque.mtw
\DataFile\Pizza.mtw
\DataFile\Cases.xls
\DataFile\CEO-COMPENSATION.xls
\DataFile\Realestate.xls
\DataFile\Oilcons.mtw
What We Will Learn
1. Correlation
– How to measure the linear relationship between two
variables
– The correlation coefficient
– Implication of the correlation coefficient “r”
2. Regression
– Definition of the regression line and how it is developed
– How to calculate and analyze a regression equation
– How to analyze relationships between an independent
variable
– and one or more dependent variables using regression
– How to interpret r2
– Understanding and analyzing residuals
– How to use the regression ANOVA table
Real World Examples

ADMINISTRATIVE

A software company wants to know the relationship between calls in


queue and service time.

MANUFACTURING

A customer and supplier disagree on the quantity received by customer


versus several months’ quantity ordered for a given lead time.

DESIGN

A chemical engineer, designing a new process, wants to investigate the


relationship between key input variables and stack loss of ammonia.
Regression and Correlation
 Regression and Correlation analyses show us
how to determine both the nature and the
strength of a relationship between two variables.

 Inregression analysis we develop an estimating


equation relating the dependent and independent
variables. i.e. how much percent of variation can
be explained by the regression equation.

 In correlation analysis we determine the degree


to which the variables are related.
Regression Analysis
 Used to fit lines and curves to data

 The fitted lines

– Quantify the relationship between the process


variables (X’s) and process performance (Y)

– Help identify the vital few X’s

– Enable predictions to be made

– Identify the impact of controlling the process


variables (X’s)

 Produces an equation to match the line


Terms
Correlation
– A measure of linear association
– Used when both X and Y are continuous
r values range from:
– Perfect positive relationship = 1
– No relationship = 0
– Perfect negative relationship = -1
Regression
– Provides the basis for predicting the values of a variable
from the values of one or more other variables
– Used with a continuous Y and continuous Xs, or
continuous Y and categorical Xs
r2 - proportion of variation of Y explained by the prediction
equation
A Word of Caution
• It is important that we consider the relationships
found by regression to be relationships of
association but not necessarily of cause & effect.

• That is unless we have specific reasons for


believing that the values of the dependent
variable are caused by the values of the
independent variable(s), do not infer causality
from the relationships we find by regression.
Correlation Illustrated
20 0 2 4 6 8 10
0
15
-5
10 -10
-15
5
-20
0
-25
0 2 4 6 8 10 -30

Correlation = 1 Correlation = - 1

12
10
8
6
4
2
0
0 2 4 6 8 10

Correlation = 0
Correlation Example
\DataFile\Correlat.mtw
Two test stations are used to measure power supply
voltage.
Is there a correlation?
Minitab: Stat>basic stat>correlation
Correlation of Station 1 and Station 2 = 0.959, P-Value = 0.000

The two are highly correlated (.959)

Is this reasonable?

? Are you comfortable with .959?


What does it mean to you?
How does the data actually look?
How would you find out?
Plot the Data
Graph>ScatterPlots
The Data
Scatterplot of Station 1 vs Station 2
9.4

9.3

9.2

If all of the 9.1


data points
Station 1

9.0
were on the diagonal
line, would8.9we have
8.8
perfect correlation?
8.7

8.6

8.5
8.6 8.8 9.0 9.2 9.4 9.6
Station 2
Worksheet: Correlat.MTW

Let’s try regression


Regression

Fitted line plot is used when there is only one predictor.


Example 1 (cont.)
In what ways is
Fitted Line Plot
Station 1 = 1.020 + 0.8729 Station 2 this graph
9.5 S 0.0557288

9.4
R-Sq
R-Sq(adj)
92.0%
91.5%
different from
9.3 the preceding
9.2
one?
Station 1

9.1
9.0
8.9
8.8
8.7 What are the
8.6 implications?
8.6 8.8 9.0 9.2 9.4 9.6
Station 2
Worksheet: Correlat.MTW

What action
Slide 225 was actual line – this is a would you
fitted line. Can be used for prediction. take?
Best Fit Line
Fitted Line Plot
Cases = 22.47 + 0.7546 Test Piece
100 S 11.5131
R-Sq 49.4%

90
R-Sq(adj) 47.7% By minimizing the
80
residual sum of
squares, we get a
Cases

70

60
best fit line of the
50
form:
40
40 50 60 70 80 90 100 Yi = a + bXi
Test Piece
Worksheet: cases.MTW

a = coefficient of the constant term or intercept


b = coefficient of the predictor, X
Statistical Process Control
About This Module…
Control charts portray process performance and
separate causes of variation:
• Random
• Assignable

Control Chart Systems are:


• A proven technique for improving productivity
• Effective in defect prevention
• Prevent unnecessary process adjustments
• Provide diagnostic information
• Provide information about process capability

Six Sigma, A Quest for Process Perfection


Meet Goals and Attack Variation

\DataFile\Attribut mtw
\DataFile\Variable.mtw
What We Will Learn.

1. Control charts are a powerful tool to hold the gains.

2. How control charts discriminate between common


cause and assignable cause variation.

3. Why control charts must be designed to fit the data


type and the control purpose.
Process Variation

Process variation is the


result of:
• Common causes.
• Special (assignable)
causes.
Common Causes

• Result in normal process


variation.
• Are specific to each process.
• Can be reduced by changing
the process.
Special (Assignable) Causes

• Are attributed to something outside of the


process.
• Result in abnormal process variation.
• Do not result in process improvement
if eliminated.
Uses of Control Charts
1) Attain a state of statistical control:
• All subgroup averages and ranges within control limits - no
assignable causes of variation present

2) Monitor a process

3) Determine process capability

Juran’s Quality Control Handbook, 4th edition, page 24.7

What happens after an out-of-control situation occurs at the


core of a successful SPC program?
General Concepts
w = some characteristic of interest

XW = mean of each sample

Sw = standard deviation of w

Upper Control Limit UCL = X + 3Sw

Centerline = X
Lower Control Limit LCL = X  3Sw
Therefore 99.73% of points will be within the control limits
unless there is an assignable cause
Control Chart Selection Tree
Type of data

Discrete Variable

Count Count or Attribute


No Yes
Classification Subgroup >1?

X Bar and R
Fixed or Variable Fixed Fixed or Variable IMR Chart
Fixed or
variable variable
X Bar and S
opportunity? opportunity?

Supplement
with EWMA if
CTQ is
P Chart
C Chart U Chart NP Chart sensitive to
small process
shifts
X and R Control Chart Formulae
& Constants
X Control Limits =X ± A 2 R
R Upper Control Limit = D 4 R
R Lower Control Limit = D3 R
Sample A2 D3 D4 d2
Size
2 1.880 - 3.267 1.128
3 1.023 - 2.574 1.693
4 .729 - 2.282 2.059
5 .577 - 2.114 2.326
6 .483 - 2.004 2.534
7 .419 .076 1.924 2.704
8 .373 .136 1.864 2.847
9 .337 .184 1.816 2.970
10 .308 .223 1.777 3.078
Creating Control Charts for Variables
\DataFile\Variable.mtw
Creating an X-bar and R Chart
An X-Bar and R Chart
Xbar-R Chart of measure1, ..., measure5
1 1
1

5
UCL=41.294
41
Sample Mean

_
_
40 X=40.000

6
39 6
LCL=38.706
1
1 1
38
1
5 10 15 20 25 30 35 40 45
Sample

UCL=4.743
4.5
Sample Range

3.0
_
R=2.243
1.5

0.0 LCL=0

5 10 15 20 25 30 35 40 45
Sample

Worksheet: Variable.MTW

The numbers show violations of the assumption of


control. The nature of the violation is given in the
session window.
X-Bar and R Chart Session Window
Test Results for Xbar Chart of measure1, ..., measure5
TEST 1. One point more than 3.00 standard deviations from center line.
Test Failed at points: 7, 10, 13, 16, 17, 29, 46

TEST 5. 2 out of 3 points more than 2 standard deviations from center line (on
one side of CL).
Test Failed at points: 10, 17, 47

TEST 6. 4 out of 5 points more than 1 standard deviation from center line (on
one side of CL).
Test Failed at points: 7, 32, 34

* WARNING * If graph is updated with new data, the results above may no
* longer be correct.
StatGuide Interprets the Tests
Comparing the Suppliers
Xbar-R Chart of measure1, ..., measure5 Xbar-R Chart of measure1, ..., measure5
42 1 1
UCL=41.207 1
41 5 UCL=41.384

Sample Mean
5
Sample Mean

_ _
_ X=40.106
40 X=39.894 40
6
5 5 LCL=38.829
39 1
LCL=38.580 38 1
1 1
2 4 6 8 10 12 14 16 18 20 22 24 2 4 6 8 10 12 14 16 18 20 22 24
Sample Sample

UCL=4.816 UCL=4.683
Sample Range

Sample Range
4 4
_ _
2 R=2.278 2 R=2.214

0 LCL=0 0 LCL=0
2 4 6 8 10 12 14 16 18 20 22 24 2 4 6 8 10 12 14 16 18 20 22 24
Sample Sample

Worksheet: Variable.MTW(Supplier = 1) Worksheet: Variable.MTW(Supplier = 2)

Supplier 2’s process is less stable


than Supplier 1’s process.
Add Dates to Control Charts

Adding dates to the control


charts may help identify
potential sources of shifts.
Adding Dates Indicates When the
Shift Occurred
Xbar-R Chart of Stacked Data 1s
12 5 UCL=11.850
6 2
2
2 2
Sample Mean

11 2 2 _
_
X=10.568

10
6
6 6
LCL=9.286
9
7/31/2006 8/4/2006 8/10/2006 8/16/2006 8/22/2006 8/28/2006 9/1/2006 9/7/2006 9/13/2006 9/19/2006
Date

4.5 UCL=4.699
Sample Range

3.0
_
R=2.222
1.5

0.0 LCL=0
7/31/2006 8/4/2006 8/10/2006 8/16/2006 8/22/2006 8/28/2006 9/1/2006 9/7/2006 9/13/2006 9/19/2006
Date

Worksheet: Variable data.MTW


Detecting Gradual Process Drift
Xbar-R Chart of Stacked Data 1s
12
5 UCL=11.850
6 2
2
Sample Mean 2 2
11 2 2
_
_
X=10.568

10 6

6 6
LCL=9.286
9
4 8 12 16 20 24 28 32 36 40
Sample

4.5 UCL=4.699
Sample Range

3.0
_
R=2.222
1.5

0.0 LCL=0

4 8 12 16 20 24 28 32 36 40
Sample

Worksheet: Variable.MTW

The tests detect the process shift but it would have been
detected earlier if the control limits had been based on the
first 20 data points.
Detecting Gradual Process Drift
Minitab calculates control limits based on the entire data set be default.
Freezing control limits detects shifts earlier than using the entire data set
to calculate limits.
Xbar-R Chart of Stacked Data 1s
12
1 1

2 2 UCL=11.360
2 2 2
5 6 2 2
11 2
Sample Mean

6 2 2 2
6
_
_
10 X=10.112

9
LCL=8.864
4 8 12 16 20 24 28 32 36 40
Sample

4.5 UCL=4.575
Sample Range

3.0
_
R=2.164
1.5

0.0 LCL=0

4 8 12 16 20 24 28 32 36 40
Sample

Worksheet: Variable.MTW
Rational Subgrouping
• Usually consecutive units
• Must come from a single distinct population
• Within subgroup, variation should be white noise only
• Between subgroups should capture variation due to black noise

1) Subgroup needs to represent the distinct population


2) Establish minimum subgroup size to reflect the within variation
3) Establish sample frequency to capture the between variation
4) Collect data maintaining the sequential information

• Never knowingly subgroup unlike things together


• Minimize variation within each subgroup
• Maximize opportunity for variation between subgroups
• Average across noise not across signals
• Treat charts in accordance with the use of the data
• Establish standard sampling procedures
Donald J. Wheeler has six guiding principles for subgrouping in a rational manner.
When to Increase Sample Size
If the sample size is too small, assignable causes may
produce real effects that are relatively small and
unimportant. In this case, it may not be economical to
take action.

To minimize the number of these “nuisance” causes,


the sample size should be increased using the
following techniques:
• Identify characteristics
• Determine logical nature of subgroups
• Identify sampling sequence
• Measurement methods proven to be accurate
What We Have Learned.

1. Control charts are a powerful tool to hold the gains

2. How control charts discriminate between common


cause and assignable cause variation

3. Why control charts must be designed to fit the data


type and the control purpose
Lean Enterprise
A Formula for Organizational Success
Agenda

1. Introduction to Lean, Wastes


2. 5S WPO & Visual Management
3. Standard Work
4. Value Stream Mapping
5. Quick changeover
6. Poka-yoke
7. Continuous Improvement
8. Kaizen Blitz
9. Starting Lean
Introduction to Lean and 7 Wastes

• Identify and Eliminate


the 7 Wastes
Brief History of Lean
• Craftsman to mass production
• Mass production to lean production
• Lean production: Ford to Toyota
– Frederick Taylor
– Taiichi Ohno
– Shigeo Shingo
– James Womack

• From shop floor to office and support functions


then to service industry
Craft Manufacturing
Late 1800’s

 Car built on blocks as workers walked around


 Built by craftsmen with pride
 Components hand-crafted, hand-fitted
 Excellent quality
 Very expensive
 Few produced
Mass Manufacturing

• Assembly line - Henry Ford 1920s

 Low skilled labor, simplistic jobs, no pride in work


 Interchangeable parts
 Lower quality
 Affordably priced for the average family
 Billions produced – identical
Model ‘T’
Lean History – Japan
• Sakichi Toyoda at his textile mills (20’s–30’s)
• Toyota Motor Company’s Kiichiro Toyoda and
Taiichi Ohno made innovations (40’s) in
assembly lines that provided efficient,
customer-focused, streamlined processes
with flow, variety, and short lead time—
Toyota Production System (TPS); to face
competition of GM and Ford after WWII when
key need was flexibility.
• Taiichi Ohno and Shigeo Shingo developed
Lean based on TPS
Lean Manufacturing
 Cells or flexible assembly lines
 Broader jobs, highly skilled workers, proud of
product
 Interchangeable parts, even more variety
 Excellent quality mandatory
 Costs being decreased through process
improvement
 Global markets and competition.
What is Lean?

“A business system for organizing and


managing product development, operations,
suppliers, and customer relations that requires
less human effort, less space, less capital and
less time to make products with fewer defects
to precise customer desires compared with the
previous system of mass production.”

Lean Lexicon, Lean Enterprise Institute, 2003


Definition
Lean manufactuirng is aimed at the elimination of waste in
every area of production including customer relations,
product design, supplier networks and factory management.

The goal of Lean Manufacturing is to incorporate less


human effort, less inventory, less time to develop products,
and less space to become highly responsive to customer
demand, while at the same time producing top quality
products in the most efficient and economical manner.
Why Call it “Lean”?

• The term “lean” is used because lean


uses “less”…
Labor
Space
Capital investment
Materials
Time between the customer order and the
product shipment
Definition of Lean
Lean has been defined in many different ways –

“A systematic approach to identifying and


eliminating waste (non-value-added activities)
through continuous improvement by flowing the
product at the pull of the customer in pursuit of
perfection.”
Definition by the MEP Lean Network

Give the customers what they want, when they


want it, and do not waste anything.
Definition of Value Added

Waste is any activity that does not add value to the


final product for the customer.
• Value-added is an activity that transforms or shapes
raw material or information to meet customer
requirements.
• Non-value added is an activity that takes time,
resources or space, but does not add to the value of
the product or service itself.
• Non-value-adding, but necessary – does not add
value to the product or service but is required (e.g.,
accounting, governmental regulations, etc.).
Waste

“Anything that adds Cost


to the product
without adding Value”
Toyota Way – the 14 Principles by
Jeffrey K. Liker
1. Base management decisions on long-term philosophy, even
at expense of short-term financial goals.
2. Create continuous process flow to bring problems to surface.
3. Use pull systems to avoid overproduction.
4. Level out workload (heijunka); work like a tortoise, not a hare.
5. Build culture of stopping to fix problems, to get quality right
first time.
6. Standardized tasks are the foundation for continuous
improvement and employee empowerment.
7. Use visual controls so no problems are hidden.
8. Use only reliable, thoroughly tested technology that serves
people and processes.
Toyota Way - 14 Principles (cont.)
9. Grow leaders who thoroughly understand work, live
philosophy, and teach others.
10. Develop exceptional people and teams that follow company’s
philosophy.
11. Respect extended network of partners and suppliers by
challenging them and helping them improve.
12. Go and see to thoroughly understand situation (genchi
genbutsu).
13. Make decisions slowly by consensus, thoroughly considering
all options; implement decisions rapidly.
14. Become a learning organization through relentless reflection
(hansei) and continuous improvement (kaizen).
372
Liker Model
Toyota Terms
Genchi
Genbutsu
Problem
Solving
Kaizen
Respect and
People and Partners Teamwork

Process

Challenge
Philosophy

373
Liker Model - 4 Ps
1. Philosophy
• Long-term thinking even at expense of short-term financial
goals
2. Process
• Eliminate waste by focusing on flow, pull, workload
balance, error reduction, standardization, visual controls,
and jidoka or use of reliable, tested technology/automation
with mistake proofing and human touch
3. People and Partners
• Respect, develop, and challenge people
4. Problem Solving
• Fix and prevent problems by continual learning, going to
place where problem occurs, getting hands dirty, and
making good decisions based on fact
374
What Is Lean?
Lean is a methodology
• that allows organizations to drastically improve
bottom line
• by improving processes and monitoring everyday
business activities to reduce errors
• in ways that increase value and minimize work, non-
value-add tasks, and waste while increasing
customer satisfaction
- Based on idea that faster processes yield less waste,
less cost, less work in process, less complexity,
higher quality, and happier customers
Womack and Jones Model

1. Define value from customers’ perspective

2. Document value stream

3. Improve flow of value stream

4. Drive for pull versus push

5. Continuously improve
Lean Principles
• Create value for customer
• Understand:
– Who is customer – person or entity who is
recipient of product or service; one who places
value on output; catalyst/trigger in value chain
• Another business
• Someone inside own business
• Specific individual, group, or team
• Consumers—ultimate customer
– What customer considers valuable
• Make value flow based on customer’s needs
Lean Principles (cont.)
• Consider life cycle of information, materials,
processes, products, and services
• Look for process problems that prevent people
from performing best work
• Eliminate waste
– Non-value-add steps
– WIP
– Cost
• Standardize work
• Do not become distracted by other stakeholders
Value-Add Quiz
In which category should the following be
placed?
Activity Value Add Type 1 Type 2
Attending weekly team coordination meeting
Filtering through daily e-mail list
Reporting status to upper management
Gaining multiple approvals on documents
Gaining management approval for routine actions
Expediting document through approval list
Writing formal policies and procedures
Writing brief work-method instructions
Gaining regulatory or agency approvals
Creating ISO 9000 documentation
Hunting for needed information to do your job
Building “best practices” database
Holding lessons learned meeting
Spending time on process improvements
Lean vs. Traditional
Lean Traditional
• Simple and visual signals • Complex
• Demand driven • Forecast driven
• Inventory as needed • Excessive inventory
• Reduce non-value added • Speed up value-added
• Small lot size work
• Minimal lead time • Batch production
• Quality built • Long lead time
• Value stream managers • Inspected-in
• Functional departments
Benefits of Lean Manufacturing

Helps in – Real Results


0 50 100
• Cost reduction
• Cycle time reduction Lead Time
Reduction
• “Waste”
Productivity
minimization Increase
• Elimination of non-
WIP
value-added Reduction
activities
Quality
• Resulting in a more Improvement
“lean,” competitive, Space
agile, and market- Utilization
responsive company
Why the Emphasis on Lean Now?
• Global economy
• Pressure from customers for price reduction
• Fast-paced technological changes (e.g. Internet auctions)
• Continued focus on quality, cost, delivery
• Higher and higher expectations of customers
• Quality standards, such as QS-9000 (or TS 16949), the
new ISO 9000:2000
• Holding on to “Core Competencies,”outsourcing the rest
• Market-driven pricing: Customers expect better
performance at lower prices year after year
Evolution of Lean Across Markets
• Proven global concept since 1980s
• Transformed business processes across
many industries:
– Automotive
– Aerospace

• Other industries beginning to embrace


Lean concepts with excellent results:
– Construction
– Hospitals
– Pharmaceutical Manufacturing
– Service Organizations
Pricing Model
Costs Maintain Increase
Old Way Increase Profits Price

Cost + Profit = Price


Customers What
Demand If Costs
Profits can we
Lower Stay the
New Way Same
Decrease do?
Prices

Price - Cost = Profit


Pricing Model
Costs Maintain Increase
Old Way Increase Profits Price

Cost + Profit = Price


Customers
Demand
Implement Increased
Lower
Profits
New Way Prices Lean

Price - Cost = Profit


Core Concepts of Lean
• Creativity before Capital
• A solution that is not-so-perfect implemented
today, is better than a perfect solution that is late.
“Just do it.”
• Inventory is not an asset, but a waste/cost.
• Typically, 95% of lead-time is not value added.
• Lean implementation using the Plan-Do-Check-
Act methodology
• Continuous Improvement environment: both
incremental and breakthrough.
• Lean is a never-ending philosophy.
Aluminum Can Example

• From aluminum ore to usable cans, it typically


takes about 300 days

Bauxite Cryolite Aluminum Cast Product

3 hours
Cans Sheet Rolled Plate

Guess what the total value-added time is?


Video

• Introduction to Lean
7 Wastes of Lean

“OMIT What U DO” C


• Overproduction O
• Motion
• Inventory M
• Transportation (Movement) M
• Waiting W
• Defects (Correction)
• Over-processing I
• Underutilized People P
Overproduction

Making more-earlier-faster than the


next process needs it
• Printing 20 copies
• Just in case logic of a report that only
• Unbalanced workload 3 people look at
• Unleveled scheduling • All-staff e-mails
• False sense of efficiency when it pertains to
only a few
• Waiting to “batch”
work
Motion

Any movement of people that does not


add any value to the product or service
• Poor layout • Where is are copier,
• Inefficient Workplace printer, files and
Organization coffee-maker located?
• Lack of Standardization, • How far does the
inconsistent work paperwork travel?
methods
• People, Material and
Machine Ineffectiveness
Inventory

Any supply in excess of one-piece flow


• Just in case logic • Printed forms or tags
• Unbalanced that become obsolete
workload • 8 weeks of paper
• Unleveled scheduling located by the copier
• Unreliable suppliers • How many pens do
• Reward system you have in your desk
• “Pack rat” mentality drawer?
Transportation

Moving people, materials and


information around the organization
• Poor layout • Moving “banker’s
• Inefficient “flow” boxes to a storage
• Carrying large area
quantities • Mail carts
• Messenger services
Waiting

Waiting for… man, machine, materials,


information etc.
• Just in case logic • Waiting for files or
• Unbalanced workload information
• Unleveled scheduling • Need a signature
• Unplanned downtime • Customer reply to a
• Needs not voice-mail or e-mail
understood • Someone is printing
50 copies of a 70
page report
• Starting a meeting
Defects

Information, products and service that


need correction
• Not using Jidoka or • Have to fix paperwork
Poka-yoke that is not completely
• Lack of Standardization, filled in or track down the
inconsistent work right person to get the
methods information
• Ineffective • An entry error causes
communication the wrong actions like
• Little investment in shipping too many,
training or too few to the
wrong address, etc.
Over-processing

Effort that adds no value to the product or


service from the customer’s standpoint
• Just in case logic • Multiple sign-offs or
• Inconsistent work checks
methods
• Ineffective
communication
• Redundant approvals
• Excessive information,
extra copies
Underutilized People

Not utilizing people’s experience, skills,


knowledge, creativity
• Not utilizing Teams • Lack of suggestions
• Organization • “That’s not my job”
structure attitude
• Poor hiring practices • Waiting for lead from
• Little investment in management
training
Mura and Muri
• Mura (unevenness) – variation in operation,
wasted resources when quality, cost, or delivery
cannot be predicted
– Testing/inspection, Containment, Rework,
Returns, Overtime, Unscheduled travel
• Muri (overdoing) – unnecessary or unreasonable
overburdening of people, equipment, or systems
when demand exceeds capacity or tasks are not
designed properly including harmful, wasteful, or
unnecessary tasks
Building Blocks of Lean

Continuous Improvement & Kaizen Blitz


TPM JIT Cellular & Flow Pull System & Kanban

Poka-yoke Self Inspection Autonomation

POUS Batch Size Reduction Quick Changeover V


S
Layout Standard Work Visual 5S M

Change Management Teams


Building Blocks of Lean

Continuous Improvement & Kaizen Blitz


TPM JIT Cellular & Flow Pull System & Kanban

Poka-yoke Self Inspection Autonomation

POUS Batch Size Reduction Quick Changeover V


S
Layout Standard Work Visual 5S M

Change Management Teams


5 Words that begin with “S”

Japanese Translation Conversion *Other


Seiri Organization Sort Sorting
Simplifying
Seiton Neatness Set in order
access
Seison Cleaning Shine Sweeping
Seiketsu Standardization Standardize Standardize
Self-
Shitsuke Discipline Sustain
discipline

* There are several other conversions


5 Steps to Workplace
Organization
1
Clear/Sort
By red tagging

5
Continuous Improvement 2
Sustain Organize/Straighten
Discipline A place for

Waste everything

3
4
Clean/Sweep
Maintain/Standardize
Housekeeping /
Establish standards
Inspection
Workplace Scan
“Understand your Current State”
• Start with a workplace Scan
• Team Based Workplace Scan Display

Area Checklist
• Define the boundaries Details Score

• Complete a Diagnostic Checklist Spaghetti Diagram

• Draw a Spaghetti Diagram Before


Photos
After
Photos

• Take “Before” Photos


• Starts the 5S Program
Sort
“When in doubt, move it out”
• Move unneeded items out of the area
• Use the Red Tag Technique
• Use a Temporary Red Tag Holding Area
Name____ Date___
• Criteria for unneeded items Item _____________

– “30-day Rule” Reason _________

• Keep only what you need in the area


Set in Order
“A place for everything and
everything in its place.”
• Make it easy for anyone to find
– “30-second Rule”

• Make it obvious if an item is out of place


• Decide where to keep items, how many items to
keep, how and when to replenish items
• Make it Visual
Shine
“Clean and Inspect”
• Get items to a like-new condition
– “10 Second Rule”

• Must plan Shine – assignments & supplies


• Perform as a Team
• Prevent dirt, grime, or contamination
• Repair as needed
Standardize

“Create the rules and follow them”

• Determine how the first 3S conditions are


met

• Use “One-Point Lessons”

• Maintain and monitor the conditions

• Use Visual techniques


Sustain
“Make 5S a habit”
• 5S is not something additional, it is part of
everyone’s daily job
• Supports discipline
• Train
• Communicate
• Support from Management
• Reward and recognition
5S is Fundamental to Lean

• 5S is directly related to other Building


Blocks
– Teams

– Visual

– POUS

– Standard Work

– TPM
Point-Of-Use Storage

• Raw material and WIP are stored at


workstation where used, which reduces the
inventory that can be carried.
• Works best if vendor relationship permits
frequent, on-time, small shipments (JIT).
• Simplifies physical inventory tracking,
storage, and handling.
Carpenter Story

• Does the
carpenter walk
back to the
toolbox every
time a tool is
needed?
• Which waste is
this?
• What does the
carpenter do?
Proximity
• Typically, up to 60% of time is spent on
finding/collecting items needed.
• Minimize non-value activities
• Store as close as possible and within reach
• Layout and workstation design should
accommodate required materials
• Try to use the packaging from the supplier
or have the supplier change packaging
POUS Components

• Have the:

– Information

– Parts & materials

– Tools & equipment

that you need to perform you tasks within reach


POUS Workplace Zones

• Items used most often


(i.e., daily) should be
kept within reach
• Items used less often
(i.e., weekly) should be
kept close-by
Daily
• Items used rarely (i.e.,
Weekly
monthly) should be kept
in the vicinity Monthly
Location of items

• Use horizontal
transfers and gravity
feeds when possible
• Support heavy objects
• Set items
ergonomically
• Must be comfortable
for a day’s work
Benefits

• Supports 5S & Visual and other Building

Blocks

• Simplifies inventory tracking and accuracy

• Reduces waiting, inventory, motion and

transportation waste
Benefits of 5S
• Improved equipment reliability
• Superior quality
• Increased productivity
• Better workflow
• Enhanced Safety
• Reduced inventory
• More pleasant place to work
• Impress customers
Video

• Introduction to 5S
Building Blocks of Lean

Continuous Improvement & Kaizen Blitz


TPM JIT Cellular & Flow Pull System & Kanban

Poka-yoke Self Inspection Autonomation

POUS Batch Size Reduction Quick Changeover V


S
Layout Standard Work Visual 5S M

Change Management Teams


Visual

Signs, lines, labels and color coding


Why Visual?
• What you need to know
• Cockpit view
• Information sharing

How do you know where to park when you


drive to a shopping mall?
Does someone have to tell you where to
park?
How to Apply Visual
• Use Signs, Lines, Examples
Labels and Color- • Productivity Goals
coding • Quality Goals

• Charts, pictures, • Delivery schedules

lights, scoreboards • Set-up specification


• Safety Initiatives
• Kanban, Andon lights
• Attendance Goals
• Inventory Levels
• Team Objectives
More Examples of Visual
• Signs • Tags
• Forms
• Charts
• Training hours
• Goals
• Employees’ suggestions
• Pictures • Cross trained skills
• Color coding • Employee awards
• Lights • Absenteeism
• Scoreboards • Critical maintenance
points
• VSM Current & Future
• Customer satisfaction
State goals
• Standard Work • Performance targets
instructions
Visual Controls as Communication
Tools
• Visual controls expose waste so we can
reduce or eliminate it
• Visual controls help in:
– Improving motivation & morale
– Focus on safety
– Pride of workplace & workmanship
• Visuals and performance metrics
– What gets measured, gets done
– Policies drive behaviors
World Class Visual Controls

• Anyone knows what’s going on by looking

around

• You do not have to wait for information to

do your job

• Everyone passes the 30 second test


Visual Examples

Andon Lights

Display Panels

Range Markings
On Gauges
Shadow Boards
Standard Work

Reduce task variability


Standard Work

Best sequence of operations, using the most


productive combination of resources:
• Man, machine, materials, changeovers,
etc.
• Details any special skill/knack needed
• Safety, ergonomics are integrated
• Tool for perfect quality and efficiency
Standard Work Properties

• Specific

• Measurable

• Repeatable

• Documented
Standard Work
Identifies value added versus non-value added
activities
• Reduce or eliminate non-value added activities
• Convert Internal Time to External Time,
wherever possible
• Continuous Improvement: once Standard Work
is established as a base and displayed at
workstations, operators monitor and implement
improvements
• Use as a training tool for new employees
• Created by input from the people who actually
work in the process
Philosophy of Standard Work

Be specific about: Example: Installing a Car Seat


• Content • Bolts are installed and
• Sequence tightened in the same exact
• Timing order
• Outcome • The time required to tighten
bolts is stated and followed
• Specified torque is applied
and checked
Used as the basis from where the next level of
improvement is made.
Types of Standardized Work Forms

• Process Capacity Table

• Work Combination Sheet

• Standard Work Sheet

• Others forms may be used based on your

organization’s needs
Process Capacity Table
Times for: Changeover Analisys Chart

Process: Date: Name: Shift:

NVA = X
• Elapsed time Seq

1
No. Element
Review work order for die number and material
Symbol
Elapsed Element Internal External
Time
600
Time
600
Time Time Comments
300 Convert to External
2 Locate the die 1800 1200 300 Convert to External
• Element time 3
4
Locate material
Get tools
3000
3600
1200
600
300 Convert to External
300 Convert to External

• Internal Time 5
6
Tagout machine
Loosen bolts
4500
4980
900
480
60
60
Prepare Tags before
Use quick connects
7 Disconnect hoses 5280 300 60 Use quick connects

• External Time 8
9
Remove die from press
Return die to Tool Room
5580
6180
300
600
120 Position die cart
300 Convert to External
10 Load die into press 6480 300 90 Use positive stops

• Manual Time 11
12
Align die
Tighten bolts
6960
7440
480
480 60
0 Eliminate
Use quick connects
X

13 Connect hoses 7740 300 60 Use quick connects


Step Identification 14 Position material 8340 600 300 Convert to External
15 Clear tagout 8940 600 60

• Transportation 16
17
Make sample piece
Take first piece sample to QC
9000
9600
60
600
60
120 Have QC Tech ready
18 Adjust die and press 10800 1200 0 Pre-adjust die X

• Process 19 Make sample piece


20 Return tools
10860
11460
60
600
0 Eliminate
600 Convert to External
X

• Inspection Transport Process Inspection Storage Total Time 11460 750 2400 Page ____ of _____

• Storage
Use to identify bottlenecks Use to calculate the capacity of
machines, and identify bottlenecks
Work Combination Sheet
Sequence of: Product Fam ily: B - Generators Tim e Available: 24,600 seconds Date: January 31, 200X

• Manual work time Process: Assembly Cell #23


Description: Final Assembly
Cyclical Work Elem ent
Dem and:
Takt Tim e:
Element Time
200
123
pieces
seconds per piece
Nam e: Eileen N. Terprise
VS Mgr: I.C. Flow
Graph Data
Operation Times

• Machine
Manual Machine
ID Description of Work Element Operation Operation Walking Start Manual Operation
Machine Operation
Walking
1 Manually load WC-1 12 32 0 1 44 0 0 12 32 0
2 Start WC-1, go to WC-2 3 0 2 2 5 44 49 3 0 2

operations time 3
4
5
Unload part, Load Part, Start Machine, go to WC-3
Unload part, file corner, inspect to print
Load next part, start machine, go to WC-1
1
45
2
16
120
15
2
0
3
3
4
5
19
165
20
49
68
233
68
233
253
1
45
2
16
120
15
2
0
3

• Walking
6 Fasten parts A-1, and B-1 together, go to AS-2 23 41 2 6 66 253 319 23 41 2
7 Fasten Housing and Base 42 0 0 7 42 319 361 42 0 0
8 Package generator 35 0 0 8 35 361 396 35 0 0
9 Load on cart 5 0 4 9 9 396 405 5 0 4
10 0 0 0 10 0 405 405 0 0 0
11 0 0 0 11 0 405 405 0 0 0
12 0 0 0 12 0 405 405 0 0 0
13 0 0 0 13 0 405 405 0 0 0

Shows the
14 0 0 0 14 0 405 405 0 0 0
15 0 0 0 15 0 405 405 0 0 0
16 0 0 0 16 0 405 405 0 0 0
17

interactions
17 0 0 0 0 405 405 0 0 0
18 0 0 0 18 0 405 405 0 0 0
19 0 0 0 19 0 405 405 0 0 0
20 0 0 0 20 0 405 405 0 0 0

between machines 21
22
23
0
0
0
0
0
0
0
0
0
21
22
23
0
0
0
405
405
405
405
405
405
0
0
0
0
0
0
0
0
0

and operators 24
25
26
0
0
0
0
0
0
0
0
0
24
25
26
0
0
0
405
405
405
405
405
405
0
0
0
0
0
0
0
0
0
27 0 0 0 27 0 405 405 0 0 0
28 0 0 0 28 0 405 405 0 0 0
29 0 0 0 29 0 405 405 0 0 0
30 0 0 0 30 0 405 405 0 0 0

Allows to recalculate Total Time:


Percent Operator Time:
168
41%
224
55%
Total Cyclic Time:
13
3%
405
0 100 200 300 400 500

operator work 1
Non-cyclical Work Elem ents
Time
Seconds
PCS Rate
Total Time (Cyclical + Non-cyclical):
Throughput per shift = Available Time / Total Operating Time:
Work Flow Diagram
405

content as takt 2
3
4
WC-1 WC-2 WC-3

AS-1

time changes 5
Total Non-cyclic Time:
AS-1 AS-1
Standard Work Sheet

Sequence of processing steps


• Worker Standard Work Sheet

Product Family

• Machine B - Generators

Process
WC-1 WC-2

Assembly Cell #23

• Tools Description

Final Assembly
1 3 2

• Layout

WC-3
Supplier

Fab, Assm #17, SM-29


1
Customer 3

• Material location SM-FG48, Shipping


2

(Standard stock) 2 1

-1
AS
FS-1
AS-2

Displayed at Workstations

Continuously reviewed and updated


Uses
• The Process Capacity Table can be used
to reduce changeover times

• The Work Combination Sheet can be


used for line balancing when creating a cell

• The Standard Work Sheet can be used for


training and team development
Standard Work Examples
Standard Work and Training
Questions to ask Operators about Standard Work
• How do you do this work?
• How do you know you are doing this work
correctly?
• How do you know there are no defects?
• What do you do if you have a problem?

If these questions cannot be answered


satisfactorily, then either the Operator needs
additional training or the Standard Work is
unclear
Uses of Standard Work

• Consistent performance of tasks = better


quality
• Track performance = actual versus standard
for continuous improvement
• Easy to Train = reduced learning cycle time

Old Learning Curve New Learning Curve


Productivity

Productivity
Time Time
Benefits of Standard Work
• Standard documentation for all shifts
• Reductions in injuries and strain
• Employee ownership of process
• More pleasant working conditions; higher
morale
• Better than traditional time and motion
studies
• Reduced variability
Poka-yoke
• Error proof (mistake proof) takes away the
possibility of human error
• The term Poka-yoke was made popular by
Shigeo Shingo
• Fail-safe devices
• Low cost, highly reliable mechanisms
• Detects abnormal situations before they occur,
or
• Once they occur, will stop the equipment from
further production. The machine stoppage
makes the problem visible.
Other Poka-yoke Examples

• USB ports on computers

• Re-typing passwords to verify

• Computer prompts before deleting file

• Bar codes & scanning

• ATM swipe card or beep


Benefits of Poka-yoke
• Gives immediate feedback for root cause
analysis & correction (and prevention for the
future).
• Failure Mode and Effect Analysis (FMEA)
solution can be Poka-yoke
• Some examples of Poka-yoke devices are:
sensors, counters, feelers, limit switches, electric
eyes, probes, automatic stops.
Building Blocks of Lean

Continuous Improvement & Kaizen Blitz


TPM JIT Cellular & Flow Pull System & Kanban

Poka-yoke Self Inspection Autonomation

POUS Batch Size Reduction Quick Changeover V


S
Layout Standard Work Visual 5S M

Change Management Teams


Value Stream Mapping

See the Flow


Value Stream Analysis
Value stream analysis encompasses all activities
company must do to design, order, produce, and deliver
its products and services to customers
– Flow of tasks, from request for service (trigger event)
to service complete, from receipt of materials or
information from suppliers to delivery of finished
product or service to customers
– From viewpoint of customer, service, or transaction
– Flow of information that supports and directs both flow
of materials and transformation of raw materials or
information into finished goods or services
Purchasing Human Operations Finance Sales
Resources
Value Stream

Value Stream
Purpose of Value Stream Map
• Has customers’ perspective and focuses on meeting
customers’ wants and needs
• Starts with immediate customer and maps back to
receiving inputs from suppliers and shows how fits
into overall value stream
• Provides single view that is a complete, fact-based,
and time-based representation of stream of activities
• Provides common language and view for analysis
• Shows how information triggers and supports
activities
• Shows time for activities and whether they add value
Elements of Value Stream Map
• Process steps
• Value-add classification for each step
• Information flow such as orders, requirements, schedules,
messages, approvals, specifications, kanban signals, shipping
information, standard procedures
• Box score of key operational metrics including cycle time, waiting
time, working time, conveyance time, distance traveled, items
per shift, items processed per hour,setup time, backlog/work-in-
process, amount of inventory between last step and consumer,
defects, cost information, resource availability and active time,
process variations
• Lead time is amount of time for one item to flow completely
through process, noted along bottom of flow
• Takt time showing customer demand rate, in upper right corner
of flow
Value Stream Analysis Steps
• Identify deliverable, value stream, and sponsor who has authority and
responsibility to allocate resources and make changes across
organization
• Identify customer and value from customer’s perspective as well as
regulatory, legal, and compliance requirements
• Draw visual representation of process current state, generally draw
steps starting at consumer’s view working back through steps to
sources of material and labor; flows from left to right with time, with
steps in order of occurrence
• Add metrics and observations like Takt time/throughput, cycle times,
defect rates, and inventory/work-in-process and information flows to
identify magnitude and frequency of waste
• Use lean principles to reduce or eliminate waste and reduce cycle time
• Develop future state map, document steps of process that need to
happen, and prioritize and implement action plans to achieve future
state
Value Analysis Matrix Steps
• Structure value analysis matrix
• Number process steps on sub process map
• Have column for each process step
• Estimate time for each process step
• Place check in category for each process step,
either value-add or one of non-value-add
categories
• Total number of hours or number of checks for
each row
• Report percentages of value-add and non-
customer required
Value Analysis Matrix Example
Process Step 11 22 33 44 55 66 77 88 99 10
10 Total % Total

Time (Hours) 1212 1010 1 1 1010 2020 6 6 1010 1 1 1010 2020 100
100 100%
100%

Value -Add
-Added   22 2%
2%

Non Customer Required

Internal Failure

External Failure  1010 10%


10%

Control/Inspection  66 6%
6%

Delay    5252 52%


52%

Prep/Set -Up
Prep/Set-Up

Move    3030 30%


30%

Total 100
100 100%
100%
RFQ Creation Value Stream Map

Iterate

Revise

Triggering
Event Measurable
Deliverable
Continue Continue As-Is Process
Cycle Time*:
C/T = Calendar Time
W/T = Work Time C/T = 58 days
VA/T = Value-Add Time W/T = ~14 days
VA/T = 5 days
Revise Assumes no revisions!
Sales Order Processing Value
Stream Map C/T = Calendar Time
W/T = Work Time
VA/T = Value-Add Time
Time Customer is On Telephone

Change Ship Date


Triggering
No
Event
Yes
Mtl.
Measurable Available
Deliverable ?

While customer C/T = 60 min. From Contact C/T = 17 days


is on telephone: W/T = 55 min. to Order Launch: W/T = ~ 1 day
VA/T = 20 min. VA/T = 0
Buying a Salad Process Flow
Customer Grocer offers Salad company Salad company
buys salad premade makes salad and buys supplies
with salmon salads delivers to grocer

Processed Salmon is Mother


Fishers catch
salmon is shipped processed at nature makes
salmon
to fish markets fishery salmon

Produce is package Farmers grow and


Restaurant and shipped harvest produce
supply company
distributes food Meat companies Farmers raise
process animals meat animals

Container Manufacturers Chemical producers Oil refined for


company sells make make plastics from petroleum
containers containers petroleum products
Cycle Time Definition
“One of the most noteworthy accomplishments in keeping the
price of products low is the gradual shortening of the cycle
time. The longer an article is in the process and the more it is
moved about, the greater is its ultimate cost.”
Henry Ford, 1926

Work Errors, waiting, transportation, movement


etc…..

Total Cycle Time


• Time that elapses from beginning to end of process
• Ultimate objective or goal of Lean processes is to reduce cycle
time by eliminating waste
Benefits of VSM
• Helps you visualize more than the single
process level

• Links the material and information flows

• Provides a common language

• Provides a blueprint for implementation

• More useful than quantitative tools

• Ties together lean concepts and techniques


4 Steps for VSM

1. Determine the Product Family

2. Draw your Current State Map

3. Create the Future State Map

4. Develop your plan to get there


Current State Map

• Understanding how the floor currently


operates
– Material and Information flows
– Draw using symbols
– Start with the “door to door” flow
– Have to walk the flow and get actuals
• No standard times
• Draw by hand, with pencil and eraser
– Foundation for the future state
Current State Icons
Process Box
Customers Go See
Painting
Suppliers

Mon., Shipment- Operator


Wed., Truck Data
Fri.
Box
I Inventory C/T=1 sec Cycle Time
C/O= 1 hr Changeover
Push System Rel.= 98% Reliability
FPY = 95%
Quality
More Current State Icons

Hardcopy
Train Electronic

Cell

Boat
Person
Plane
Fun Current State Icons

?
?
?
Fax

o
&@#$
%!
Current State Map Setup
Tips
•Use 11” x 17” paper, landscape
•Use pencil and eraser
•Draw by hand
•Don’t waste time putting it on a computer just to
make it look nice (non-value added time)
•Practice, practice, practice
Steps
•Customer
•Supplier
•Process
•Information flow
•Calculate process time and lead-time
Current State Map Setup

Supplier Customer
information Information Flow Area information

Process Flow Area

Process Time and Lead-time Area


Title Block
Current State Map
2 days
20 min
30/60/90 Prod Ctrl Order Entry
Forecast Monthly Dewey,
Phlye- MRP MRP
Weekly Cheatem &
Biknight L/T= 2 days Howe
Weekly 5,300 pcs/mo.
P/T = 20 min 265 pcs/day
Weekly
Weekly Schedule Daily
Daily

Stamping Spot Weld Deburr Assemble


Shared
I I I I
2 Weeks =1 5,425 =1 1,400 =1 1,225 =2
C/T=1 sec C/T=39 sec C/T=17 sec C/T=48 sec
C/O= 4 hrs C/O= 11 min C/O= 0 min C/O= 5 min
Rel.= 98% Rel.= 99% Rel.= 80% Rel.= 100%
FPY = 95% FPY = 90% FPY = 100% FPY = 98%
10 days 20.5 days 5 days 4.5 days 40 days
1 sec 39 sec 17 sec 48 sec 105 sec
Future State Questions
1. What is the Takt Time?
2. Will we build to shipping or to a supermarket?
3. Where can we use continuous flow?
4. Where do we have to use supermarket pull system?
5. At what single point in the production chain do we
trigger production?
6. How do we level the production mix at the pacemaker
process?
7. What increment of work will we release and take away
at the pacemaker process? (Leveling the volume)
8. What process improvements will be necessary?
(e.g. uptime, changeover, training)
1. What is the Takt Time?
• Takt means drumbeat

• Ability to meet customers’ demand

• Formula
Time Available
Takt Time =
Demand
Takt Time Calculation

Time available
Shift (8 hours) = 480 mins
Breaks (2 x10) - 20 mins
Lunch - 30 mins
Meetings - 5 mins
C/O - 5 mins
Total Time = 420 mins = 25,200 sec
Demand = 265 parts
Takt Time = 95 sec/part
Takt Time Calculation
• Takt Time = Demand Rate
– Goal: Produce to demand with no excess
capacity
• Takt Time = work time available number of
units sold
• Assume 5 people work, sell 500
units/week:
– Takt Time = (5 x 40 x 60) / 500 = 24 min/unit
– Set cycle time to match personnel/operation
– For three-step process, perfect is 8 min/step
Exercise

• Takt time calculation


Takt Time Calculation
• Old requirements: 847/day * 240 workdays/yr =
203,000/yr
• 10% growth = 223,600/yr
• New requirements: 223,60/yr/240 workdays/yr =
931/day
• Time available: 8.5 hrs/day - .5 hrs (lunch) - .33 hrs
(breaks) = 7.67 hrs/day
• 3,600 secs/hr * 7.67 hrs/day = 27,612 seconds/day
• 27,612 seconds/day divided by 931 units/day = 29.3
secs per unit
• Cycle time (actually 116 secs/unit) divided by Takt
time (29.3 secs/unit) = 3.95 = 4 operators required
Quick Changeover

Changeovers in less than 10 minutes


Quick Changeover
• Factory definition – the time from the last
good piece of previous run to the next good
piece of new run

• Office definition – the time it takes to switch


from one task to a new task
– Typically, in the office the time savings is not
as significant as in manufacturing
2 Hour C/O – Large Batch Size

8
7
6
5
4 A B C D E
3
2
1
0
Mon Tue Wed Thu Fri
1 Hour C/O – Half Batch Size

8
7
B D A C E
6
5
4
3 A C E B D
2
1
0
Mon Tue Wed Thu Fri
10 Minute C/O – Small Batch Size

8
E E E E E
7
6 D D D D D
5
4 C C C C C
3 B B B B B
2
1 A A A A A
0
Mon Tue Wed Thu Fri
10 Minute C/O – Many Different Small
Batches

8
B
7 D E
6 D C D
E F B
5
H
4 C C
3 B
B A H G
2
1 A F G A B
I
0
Mon Tue Wed Thu Fri
Changeover Summary
Based on 40 hours per week
Number of Production
Number of
C/0 Time Production Time
Changeovers
Runs Available

2 Hour 5 5 30 hours

1 Hour 10 10 30 hours

30 Minute 15 15 32.5 hours

10 Minute 25 25 35.8 hours


SMED

• Single Minute Exchange of Dies (SMED)


• Shigeo Shingo (1970)
1. Separate internal steps and external
steps
2. Convert internal steps to external where
ever possible
3. Streamline all steps
4 Categories of SMED time

1. Preparation, after-
process adjustments,
30% 5%
checking of material
and tools (30%)
15%
2. Mounting, removing
tools and parts (5%)
3. Measurements,
settings and 50%
calibrations (15%)
Typical proportions
4. Trial runs and
adjustments (50%)
Step 1. Separate Internal and
External Times

• Checklists

• Functional checks

• Transportation of parts and tools


Step 2. Convert Internal Time to
External Time
• Preparation conditions
– Pre-heat, correct air pressure, stage materials
• Standardize
– Centering, gripping, securing, replace fewest
parts, standardize heights, standardize bolts or
fasteners
• Intermediary jigs
– Mounting plates
Step 3. Streamline
• Parallel operations
– More than one person working at the same
time

• Eliminate adjustments
– Markings
– Scales

• Functional clamps
• Mechanization
Changeover Cart Example

Before After
• Waste of time to find • Saves time
correct tools • Do not have to replace
• Tools can become tools as often
damaged • Have what you need
• Waste of money for extra where its needed
tools
QCO and Other Building Blocks
• 5S, Visual, POUS, Teams and Standard
Work
• VSM can discover opportunities for QCO
• As Batch Size Reduction continues, QCO
becomes more important
• Kaizen Blitz is a great method to implement
QCO
• Must sustain the gains
Benefits of Quick Changeover
• Shorter lead time
• Less material waste
• Fewer defects
• Less inventory
• Lower space requirements
• Higher productivity
• Greater flexibility
• Better Teamwork
Building Blocks of Lean

Continuous Improvement & Kaizen Blitz


TPM JIT Cellular & Flow Pull System & Kanban

Poka-yoke Self Inspection Autonomation

POUS Batch Size Reduction Quick Changeover V


S
Layout Standard Work Visual 5S M

Change Management Teams


Continuous Improvement
Kaizen vs. Kaizen Blitz, or Incremental vs. Breakthrough
Improvements
Kaizen Incremental Improvements:
• Are continuous, since there is always room for
improvement in any process
• It is never-ending
• Many small improvements throughout the enterprise
• Done by individuals or small teams
• Could be functional, departmental, or task-oriented
• Part of the “useful many”
• A little time spent on an ongoing basis
• Standardization of processes
(i.e., process improvement oriented)
• Plan-Do-Check-Act methodology
Continuous Improvement

Ideas for continuous improvement could come from:


– Employee suggestions
– Corrective & Preventive actions
– Non-conformities, defects
– Customer complaints, returns
– Benchmarks
– The Lean “wastes”
– Variations from the standard
– Assessments, audits & competitive analyses
– Research & Development activities
Continuous Improvement & Continuous Learning

• Continuous Learning goes hand in hand with


continuous improvement
• Management should have training given to
employees in Lean and Quality tools, problem
solving and root cause analysis, the process
model, concepts of Theory of Constraints, basic
statistical techniques, graphical tools, etc.
• Understanding of Plan-Do-Check-Act and
Standardize-Do-Check-Act (SDCA) will be
beneficial
Continuous Improvement & Continuous Learning

• Continuous improvement and learning:


– Becomes part of daily work life
– Is practiced at both personal, functional and
organizational levels
– Is result oriented
– Is shared within the enterprise
– Becomes part of institutional memory and
knowledge, even after employees retire, move
up/laterally or leave
Why C. I.?
• Standing still is not an option:
– Competitors will overtake us
– Globalized economy
– Higher customer expectations
– Technical and breakthrough changes
– Tapping into human potential and creativity
• Improvements based on:
– Cost and cycle time reduction
– “Waste” minimization
– Defect prevention
– Enhancing customer satisfaction/delight
– Attaining competitive advantage
Kaizen Blitz

• Breakthrough strategies for lasting


results
Building Blocks of Lean

Continuous Improvement & Kaizen Blitz


TPM JIT Cellular & Flow Pull System & Kanban

Poka-yoke Self Inspection Autonomation

POUS Batch Size Reduction Quick Changeover V


S
Layout Standard Work Visual 5S M

Change Management Teams


Kaizen Blitz
• Kaizen Blitz is a combination of the Japanese word
Kaizen for “continuous improvement” and the German
word Blitz for “lightning.” It is a focused, week-long
workshop where a cross-functional team reviews a
process, identifies and eliminates waste, thereby
achieving dramatic and tangible breakthrough (rather
than incremental) improvement results.
• Kaizen Blitz now stands to mean the improvement
activity itself.
• It is treated more as a “Project” (rather than a
“Process”).
Why Kaizen Blitz?
• Major benefits in a flash.
• Can use benchmarking for setting goals.
• Innovation has become indispensable in today’s
competitive world economy.
• Cycle Time Reduction translates directly to cost
savings.
• Kaizen Blitzes typically attack wasted time.
• Positive impact on organizational culture through
“breakthrough” type improvements.
Who Will be Involved?
• Kaizen Blitz teams that come together for one week to
implement improvements in a pre-selected bottleneck project
or process.
• Cross-functional teams (seven to ten persons)
• Hourly and salaried personnel
• Operators, engineers, supervisors, maintenance persons,
managers, technical experts, material handlers, quality
personnel, business support personnel, participants from the
outside
• Typically, the Team Leader is person with clout and is the
highest stakeholder in the process who possess leadership
skills, open minds, strong desire to succeed and some prior
Lean experience.
Kaizen and Cycle Time Reduction

• Focus on Process.
• Focus on Elimination of Waste.
• Focus on Speed.
• Time improvement translates directly to
cost savings and customer satisfaction.
Kaizen Blitz Steps
1. Select specific area for improvement.
2. Define current situation in measurable
terms
3. Set aggressive goals (stretch goals).
4. Identify team members.
5. Conduct training on the first day of the
project.
6. Do it (in three to five days).
Step 1 – Select Project
• Value Stream Map
• Bottlenecks
• Customer or quality related issues
• Interdepartmental
• Long lead-times or setup times
• Competitive advantage
• Cost reduction or avoidance
Step 2 – Define Current State
• Video tape
• Time study
• Flowchart
• Historical information
• Observations and interviews
Step 3 – Identify Team Members
• Cross-functional teams (seven to ten persons)
• Hourly and salaried personnel
• Operators, engineers, supervisors, maintenance persons,
managers, technical experts, material handlers, quality
personnel, business support personnel, participants from
the outside
• Typically, Team Leader is person with clout and is the
highest stakeholder in the process. Ideally, the team
leader must possess leadership skills, open minds, strong
desire to succeed and some prior “Lean” experience.
Step 4 – Set Aggressive Goals
• Define the purpose or objective and set stretch
goals.
• Goals should be clearly defined and quantifiable
(e.g., reduce machine set up time by 75%,
increase throughput by 40%, reduce floor space
by 30%).
• Emphasis should be on identifying and
eliminating waste, and then standardizing at the
improved level.
• Benchmark when possible
Step 5 – Conduct Training
• Select hands-on training that is compatible
with the project
• Do training on the first morning
• Provided by an internal or external expert
Step 6 – Do it!

• Perform in 3 to 5 days
Example 1 – Universal Joint QCO

Category Changeover Tool Change


Target savings 4.9 hours (75%) 14.8 hours (75%)
Actual savings 4.5 hours (69%) 13.8 hours (70%)
Target savings
$291 $879
per day
Actual savings
$268 $820
per day
Total savings $272,041 per year
Example 2 – Tube Mill SMED

• Changeover savings per year: $510,000


• Reduced changeover time from 4 hours, 40
minutes to 2 hours, 11 minutes
• Employees happier – bonuses based on
changeovers
• Management happier – more changeovers,
more production, more cash in the door
How to Start and Sustain your Lean
Journey

• A journey of one thousand miles, starts with


a single step
8 Ways to Get Started
1. Baseline Assessment or Gap Analysis
2. Value Stream Map
3. Training in Lean
4. Basic Building Blocks
5. Kaizen Blitz
6. Pilot Projects
7. Change Management
8. OEE
1. Baseline Assessment

Baseline Assessment or Gap Analysis performed


by experienced Lean experts
• Use
– Interviews
– Observations
– Process mapping
– Analysis of reliable data
• Create a “Gap Analysis” with focus on
eliminating the Eight Wastes
• Generate an Action Plan for implementing Lean
improvements
2. Value Stream Mapping
Value Stream Mapping
• Assemble the cross functional team
• Have a Value Stream Manager
• Determine a Product Family
• Create the Current State Map
• Create the Future Stats Map
• Develop the Plan to get there, tie-in with business
objectives
• Review the Plan, stay on course
• Your Future State then becomes your Current State
• Expand to Multiple Value Streams
3. Training in Lean
Training in Lean
• “Massive” training in Lean
• Need to build a critical mass of trained
employees
• Perform the training just before
implementation
• Lean Champions should have advanced
skills in Lean
4. Basic Building Blocks
• Start with one of the Basic Building Blocks
of Lean

Continuous Improvement & Kaizen Blitz

TPM JIT Cellular & Flow Pull System & Kanban

Poka-yoke Self Inspection Autonomation

POUS Batch Size Reduction Quick Changeover V


S
Layout Standard Work Visual 5S M

Change Management Teams


Basic Building Blocks
Basic Building Blocks
• Start with the implementation of the Basic
Building Blocks
• Build up layer by layer until TPM, Cellular
Manufacturing and Pull/Kanban are
established
• Then continuously improve using Kaizens,
suggestion systems and periodic Value
Stream Maps.
Example – Implementing 5S

5S Implementation Map
• Plan
– Identify 5S Champions 8 9
7 8 8 9 9 10 13
& Teams 8 8 9 9
– Decide how to roll-out 1 5 4 2
10
to entire organization 11
1 4 3
– Resources required 14 14
5
6 12
• Train
– Train just before
Kaizen
– Train-the-trainer
• Do
• Improve & Repeat
5. Kaizen Blitz
Select a Kaizen Blitz project
• Perform in 3-5 days
• Focus on speed and elimination of waste
• Can perform on “low hanging fruit”
• Generates quick victories and
improvements
• Do not continue in an Ad-hoc approach,
use your Value Stream Map
6. Pilot Projects
Pilot Projects
• Implement Lean Pilot Projects where bottlenecks
have been identified
• Use cross-functional teams
• PDCA methodology is best
• Can use benchmarks and best practices for goal
setting
• Communicate results
• Migrate lessons learned to other areas
7. Change Management
Change Management
• Begin with cultural Change Management before
rolling out Lean
• Address the human side of Lean in your three to
five year Master Plan
• Use internal/external change agents
• Communicate the need for change: ultimately,
Lean has to become integrated into daily work life
• Open up channels for sharing ideas
8. OEE

Overall Equipment Effectiveness (OEE) analysis


can identify where to start your Lean journey
• Pareto the time spent on:
– Breakdowns
– Setups
– Tool changes
– Idling time
– Slower speed
– Minor stoppages
– Producing defects, rework
– Start-up issues
• This exercise will self-identify the “biggest bang
for the buck” and where to start
Plan
• You must have a plan

• Utilize a Steering Committee, Design


Teams and Lean Champions

• Tie the Lean Objectives with the Business


Objectives

• Commit resources (time, people, budgets)


Stages of Lean Implementation
Generally, organizations can use this model
for the stages of implementing Lean
• Takt
– Establish takt time and meet it
• Flow
– After meeting takt time, then create Flow
• Pull
– Where you can’t Flow, Pull
How to Sustain Lean?

• Lean will not be sustainable without proper


training in Lean and satisfied employees

• Internalize into daily work

• Understand that it is a never-ending process or


philosophy: no turning back

• Create discipline/motivation/incentives

• Standardize so as not to slip back


How to Sustain Lean?

• Continued, visible management commitment

• Open communication channels

• Emphasize accountability

• Use Lean performance metrics

• Role of Lean champions

• Job rotation
How to Setup the Lean Team

• Steering Committee
• Design Teams
• Champions
Lean Team Roles & Responsibilities

Design Team
Lean Steering HPT
Committee Design Team

5S, Visual, POUS


Design Team Pull/Kanban
Cellular
Design Team
Design Team

Champions Champions
Champions
Steering Committee

Roles & Responsibilities


• Set the Lean policy
Lean Steering • Provide resources – time, people,
Committee budgets & remove barriers
• Develop and share the Lean Vision
• Develop the Communication Plan
• Members usually and then deploy it
from top • “Walk the talk” everyday and fully
management, but support the Lean initiatives
can include “Value
Adders” • Determine the Design Team make-
up and members
• Review the work of the Design
Teams
Design Teams

Roles & Responsibilities


• Deploy the Lean policy
Design Team • Determine the resources
required
– Time, people, budgets, etc
• Determine the best way to
• Members usually implement the Lean Building
from management or Blocks in your organization
“Value Adders”
• Report to the Steering
• Group like Building Committee on progress
Blocks together
• Support your Lean
Champions
Design Team Agenda

• Design Teams decide how • Design Teams may change


Lean will be implemented at over time
their facility – At first they oversee the
• Take into account: implementation plan
– Organizational Culture – Then they support the
– Change management sustaining efforts
– Resources – They may disband, be
– Current skills & needed absorbed into another
skills Design Team or morph
into a new design Team
– Size and timing of
projects – Allows members to try
different aspects of Lean
– Metrics & Goals
Lean Champions

Roles & Responsibilities


• Deploy the Lean policy via
training, implementation and
Champions Kaizen Blitz
• Feedback information to the
Design Team on progress
• Be given the time to support
• Members usually
the Lean efforts
from management or
“Value Adders” • Make presentations and
communicate the results of
• Motivated to learn,
the Lean projects
lead and improve
their organization
Champion’s Agenda
• Be ready to commit WIIFM
time to Lean projects • Gain new skills
• Be willing to learn and • Valued by the
continually improve organization
• Become a Lean • Exciting, new
content expert assignments
• Have skills in training, • Learn other aspects of
public speaking, Lean
project management
and be a team player
Multi-facility Deployment Example
Organization
Lean Steering
Committee
Small Facility Large Facility
Lean Steering Lean Steering
Committee Committee

Design
Design Team/ Design Design Design
Team
Champions Team Team Team

Champions
Momentum

• Have to build a
“critical mass” of
employees trained in
Lean and apply
principles
• Build “buy-in” and
get people onboard
• “Bandwagon” affect

Organizational Alignment
Dealing with Objections to Lean

• Put yourself in their shoes

• Help answer “WIIFM”

• Communicate, communicate, communicate


and then communicate some more!

• Create an “Elevator Speech”


Getting People on Board

! ?

Status Quo
Land
Lean Enterprise vs.
Lean Manufacturing

• Taking Lean beyond the shop floor


Lean Enterprise
• Move from the shop floor to Enterprise-wide Lean
implementation
• Many of the Building Blocks are essential for
efficient office functions – 5S, Visual, POUS,
Standard Work, Layout, Self Inspection, Poka-
yoke
• The goal is to reduce or eliminate the wastes to
reduce lead times and to enhance
responsiveness, competitiveness and customer
satisfaction
Thank you!

www.asq.org

You might also like