Professional Documents
Culture Documents
06-Lssbb-Pg1a Vol 1 (Mod. 1-13) (Ingles)
06-Lssbb-Pg1a Vol 1 (Mod. 1-13) (Ingles)
BLACK BELT*
(Volume 1)
Participant
ParticipantGuide
Guide
Software: SPC XL
DOE KISS
DOE PRO
SimWare PRO
Copyright © 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005
06-LSSBB-PG1A
NAME:
Job:
Your Phone / E-mail:
NAME OF SPONSOR:
Phone / E-mail of Champion/Sponsor:
NAME OF PROJECT:
Brief Project Description (25 Words or Less):
1.
2.
3.
Goals
Outcomes
Course Guidelines
Introductions
This training will provide each participant with the ability to:
• Breaks
• Active participation
INTRODUCTIONS
• Name
• Expectations
– IMPORTANCE OF KNOWLEDGE
• Role of Questions
APPROACH
Providing a
BETTER product or service,
FASTER, and
at a LOWER COST
than our competition
Knowledge
• BUSINESS RESULTS
– Generate business growth and improve profitability
– Achieve total customer satisfaction resulting in
increased market share and improved profit margin
• Remove wasteful/non-value added activities
• Decrease defects, cycle time, and inventory;
increase first pass yields and improve quality
• Expand knowledge of products and processes
through characterization and optimization
• BREAKTHROUGH TECHNOLOGIES
– Develop leaders in breakthrough technologies to
meet stretch goals associated with Better, Faster,
Lower Cost Products and Services
– Provide a common set of tools and techniques (a
common methodology)
Lower Upper
Spec Spec
Limit Limit
(LSL) (USL)
S ix
Six
ean
LLean S igm
Sigma a
Standard
Standard Model
Model Jidoka
Jidoka
Interchangeable
Interchangeable Costing
Costing Variety
Variety S.
S.Toyoda
Toyoda
Parts
Parts
Eli
EliWhitney
Whitney
Mass Just
Just––In
In––Time
Mass/ /Batch
Batch Time
K. Toyoda
K. Toyoda
Alfred
AlfredP.
P.Sloan
Sloan
Assembly
Assembly
Line
Line Toyota
Mass Toyota
MassProduction
Production Production
Waste Production
Waste System
Henry System
Elimination
Elimination HenryFord
Ford T.
T.Ohno
Ohno
System
System
Supermarket
Supermarket Lean
Synchronization Systems
Systems Employee
Synchronization Employee Six Sigma
Partnership
Partnership
Total
TotalQuality
Quality Drucker
Drucker
Time E.
E.Deming,
Time&&Motion
Motion Deming,
Division et
etal
DivisionofofLabor
Labor al
F.
F.Taylor
Taylor Six
SixSigma
Sigma
SQC
SQC Motorola
Motorola
Shewhart
Shewhart
Western
WesternElectric
Electric
DOE
DOE
Taguchi
Taguchiet
etal
al
Customer
Better, Faster, Lower Cost
Capability Culture
Tools and Values,
Methodology Leadership,
The Way We Work
Zero
Defects
TQM
CPI
Quality
Circle
Re-
Engi-
neering
• Executive Ownership
• Motivation
• Financial Accountability
• Supply Chain
PROJECTS
• Master Black Belts (Masters)
• Team Members
• Schedule training
1 - 14
(ROT) employees
CHARACTERISTICS OF
LEAN SIX SIGMA PROFESSIONALS
(Champions, Masters, Black Belts, and Green Belts)
• Vital Role
– Learn and accelerate use of tools and methods
– Focus on knowledge
– Linked to business goals, bottom-line improvement,
and customer satisfaction
Define
Pr
oj thi
ec s
t s wa
uc y
ce
ss
…
Measure
I mprove
Control
Measure success. Plan for
holding the gains. Monitor
critical outputs to prevent
repeat problems.
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 1 - 18
DEFINE
Define
Control
Build Knowledge /Project Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go
Yes
MEASURE
Close Down Project
Measure
Plan for data collection; Establish or
refine metrics if necessary
Update Knowledge/Project
Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
ANALYZE Close Down Project
Measure
MEASUREMENT METHOD MACHINE
Carburetor Transmission
Gage for
Ma
nu
Tire Press.
Tire Press.
to
al
Estimate of
Full Tank Maint. Sched. Tune-Ups Weight Tire Type
Ra
Re
gu
WX
Engine Warm-
Type Fast Start Up Time Radio
Air Conditioner
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Ice
Oil Type
Passengers Snow
Terrain
Pr
A
Rain
Hu
ec
Te
ipi
mi
s
Wi
mp
ain
tat
dit
e
t
Fla
nd
tud
era
ion
y
unt
Driver
tur
Alti
Mo
Fuel Additive
Acctng
Identify the non-value added activities,
HR
Legal
sources of waste, and defects and/or Analyze
Functional
Area
Benefits
problems
Frequency or
Financial Measure
Control
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
IMPROVE Close Down Project
Copyright Air Academy Associates 1 - 21
IMPROVE
Define
170
150
140
130
120
110
100
90
80
validate improvement
1
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
Month Number
Analyze
Control
Update Knowledge/Project Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
CONTROL Close Down Project
CENTERLINE
Monitor performance and measure
success
LCL
Measure
Performance No
confirmed and Gap Analysis
project goals
achieved? Analyze
Yes
ZERO
DEFECTS ISO 9000
TQM
BALDRIGE
CRITERIA LEAN
QFD
Which steps
have you Job Security
taken?
Long Term Success
Return On Investment
(Improved Bottom Line)
Quality Improvement
(Better, Faster, Lower Cost)
Knowledge
PROFOUND KNOWLEDGE
1. Which value stream are you supporting and who is the recipient of the value, i.e.,
who is the customer? Who is the value stream owner and who are the players or
DEFINE
process is improving? Why did we choose those? How accurate and precise is
the measurement system? Show me the data.
5. What are the customer-driven requirements or specifications for all of the
performance measures? Are the process performance measures in control and
how capable is the process? Show me the data. What are the improvement
goals for the value stream or process performance measures?
6. What kinds of waste and cost of poor quality exist in the value stream or process
and what is the financial and/or customer impact? Show me the data.
7. What are all the sources of variability in the value stream or process and which of
those do we control? How do we control them and what is our method of
documenting and maintaining this control? Show me the data.
ANALYZE
8. Are any sources of waste or variability supplier-dependent? If so, what are they,
who are the suppliers, and how are we working together to eliminate waste and
variability? Show me the data.
9. What are the key input variables that affect the average and standard deviation of
the measures of performance? How do you know this? Show me the data.
10. What are the relationships between the measures of performance and the key
input variables? Do any of the key input variables interact? How do you know
for sure? Show me the data.
IMPROVE
11. What settings or values for the key input variables will optimize the measures of
performance? How do you know this? Show me the data.
12. For the optimal settings of the key input variables, what kind of variability still
exists in the performance measures? How do you know? Show me the data.
13. Have we implemented a process flow and control system to sustain the gains
CONTROL
Which steps
have you Job Security
taken? Long Term Success
Return On Investment
(Improved Bottom Line)
Quality Improvement
(Better, Faster, Lower Cost)
Knowledge
Questions Questions
Managers Managers
Need to Need to
Knowledge Answer Ask
KBM
Based
Management
(KBM)
Tools and Techniques
to Answer Questions
and Improve the
Scorecard
Knowledge Gained
KISS
LOW HIGH
Statistical Complexity
give you a chance to practice that tool in class, then have you
apply that tool to your project, and finally have you review the
– KEY TERMINOLOGY
• TYPES OF DATA
• DISTRIBUTION, MEAN, STANDARD DEVIATION
• MEASURES OF QUALITY (Cp, Cpk, FPY, RTY, sigma
capability)
– PF/CE/CNX/SOP
• THE FIRST LINE OF DEFENSE AGAINST VARIATION
X Vision
X Benchmark
X Philosophy
5. Continuously __________________________
..
value |’val-(,)yu , n
1. : relative worth, utility, or importance; degree of excellence
2. : the monetary worth of something : marketable price
- Webster’s New Collegiate Dictionary
Suppliers
Suppliers Design
Design Procure
Procure Make
Make Sell
Sell Customers
Customers
Total Value
Stream
External External Internal Your Ware- Distributor Retailer Customer
Supplier Supplier Supplier Facility houses
X X
Eliminate Non-value Added Steps
INPUTS OUTPUTS
People
Equipment
PROCESS
VALUE STREAM
(also known as a SIPOC model)
Customer
om
e ry
Inc H isto
dit
Cr e
U/W accuracy
Broker/Correspondent
Application forms
r
pe -mail Fax
Pa E
Criteria
Procedures Underwriting
Regulations Process Cycle Time
Policies
Software
Underwriter
e /
i enc ing e w ny
n
er rai Tim mpa
p T
Ex co Cost / per
Property application
e n
ue
Ag tio
l
a
Va
c
Lo
Application
Under-
writing Approval
Fund
Service
Coding SOPs
Cost per Line of Code
Programming Language
Experience of Programmer
Software Time to Code (Lines per day)
Coding Process
Design Documentation Errors Detected in
Internal Testing
Module Length
Time to Fix Errors
Module Complexity
Str of P
Pro cisio
Da Va
I/O
uc ath
ta riab
C
c ed n
tur s)
(# les
ur a
al
of
l
Analysis
Design
Phase
Phase Coding
Phase Testing
Phase Operational
Phase
PROCESS
Describe the
process Keep it
short, using 2-10
words as a rule of
thumb. A blending of
inputs to
achieve the
desired outputs
Purchasing Payroll
Accounts Receivable Help Desk
Accounts Payable Shipping
Hiring Scheduling
Training Web Site Creating
Documentation Management of Lean Six Sigma
Advertising Database Management
Drug Testing Personnel Placement
OUTPUTS
Perform a service
Describe the PROCESS
measurable outcomes
of the process. These
should be measures
related to process Produce a product
performance, speed, A blending of
cost, etc. inputs to
achieve the
desired outputs Complete a task
INPUTS
People
Material PROCESS
Describe the process
inputs. These are the
Equipment sources of variation
that impact the output.
Policies A blending of
inputs to
Procedures achieve the
desired outputs
Methods
The inputs to a process are items that affect the outcome of the
process. You may or may not have control over these items.
Satellite View
People
Market Share
Technology
Corporate Profit
Policies
Aircraft View
In the Trenches
People
Methods Cost
Materials
Process Variables Performance
My Process
Environment
Time
Measurement
Errors in Hiring
# of people
Recruitment Method
# of Errors
Payment method
Time to Hire
Job Descriptions Hiring
Payroll
Tax Tables Process Deposit Dates Process
Cost per Hire
Benefit Plans
Hourly Rate
Tax Amount
Orientation Cost
Vacation Policy Background Checks
(HR) (HR)
2 - 14
Purchase Order (PO)
Cycle Time Type of Raw Material
Payment Policy
Mold Temperature
Sole Source vs. Bid FPY Thickness of Molded Part
Holding Pressure
Bid Evaluation Purchasing PO Cost Manufacturing
Process Holding Time Injection
Molded Parts % Shrinkage from Mold Size
Negotiations Time Until Delivery Process
Gate Size
Coding SOPs
2 - 15
Monitor Size
Mouse Type
Feed Rate # of Sorted Errors # Pages Per Day
Type Keyboard
Automatic Desktop
Mail Sorter Years Experience
Speed Cycle Time Publishing
Process Process Defects Per Page
(R&D)
Computer Speed
Type Computer
(Design)
Automated or
Manual Type of Wire Chip Inductor
Accounts
Design
Receivable % defaulted Inductance
Process
Wire Insulation Level
Net Policy
(Accounting) (Design)
2 - 16
Type of Training
Instructional Materials # of Calls
# of Students Ratio of Abandoned
Like (Course Evaluation) Type of Call Calls
Site of Training
# of Operators
Accountability of Training
Student & Trainer Process Learn (Pre vs Post Test) Help Desk Time per Call
Operator Call Center
Student Background Kn Tr Ex
o w
ai
ni pe
Amount of Time for le ng rie
Use (# of Applications) dg nc
Training e e Cost per Call
of
Pr
Kind of Trainers od
(Finance) (Shipping)
2 - 17
EXERCISE
Xs Ys
Y=f(X)
• Pronounced “Y is a function of X”
• Variables Data
– Also called continuous or measurement data
– Actual measured values are recorded
– Data can assume a range of values on a continuous
scale
– Examples:
• Dimension in thousandths of an inch
• Distance in feet
• Operating temperature in degrees Fahrenheit or Celsius
• Attribute Data
– No direct measurement is made
– The presence or absence of something is recorded
– We classify or count a feature or characteristic of
the product or process
– Examples:
• Number of nonconforming articles
• Invoice errors per week
• Test results (pass/fail, good/bad)
• Presence or Absence of a date code stamp
y (measure of performance)
130 140 150 160 170
∑y i
Center of Process: y= i=1
n
∑ ( y − y)
2
i
Process Variance: s =
2 i =1
n −1
-6σ -5σ -4σ -3σ -2σ -1σ 0 +1σ +2σ +3σ +4σ +5σ +6σ
68.27%
95.45%
99.73%
99.9937%
99.999943%
99.9999998%
20 22 24 26 28 30 32 34 36 38 40 42 44
Any bills being paid more than 40 days out or less than 30
days would be classified as an error or defect in the billing
paying process.
3σ Process Centered
• Process is WIDER
than the
Example 1: σlevel = 3
specifications,
causing waste and
LSL USL cost of poor quality
σ σ σ
Determined by Determined by
the customer the customer
-6σ -5σ -4σ -3σ -2σ -1σ 0 +1σ +2σ +3σ +4σ +5σ +6σ
WASTE WASTE
Example 2: σlevel = 6
6σ Process Centered
LSL USL
σσσσ σ σ • Process FITS well
within the
specifications, so
even if the process
shifts, the values fall
well within tolerances
-6σ -5σ -4σ -3σ -2σ -1σ 0 +1σ +2σ+3σ +4σ+5σ+6σ
Notes:
Yield is the probability that whatever we are producing (manufactured part, PO,
paperwork, etc.) will pass through the entire process without rework and without
defects.
Cp = USL − LSL
6σ
Cp PPM
15.0
7.0
4.5 << 1PPB
3.0
2.0 .0018
1.67 < 1PPM
1.33 55
1.0 2,700
.9 6,900
.8 16,400
.67 45,500
.5 133,600
LCL LSL TARGET USL UCL
⎛ USL − y y − LSL ⎞⎟
Cpk = min ⎜⎜ or
⎝ 3σ 3σ ⎟⎠
2 - 30
PRIOR TO COLLECTING ANY DATA
Output(s)
and How What Who
- - -
Specs - - -
- - -
- - -
- - -
- - -
- - -
C= Controlled (Constant) SOPs - - -
- - -
- - -
N= Noise - - -
X= Experimental
2 - 32
PROCESS FLOW DIAGRAM
Approve/Disapprove
Accept/Reject
Decision Point
Yes/No
Pass/Fail
A
Connector (to
another page or B
part of the diagram)
Represents
direction of flow
Patientcalls
Patient calls No Patienttold
Patient toldtoto
hospital IsIsan
anopen
open
holdfor
for1st
1st
hospital lineavailable?
available? hold
mainnumber
number line availableoperator
operator
main available
Yes
Patientdescribes
Patient describes
typeofofcare
type care
required
required
Operator
Operator
transferspatient
transfers patient
totoappropriate
appropriateoffice
office
Emergency
Emergency Family
Family AcuteMinor
Acute Minor
Room
Room Optometry
Optometry
Practice
Practice IllnessClinic
Illness Clinic
(E.R.)
(E.R.)
E.R.personnel
E.R. personnel Appointment
Appointment Appointment
Appointment Appointment
Appointment
consultpatient
consult patient made
made made
made made
made
Pass
Example Web
Page Creation
Management
Fail Rework
Review
Pass
Pass
Pass
Complete
STAFFING Hiring Manager Timely Requisition/Approval Staffing Q1: % of Late Filled & Open Positions
Post Job
Develop Hiring Strategy
Conduct Ext. No
Recruiting?
Yes
Submit
Conduct Ext. Recruiting Collect Candidate Responses Resume
Screen Paperwork
No Any Potential
Candidates?
Yes
Forward to Hiring Mgr.
Screen Paperwork
No Any Potential
Interviewees?
Yes
INTERVIEWING
No Found Final
Candidate?
Yes
Accept
Extend Offer Offer?
Yes
Initiate New Hire Process
CAUSES EFFECT
(Sources of Variation or Variables) (Response or Output Variable)
RESPONSE
C = Constant
• To hold a variable as constant as possible requires
controlling the variable via Mistake Proofing and SOPs
to eliminate errors and reduce variation.
• Controlling a variable or holding it as constant as
possible doesn't just happen. It must be engineered into
the process.
• Mistake Proofing: The process of eliminating conditions
(errors) that lead to variation in the CTCs and ultimately
cause defects.
N = Noise
• Noise variables are those that are not being controlled or
held as constant as possible.
• Mistake Proofing is needed to change an "N" variable to
a "C" variable.
X = Experimental
• These are key variables that can be controlled and held
constant at different levels or settings for the purpose of
determining the effect of this variable on the CTC.
Carburetor Transmission
M
Gage for Au an
Ride Brakes Tire Rotation t o
ua
Tire Press. l
Tire Press.
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Ice
Oil Type Pr
Passengers H ec Snow
Terrain
A
Te
B
m
um ip Rain
W id ita
s
in pe it y tio
d
e
ra
at
n
a in
tu
Fl
ud
nt
Driver re
u
l tit
A
Mo
Fuel Additive
PEOPLE
PEOPLE MATERIALS
MATERIALS ENVIRONMENT
ENVIRONMENT
2 - 40
Convenience Service Level of Appreciation
Did not Difficult to use Too many Keep asking
consolidate mistakes Don’t for the same
Discourteous
accounts understand information
or impersonal
Delays my needs
* As viewed by customer
2 - 41
LINKING PF/CE/CNX AND IPO TOGETHER
Process Flow
Step 1
(PF)
Step 2
No
Step 3 Step 4
Yes
Step 5
Step 6
Generating
Variables Associated with this Process
can be Done Using a Cause-and-Effect Diagram with
Different Sets of Variable Categories.
Two examples are shown here.
CE/CNX CE/CNX
Using Traditional 6 M's Using Process Steps
Materials Method Machine Step 1 Step 2 Step 3
X1 X3 N1 X1
C1 C2 N2
X2 C1
y y
N1 C2 C3
N3 C3 N3
N2 X2 X3
Manpower Measurement Mother Nature Step 4 Step 5 Step 6
C1 C2 C3
IPO
x1
x2 y
x3
2 - 44
LEAN SIX SIGMA PHILOSOPHY:
Appraisal Costs
Prevention Costs
COPQ (% of Sales)
40
35
30
25
20
15
10
5
2 3 4 5 6
Sigma Rating
Before After
PF/CE PF/CE
(from p. 2-30) (from p. 2-44)
COST COMPUTATION
Cost of Inspection (Time + Manpower)
= N • $5.00 =
Cost of Defect Disposition (Time + Manpower)
= D • $10.00 =
Cost of Rework (Time + Manpower + Material)
= R • $50.00 =
Cost of Scrap (Time + Manpower + Material
+ Removal)
= S • $100.00 =
Total Cost
4 units
rework
1 unit
scrap
• Example:
LS US LS US
Desired
Desired
... End State
EndState
Test
Test Test
Test
Analyze
Analyze Analyze
Analyze
* *
Fix
Fix Fix
Fix
Every Time a Defect is Created During a Process (Step), it Takes Additional Cycle Time to Test,
Analyze, and Fix.
2 - 52
* These Non-Value Added Activities Typically Require Additional Floor Space, Capital Equipment,
Material, and People.
EXERCISE: ILLUSTRATING FIRST PASS YIELD
(Rolled Throughput Yield) (Round 1)
Directions:
1. Each team will be given a deck of cards (52).
2. Each team will have 2 operators, each of whom will drop one
card at a time onto a target area.
Start Target Target Finished
Product to
Supplier Area # 1 Area # 2 Customer
Station 1 Station 2
3. The method of drop will be to hold the card at arms length over
the target using two fingers on edge (as demonstrated by the
instructor).
4. After release, the card will fall and either be completely in the
target area or not. Only those cards that fall completely within
the target area may move to the next station.
5. Objective: to deliver 20 completed products or units to
the customer.
6. Team Personnel: 2 Operators
2 Inspectors
1 Material Supplier
1 Timer
1 Foreman
7. Data To Track:
A. Total number of good units moved on to the next station (or
customer). Inspector at each station should record this.
B. Total number of drops at each station. Inspector should
keep track of this, or a separate auditor could perform this
function.
C. Total time from first drop to completion of customer order
(20 good units).
D. Total number of cards of the 52 used in the exercise (52 -
whatever supplier has left over).
A1 = B1 =
A2 = B2 =
Y2 = A2/B2 =
Total Cost
Average Cost Per Unit produced = =
20
C
Average Cycle Time = =
20
Directions:
1. As a team, brainstorm opportunities for improvement.
2. Have a tollgate review with your customer (instructor) to discuss
your improvement ideas and get approval.
3. Train the team on the new process and repeat the exercise at
the instructor’s signal.
4. Track the same data as in round 1 and report your first pass
yield, average cycle time per unit produced, and average cost
per unit produced.
Constraints:
1. Cards must still be dropped one at a time.
2. You cannot eliminate one of the steps in the process (i.e., must
still use two target areas)
3. You may not damage the cards. (no bending, tearing, etc.)
4. You must still hold the card at arms length over the target.
A1 = B1 =
A2 = B2 =
Y2 = A2/B2 =
Total Cost
Average Cost Per Unit produced = =
20
C
Average Cycle Time = =
20
*
Yield is defined here as the percentage of all products, services or transactions
in which all of the parts are defect-free or all steps are performed perfectly, that
is, without error.
Copyright Air Academy Associates 2 - 57
WHY LEAN “SIX” SIGMA?
Motorola
Defects / M
100K
DEFECT REDUCTION
66810
10K
Average 6210
Company
1K
233
100 >> 100x
10
“Best in Class” 3.4
1
2 3 4 5 6 7
Sigma Capability
Process Flow Diagram (Sec. 2.2.1) Cause & Effect Diagram with CNX
(Sec. 2.2.2)
Input
Input
Manpower
Manpower Materials
Materials
No
Decision?
Decision? Causes or
Sources of
Variation Effect or
Effect or
Yes Response
Response
ProcessStep
Process Step
Output
Output
Methods
Methods Machines
Machines
LSL USL
y
2 4 6 8 10 12 14 16 18
Brown Red Yellow Tan Green
• a bar chart for non-numerical categorical • a bar chart for numerical categories; bars
descriptors whose bars are in descending need not be in descending order
order
• shape provides form of distribution of data
• used to identify and separate the most
• central tendency and variability are easily
frequently occurring categories from the
estimated
less important categories
• spec limits can be superimposed to
• may be used in a stratified or nested
estimate process capability
fashion to zero in on precise problem area
Absenteeism
S = Severity of Failure
1: Low 3: Moderate 5: High
O = Probability of Occurrence
1: Low 3: Moderate 5: High
D = Probability of Escaped Detection Time
1: Low 3: Moderate 5: High
R = S•O•D = Risk estimate
1: Low 25: Moderate 125: High
• used to identify and evaluate potential • plot of data over time
failure modes in order to prioritize and • used to record and display trends in data
“fool proof” the critical ones over time
• performed on both products and • detects meaningful changes in a process
processes
UCL
CENTERLINE
LCL
Vehicle Weight
• used to identify when special causes are • shows the relationship between two
affecting the process average or variables (positive, negative, none)
variation
• provides visual estimate of correlation
• separates natural variation from special coefficient (strength of a linear
causes of variation relationship)
• monitors process performance over time • allows “eyeball” fit of the data
• helps reduce process variability • identifies outlying points (outliers)
• helps determine what kind of action
should be taken
Run A B AB y1 y2 y3 ... y s
y
1 - - +
2 - + -
20 ŷ = 18 − .65x
3 + - -
15 4 + + +
10
ΔA Δ Δ
ŷ = y + A + B B + AB A ⋅ B
5 2 2 2
δA δ δ
5 10 15 20
x ŝ = s +A + B B + AB A ⋅ B
2 2 2
• systematizes an efficient way to collect
data
• develops a model relating a response
variable to input variable(s) • used to identify important factors affecting
the process
• allows prediction of responses at levels
other than where data was collected • identifies important interactions between
process variables
• quantifies the strength of the model
• can be used to optimize a process
• identifies outliers
• used to build a model relating how inputs
affect the outputs
Gage Capability Study (Sec. 9.13) Quality Function Deployment (Chap. 11)
UCL
LCL
Be Wiser
An Incomplete Scorecard
– Operational Cost
– Advertising Cost
– Stock price – Milestones
– Materials Cost
Organization – Market share – Timelines
– Taxes
– Organizational Climate – Reaction Time
– Budgets
– Profits
– Development Cost
– Meeting Customer Needs – Design Cost
– Development Time
– Customer Complaints – Product Price
Product – Design Time
– Reliability – Lifecycle Cost
– Design Changes
– Availability – Cost of Poor Quality
– Profit Margin
– Turnover Rate
– Product Rate
– Training Cost
– Defect Rate – Training Time
– Legal Costs
People – Safety Violation Rate – Rehabilitation Time
– Absentee Costs
– Absenteeism – Learning Curve Time
– Ergonomic Costs
– Suggestion Rate
– Benefit Costs
A Balanced Scorecard
Vince Lombardi
Copyright Air Academy Associates 2 - 63
SUMMARIZING THE POWER OF
LEAN SIX SIGMA
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 3-2
DEFINE
Define
Control
Build Knowledge /Project Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go
Yes
MEASURE
Close Down Project
• Steps:
– List project opportunities
– Determine project selection criteria and importance
rating (on a scale from 1(low) to 10 (high))
– Rate the relationship between each project and
each criteria (0 = none, 1 = low, 3 = med, 9 = high)
– Calculate weighted sum by cross multiplication
• Example:
Example Project Selection Criteria
(Importance)
Doable in High Apply
2-3 Likelihood across
High ROI months of Success business
(9) ( 10 ) (8) (7)
Project #1 9 1 3 1 122
Project #2 9 9 3 3 216
Project #3 3 9 9 1 196
Project #4 1 3 1 9 110
Total: 644
Pareto Chart
250
200
Weighted Sum
150
100
50
0
Project #2 Project #3 Project #1 Project #4
– Saturn:
• moderate ROI (about $100,000)
• moderate likelihood of success (about 60%)
• moderate linkage to strategic business goals (1 of 5)
• quick to complete (< 3 months)
• lots of available resources
– Neptune:
• high ROI ($250,000)
• moderate likelihood of success (about 60%)
• high linkage to strategic business goals (3 of 5)
• moderately quick to complete (about 3-4 months)
• low availability of resources
– Jupiter:
• low ROI (<$50,000)
• high likelihood of success (>95%)
• no linkage to key, strategic business initiatives
• long to complete (>6 months)
• lots of available resources
Total:
• Example:
– “We have huge differences between our
forecasted and actual demands for software,
resulting in writeoffs of $1 million per quarter
in software worldwide. This is causing us to
miss our revenue projections for the quarter
and impacts profitability.”
Specific
Measurable
Achievable
Relevant
Timely
“in scope”
“out of scope”
• Examples
– Cost Takeout / Cost Avoidance
• reduction in scrap
• reduction in raw materials
• reduction in labor or planned headcount
• freight savings
• interest expense reduction
• advertising and agency fee reduction
• warehouse space reduction
– Cash Flow
• reduction in inventory levels
• inventory carrying cost reduction
• reduction of past due receivables
– Growth
• increase in capacity/throughput
• increase in market share / sales revenue
• Individually (5 minutes)
– How will you measure the financial impact of your
project? Document some ideas that you have right
now for potential hard and/or soft savings.
• Team Leader
• Facilitates team meetings
• Keeps Sponsor informed on team progress and
resource needs
• Purpose
• Typical Activities
– Validate the project charter
– Validate the problem statement
– Validate the project goals
– Establish team ground rules and code of
conduct
Level of Acceptance
Key Issues /
Dept. 1 2 3 Strategy
Stakeholder Concerns
blocker neutral supporter
• For each task, estimate the time duration and assign resources
• Example:
Team Resources
Assess the
1.0 measurement A
system capability
30
1.1 Brief the team I A I I
minutes
Select items to be 30
1.2 R A D
included in the study minutes
Determine the 15
1.3 I A
sample size minutes
Take data
1.4 2 days R I A
(measurements)
Determine actions
1.6 0.5 days I A, S
to be taken
S = sign-off
R = review
D = data
I = input
A = accountable
DAY #
ID TASK 1 2 3 4 5 6 7
Select items to be
1.2
included in the study
Determine the sample
1.3
size
Take data
1.4
(measurements)
Determine actions to be
1.6
taken
. .
. .
. .
• As a team,
4. “It is not the strongest of the species that survive, nor the
most intelligent, but the ones most responsive to change.”
11. “But of the good leader, when his work is done, his aim
fulfilled, the people will say “We did this ourselves.”
B. Henry Ford
• U.S. automobile industrialist
C. Arnold Bennett
• Novelist
D. Albert Einstein
• Scientist and one of the greatest thinkers of the 20th century
F. Calvin
• From the Calvin and Hobbes cartoon strip, by Bill Watterson
G. Grace Hopper
• Pioneer computer scientist
• First woman to receive the prestigious National Medal of Technology in
1991.
H. Robert F. Kennedy
• United States Senator
• Brother of John F. Kennedy; assassinated in 1968
I. The Beatles
• Famous British Rock and Roll band
J. Charles Darwin
• British naturalist
• Known for his theory of evolutionary selection
Emotional Level
(Intensity)
Transition Curve
High
• Acceptance
• Denial
• Anger
• Exploration
• Bargaining
• Depression
Low
Time
DD + VF + FS > R *
DD = ___________________________
VF = ___________________________
FS = ___________________________
R = ___________________________
* Change Model used with permission from Kelly and Frank Petrock.
1. WIIFM?
2. Organization’s Status?
a. Company in Crisis
b. Market Leader
c. Middle-of-the-Pack
UNDERSTANDING THE
VOICE OF THE CUSTOMER (VOC) AND
DEFINING A PROCESS
goals
business
value
customer
UNDERSTANDING THE VOICE OF THE
CUSTOMER (VOC) AND DEFINING A
PROCESS
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 4-2
DEFINE
Define
Control
Build Knowledge /Project Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go
Yes
MEASURE
Close Down Project
+ -
Value
Added
OUTCOME
Satis-
factory
Fails
PROCESS
Exciting Quality
(Delighters) One-Dimensional
Quality
Expected Quality
(Must Be)
Very Unsatisfied
Expected Quality
• Not specifically requested, but assumed to be present
NEEDS • If present, customer neither satisfied nor dissatisfied
• If absent, customer very dissatisfied
One-dimensional Quality
• Specifically requested items; specifications
• Items from a typical survey (stated wants)
WANTS • If present, customer satisfied
• If absent, customer dissatisfied
Exciting Quality
• Unknown to the customer
• Most difficult to define and develop
DELIGHTERS
• If present, customer very satisfied
• If absent, customer neither satisfied or dissatisfied
• Market studies
• Customer research
• Sales calls
• Customer observation
• Customer complaints
• “I don’t like to wait on hold a • hold time (goal: less than 1 minute)
Customer Service
long time.”
• “I hate getting bounced • % of customers routed to the correct
around to multiple people.” person the first time (goal = 100%)
• “I want to have good • % of customers surveyed replying
customer service.” “satisfied” or better (goal 99%)
• “The reps should be • % of customers greeted by name; %
friendly.” of customers not interrupted when
talking (goal 100%)
Functional Req.
(CTC's)
Customer Req.
Customer
Expectations
House of
Quality
Prod. Design
#1 Characteristics
Functional Req.
House of
(CTC's)
Quality
Mfg. Process
#2 Characteristics
Performance
Characteristics
Prod. Design
CTC's House of
Quality
Mfg. Process
#3 Control
Product
Characteristics
Mfg. Process
Mfg. Process
Control
HOW
What can we measure or control to ensure
customer requirements are met? (These are
the CTCs, or critical to customer performance
measures)
Relationships
IMPORTANCE
WHAT
RATING
PRIORITY / IMPACT
What do we need to focus on?
This is a weighted sum.
Ease of Use
Importance
Setup Time
Web Filter
Customer Requirements
B: Quick Logon 3 9 1 3
C: Fast Downloads 4 1 1 9 3 1 3
Rank-Ordered Sum 9 48 56 62 28 45
Customer Requirements
A: Fast Delivery 5
C: Correct order 5
Rank-Ordered Sum
# of Calls
# of Operators
Help Desk Time per Call
Operator
Call Center
Ex
Tr
Kn
ai
pe
ow
ni
nc
dg
e
e
of
Pr
Resolution
t
Information on Failures
(Service)
S I P O C
Suppliers Inputs Process Outputs Customers
Materials, Any activity Measurable
methods, that takes outcomes of
people, etc. inputs and the process
required for creates
the process outputs
Who provides Who receives
the inputs to the outputs
your process? of the
process?
4. 3. 2. 1.
Suppliers: Inputs: Process: Outputs: Customers:
Telephone
Ordering for
Take-Out
Pizza
(from receipt of call to
delivery of order to
cook)
– NUMERICAL MEASURES
Measures of Location (mean, median)
Measures of Dispersion (range, standard deviation,
variance)
Measures of Quality for Variables Data
– Cp
– Cpk
– Dpm
– Sigma level
– Sigma capability
Measures of Quality for Attribute Data
– FPY
– dpu
– dpmo
– Sigma capability
Measuring Correlation
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 5-2
MEASURE
Define
Understand the process and create
detailed process maps
Measure
Plan for data collection; Establish or
refine metrics if necessary
Update Knowledge/Project
Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
ANALYZE Close Down Project
Approve/Disapprove
Accept/Reject
Decision Point
Yes/No
Pass/Fail
A
Connector (to
another page or B
part of the diagram)
Represents
direction of flow
• High Level
no
yes
Unload Determine
and Scan Bag Method of
yes groceries Groceries Payment
Proceed Want to use
to Check self-service Put
Pay for items in
Out lane? Proceed Unload
Determine groceries car
no to Check items; wait
Method of
Out Lane for clerk to
Payment
scan/bag
Pass
Example Web
Page Creation
Management
Fail Rework
Review
Pass
Pass
Pass
STAFFING Hiring Manager Timely Requisition/Approval Staffing Q1: % of Late Filled & Open Positions
Post Job
Develop Hiring Strategy
Conduct Ext. No
Recruiting?
Yes
Submit
Conduct Ext. Recruiting Collect Candidate Responses Resume
Screen Paperwork
No Any Potential
Candidates?
Yes
Forward to Hiring Mgr.
Screen Paperwork
No Any Potential
Interviewees?
Yes
INTERVIEWING
No Found Final
Candidate?
Yes
Accept
OFFER / START
cycle time = 35 sec. cycle time = 15 sec. cycle time = 5 sec. cycle time = 3 sec. cycle time = 13 sec.
material availability dpm = 5,000 dpm = 10,000 yield = 99% 15% out of spec
= 85% accident rate =
throughput = 100/hr 1/1,000
Warehouse
Purchasing
Operations
Department
12 6 5
8
Purchasing WIP Purchasing WIP
Admin. Queue Director Queue
7
Non-value Added
Total active
time: 10%
Value
Adding
Non-Value
Adding Queue
118 minutes
value add
a measure of
process efficiency is
(value added time) / (total
lead time)
Activity
Time
Define
Pr
oj thi
ec s
t s wa
uc y
ce
ss
…
Measure
I mprove
Control
Measure success. Plan for
holding the gains. Monitor
critical outputs to prevent
repeat problems.
Statistical Questions;
Tools Critical
Thinking
Sampling plan
How?
Metric Formula Who (time period, Accuracy
Definition (collection
(Data) (if appl.) Responsible? sample size, of data?
method)
source, etc.)
– random sampling
– systematic sampling
5 - 21
MEASURING PROCESS PERFORMANCE
Type of Summary
LSL USL
• Cp and / or Cpk
Nor mal Distr ibution
Mean = 20.8 Histogram
Std Dev = 3.3966
KS Test p-value = .5030
8
5
# Obs er va tions
0
15.0 t o <= 17. 4 to <= 19.8 to <= 22.2 t o <= 24. 6 t o <=
17 .4 19.8 22. 2 24.6 27.0
5.52
8.73
9.8
10.9
15.2
16.2
17.3
20.5
21.6
22.7
23.7
26.9
29.1
28
6.59
7.66
11.9
13
14.1
18.4
19.4
24.8
25.9
30.2
31.2
32 3
Class
5000
4000
3000
2000
1000
0
1996 1997 1998 1999 2000 2001 2002 2003
Year
DATA
Types of Count Proportion
Shipping Errors (frequency)
Address 11 .11
Ship Early 9 .09
Ship Late 37 .37
Method 6 .06
Contents 8 .08
Packing 29 .29
100 1.00
Types of Errors and Frequency of Occurrence
40
35
30
Frequency
25
20
15
10
0
Ship Late Packing Address Ship Early Contents Method
Options window
for chart details
Chart 80
70
0.8
0.7
60 0.6
Count
1600 100%
1400
Monetary Loss ($)
80%
1200
1000 60%
800
40%
600
400
20%
200
0 0%
Packing Ship Late Ship Early Address Contents Method
Friday, June 9
Tuesday, June 13
Code Code Code Code Code
440 410 440 465 841
413 412 412 413
181 465 218 440
Wednesday, June 14
Code Code Code Code Code
465 412 413 841 440
440 413 412 465 410
413 465
30
27 INFORMATION
25
19
20
16
15
12
10 9
8 8
0
410 413 465 412 440 841 218 181
June 9 June 12
16 16
14 14
12 12
10 9 10
8 8
6 6
4 4
3 3 4 3
4
2 2 2
1 1 1
2 2
0 0
410 413 465 412 218 841 413 465 218 410 440 841
June 13 June 14
16 16
14 14
12 12
10 10
8 8
6 6
3 4 3 3
4 2
2 2 2 2
1 2 1 1
2 1 1 1
0 0
440 413 465 412 410 841 218 181 413 465 412 440 410 841
KNOWLEDGE
• After analyzing the units that had failed with the
410 error code, it was discovered that the diskette
drive design allowed a manufacturing worker to
unknowingly damage the unit. As the workers
became more experienced at installing this unit
they made this mistake less frequently. This
reduced the number of 410 errors. Since the
product could still be damaged by an
inexperienced person, a simple design change
was made to prevent the damage from occurring.
• Individually
Brown =
Red =
Yellow =
Blue =
Orange =
Green =
Total
n = 50 DATA
18 16 30 29 28 21 17 41 8 17
32 26 16 24 27 17 17 33 19 18
31 27 23 38 33 14 13 26 11 28
21 19 25 22 17 12 21 21 25 26
23 20 22 19 21 14 45 15 24 34
INFORMATION
16
14
12
Number of Cars
10
0
6 to <12 12 to <18 18 to <24 24 to <30 30 to <36 36 to <42 42 to <=48
Gas Mileage (mpg)
Number of Classes: k = n =
20
18
16
Number of Cars
14
12
10
Options window
for chart details
Normal Distribution
Mean = 22.88
Std Dev = 7.7266 Histogram
KS Test p-value = .3731
16
14
12
# Observations
10
0
8.0 to <= 13.0 to <= 18.0 to <= 23.0 to <= 28.0 to <= 33.0 to <= 38.0 to <= 43.0 to <=
13.0 18.0 23.0 28.0 33.0 38.0 43.0 48.0
Class
Skewed distribution
Parabolic distribution
Exponential distribution
Bi-modal distribution
Third Quartile
Interquartile Range
+ Median
First Quartile
35
Q1
30
Min
25 Median
+ Max
20 Q3
15
10
• Individually
INFORMATION
350
300
250
Time (hrs)
200
150
100
50
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Chronological Order
40
NUMBER
30
OF CSUs
DESIGNED 20
10
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
PDR NOW PDR PDR(R1)
CONTRACT MONTHS
TOTAL CSUs = 50 PLANNED (REV 1) = 30
ACTUAL = 30 TOTAL REDESIGNED = 7
PLANNED = 33
Design Progress
20
16
ERRORS
KLOC 12
0.18
0.16
0.12
0.10
0.08
Proportion
0.06
0.04
0.02
1
3
5
7
9
0
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
49
51
53
55
57
59
61
Day
5 - 40
#
CREATING RUN CHARTS USING SPC XL
Run Chart
170
160
150
Data can be stored in
140
separate columns.
130
Y Axis
Highlighting 2 or more
120
columns will overlay the
110
run charts.
100
90
80
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
X Axis
Run Chart
170
160
150
Data can be stored in one
140
column. This may be a
130
Y Axis
90
80
10
13
16
19
22
25
28
31
34
37
40
43
46
1
X Axis
1.05
1.04
UCL
1.03
1.02
1.01
Average (oz)
Centerline
1
0.99
0.98
LCL
0.97
0.96
0.95
0.94
1 2 3 4 5 6 7 8 9 10 11 12
Sample Number
COMING ATTRACTIONS
254
253
252
251
250
y
249
248
247
246
244
COMING ATTRACTIONS
1 3000 18
2 2800 21
3 2100 32
4 2900 17
5 2400 31
6 3300 14
7 2700 21
8 3500 12
9 2500 23
10 3200 14
35
30
25
Mileage (MPG)
20
15
10
0
2100 2300 2500 2700 2900 3100 3300 3500
Vehicle Weight
35
y = -0.0152x + 63.507
30 2
R = 0.9191
25
20
mileage
15
10
0
2000 2200 2400 2600 2800 3000 3200 3400 3600
weight
Correlation Matrix
Correlation Coefficient = -0.958680615
1-Feb 17 0.022
2-Feb 22 0.027
3-Feb 29 0.025
4-Feb 14 0.02
5-Feb 68 0.053
6-Feb 52 0.043
7-Feb 19 0.024
8-Feb 44 0.044
9-Feb 37 0.034
10-Feb 47 0.038
11-Feb 82 0.057
12-Feb 25 0.029
13-Feb 27 0.022
14-Feb 48 0.041
15-Feb 35 0.042
500
400
Points Scored
300
200
100
0
0 10 20 30 40 50 60 70 80 90
Fouls
80
City Population (1000's)
70
60
50
100 200 300
Number of Storks
• Individually
Type of Summary
LSL USL
• Cp and / or Cpk
Nor mal Dis tr ibutio n
Mean = 20.8
Histogram
Std Dev = 3.3966
KS Test p-value = .5030
8
5
# Obser vations
0
15.0 t o <= 17. 4 to <= 19.8 to <= 22.2 t o <= 24. 6 t o <=
17.4 19.8 22. 2 24.6 27.0
5.52
8.73
9.8
10.9
15.2
16.2
17.3
20.5
21.6
22.7
23.7
26.9
29.1
28
6.59
7.66
11.9
13
14.1
18.4
19.4
24.8
25.9
30.2
31.2
32 3
Class
5000
4000
3000
2000
1000
0
1996 1997 1998 1999 2000 2001 2002 2003
Year
∑x i
x1 + x 2 + Κ + x n
x= i =1
=
n n
Sample Mean: x
8 11 6 7 12 7
x = (8 + 11 + 6 + 7 + 12 + 7) 6 = 51 6 = 8.5 hours
4 5 6 7 8 9 10 11 12 13 14
(hours)
The Mean as a Balance Point
236,000
36,000
32,000 40,000 98,000
28,000
Mean: 46,000
Assume that x1, x2, … xn, is a list of the sample data sorted in
ascending order. Then
Sample Median: ~
x
SEATWORK/HOMEWORK
X: Sample #1 (8, 3, 7, 4, 5, 6, 5, 4, 5, 6, 5)
Find x and ~
x
Mean = Median
Skewed Distribution
Median Mean
DON'T WORRY,
THAT ROPE IS
ONE INCH THICK
ON THE
AVERAGE.
x y z
30 10 10
30 30 20
30 30 30
30 30 40
30 50 50
x = 30 y = 30 z = 30
~
x = 30 ~
y = 30 ~
z = 30
Three Different Data Sets
Rx =
x
10 20 30 40 50
Ry =
y
10 20 30 40 50
Rz =
z
10 20 30 40 50
Data Sets with Same Center but Different Spreads
∑ (x − x )
n 2
i
2
s = i=1
n −1
Sample Variance: s2 = σ̂ 2(an estimate of σ2)
x (x − x ) (x − x ) 2
y (y − y ) (y − y ) 2
z (z − z) (z − z) 2
30 10 10
30 30 20
30 30 30
30 30 40
30 50 50
∑ (x )
n 2
i −x
s = s2 = i=1
n −1
Sample Standard Deviation: s = σ̂
Count 50
Mean 22.88
Dot Plot Median 21.5
Mode 21
Max 45
Min 8
Range 37
Std Dev (Pop) 7.648895345
Std Dev (Sample) 7.726551096
Variance (Pop) 58.5056
Variance (Sample) 59.69959184
Skewness 0.683704873
Kurtosis 0.525173863
• Individually
• For each data set, use SPC XL > Analysis Diagrams >
Summary Stats to find the following
Round 1 Round 2
Mean:
Median:
Range:
Sample Variance:
Sample Standard Deviation:
⎛ USL − y y − LSL ⎞
= minimum ⎜⎜ , ⎟
⎟
⎝ σ σ ⎠
USL − LSL
=
6σ
σcapability = 3 • Cp *
σ =2
4 7 10 16
LSL USL
σ=2
10 15 y = 16 20
σlevel =
t3
f irs st!
he d fir
yt
Tr han
= by
Cpk =
Cp =
dpm =
σcapability =
LSL
σ=2
10 y = 16
σlevel =
Cpk =
Cp =
dpm =
σcapability =
LSL USL
σ=2
10 15 y = 16
σlevel =
Cpk =
Cp =
dpm =
σcapability =
• Individually
Round 1 Round 2
Cpk =
Cp =
dpm =
A = # of units completely
unit 1 defect (error) free without any
kind of rework or touch-up of
any kind during the process
unit 2
n units …
. B = # of units that move on
Started . but have had at least some
. Defect Defect rework or touch-up performed
REWORK REWORK on them
unit n
PROCESS STEPS
C = # of units scrapped
A +B
Yield = n
A
First Pass Yield (FPY) = n
⎛ total # of defects ⎞
= (1,000,000 ) × ⎜⎜ ⎟⎟
⎝ total # of opportunit ies ⎠
or,
⎛ dpu ⎞
= (1,000,000 ) × ⎜⎜ ⎟⎟
⎝ opportunit ies per unit ⎠
• DO NOT:
– Measure the number of ways we can mess something up
• DO:
– Measure the complexity of a product or process
• RECOMMENDATIONS:
– Keep it simple
• For products, opportunities may be the number of piece
parts making up the product
• For processes, opportunities may be the number of value-
added process steps
dpu =
dpmo =
FPY =
σcapability =
σcapability vs dpm
Total Sigma
Characteristic Defects Units Opportunities Opportunities DPU DPO DPMO Capability
documents 195 3000 4 12000 0.06500 0.01625 16250.000 3.6382
Total 195 0 12000 0.01625 16250.000 3.6382
X X
Form 9 10 11 12 13 14 15 16
X
X X X
X X X X X
X
– PROPERTIES OF A GOOD
MEASUREMENT SYSTEM
Measure
Plan for data collection; Establish or
refine metrics if necessary
Update Knowledge/Project
Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
ANALYZE Close Down Project
• ROT:
– Variables (continuous data)
(Number of people) X (Number of parts or transactions) ≥ 20
OP 1 OP 2 OP 3
5. Recommendations?
• Press OK
USL:
LSL:
Inspection
Capability
OP 1 OP 2 OP 3 Total
Effectiveness 0.75 0.75 0.9 0.8
P(FR) 0.3 0.2 0.2 0.233333
P(FA) 0.2 0.3 0 0.166667
Bias 1.5 0.666667 NA 1.4
• Using SPC XL, set up an MSA template and enter the data
from the previous page.
Yes No
•• ••
• • •
• • •
•
•
•• •
Yes
•
• •
• • •
(repeatability)
PRECISION
•
•
•
• •
•
•
• •
•
• •
• •
•
•
• •
•
•
• •
•
No
• •
σ 2repeatability + σ 2reproducibility
REPEATABILITY:
Variation obtained by the same person using the
same procedure on the same product, transaction
or service for repeated measurements (variability
within operator).
REPRODUCIBILITY:
Variation obtained due to differences in people
who are taking the measurements (variability
between operators).
σtotal
σtotal
±3σmeasurement ±3σmeasurement
For each of the scenarios below, look at the data and indicate which of
the following you think is true:
a. Repeatability appears to be more of a problem than reproducibility
b. Reproducibility appears to be more of a problem than repeatability
c. Repeatability and Reproducibility appear to be about the same
• Scenario 1:
Operator 1 Operator 2
Rep 1 Rep 2 Rep 1 Rep 2
Part 1 21 23 26 28
Part 2 19 18 24 24
Part 3 20 23 27 24
Part 4 19 22 21 20
• Scenario 2:
Operator 1 Operator 2
Rep 1 Rep 2 Rep 1 Rep 2
Part 1 21 23 24 22
Part 2 19 18 21 18
Part 3 20 23 24 22
Part 4 19 22 22 20
• Scenario 3:
Operator 1 Operator 2
Rep 1 Rep 2 Rep 1 Rep 2
Part 1 21 23 25 23
Part 2 19 18 15 17
Part 3 20 23 23 25
Part 4 19 22 18 20
USL: 35.0
LSL: 10.0
Operator 1 Operator 2
Part # Reference Rep 1 Rep 2 Rep 1 Rep 2
1
2
3
4
USL: 35.0
LSL: 10.0
Operator 1 Operator 2
Part # Reference Rep 1 Rep 2 Rep 1 Rep 2
1 21 23 26 28
2 19 18 24 24
3 20 23 27 24
4 19 22 21 20
σmeas
P/TOT =
σ total
3. Discrimination or Resolution
(# of truly distinct measurements that can be obtained by the
measurement system)
⎛ σproduct ⎞
=⎜ ⎟ × 1.41
⎝ σmeas ⎠
USL 35
LSL 10
Precision to Tolerance Ratio 0.76131465
Precision to Total Ratio 0.91492535
Resolution 0.6
BIAS ANALYSIS
Reference Bias
Not Available
Z Axis 4
Interaction
Operator
3
Reproducibility
Repeatability
Part-to-Part
X Axis Category
30
25
20
Part Average
Operator 1
Operator 2
15 UCL = 25.963
Center = 22.438
LCL = 18.913
10
0
1 2 3 4 1 2 3 4
Part Number
Operator 1
Part Range
4
Operator 2
UCL = 6.126
3
Center = 1.875
LCL = .
0
1 2 3 4 1 2 3 4
Part Number
Operator By Part
30
25
20
Measurement
Operator 1
15
Operator 2
10
0
1 2 3 4
Part #
Sigma Total
Sigma Product
LSL
USL
dpm Potentially
Misclassified = 389,571.196
Sigma Total
LSL Sigma Meas
USL Sigma Meas
LSL
USL
USL 35
LSL 10
Precision to Tolerance Ratio 0.71425004
Precision to Total Ratio 0.85748159
Resolution 0.8
BIAS ANALYSIS
Reference Bias
Not Available
4
Z Axis
Reproducibility
Part-to-Part
XAxis Category Repeatability
4. If P/TOT >.30, check the items that were part of the measurement
system study. Do they represent at least 80% of the total process
variability? If they do and if the measurement system is already
state-of-the-art and performing up to its specifications, we may have
to live with it. One possible work-around is to take an average of 4
measurements, say, and use that value as the recorded entry for
that item and that measurement. (note: This involves the property of
the Sampling Distribution of the Mean, an advanced topic, which
says that the standard deviation of sample averages is reduced by a
factor of n , where n is the sample size. If n = 4, 4 = 2 , and so we
have cut the variability in half.) This technique should be used only
as a short-term approach to a gage study and should not be used as
a mask for an inherently poor measurement system.
Things to
Check
USL: 5.0
LSL: 0.0
USL 5
LSL 0
Precision to Tolerance Ratio 0.50929145
Precision to Total Ratio 0.3241343
Resolution 4.1
BIAS ANALYSIS
Reference Bias
Not Available
1.6
1.4
1.2
0.4
0.2
0
Part-to-Part
Reproducibility
Repeatability
Category
4 Operator 1
Profile Average
Operator 2
Operator 3
3
UCL = 3.028
Center = 2.704
2
LCL = 2.38
0
1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
Profile Number
1.2
0.8
Operator 1
Profile Range
Operator 2
Operator 3
0.6 UCL = .815
Center = .317
LCL = .
0.4
0.2
0
1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6
Profile Number
Operator By Part
4
Measurement
Operator 1
3 Operator 2
Operator 3
0
1 2 3 4 5 6
Profile #
Sigma Total
Sigma Product
LSL
USL
Sigma Total
LSL Sigma Meas
USL Sigma Meas
LSL
USL
• One team will serve as the control group that supplies the
products to each of the “operators” for measurement. The
control group will receive the measured value from the
“operator” and place it in the appropriate cell on the
accompanying (next page) spread sheet.
– IDENTIFYING WASTE
– 7 classic types of waste
– Cost of Waste
– ANALYZYING WORK
– Takt time
– Cycle Time
– Operator Loading
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 7-2
ANALYZE
Define
Measure
MEASUREMENT METHOD MACHINE
Carburetor Transmission
Gage for
Ma
nu
Tire Press.
Tire Press.
to
al
Estimate of
Full Tank Maint. Sched. Tune-Ups Weight Tire Type
Ra
Re
gu
WX
Engine Warm-
Type Fast Start Up Time Radio
Air Conditioner
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Ice
Oil Type
Passengers Snow
Terrain
Pr
A
Rain
Hu
ec
Te
ipi
mi
s
Wi
mp
ain
tat
dit
e
t
Fla
nd
tud
era
ion
y
unt
Driver
tur
Alti
Mo
Fuel Additive
Acctng
Identify the non-value added activities,
HR
Legal
sources of waste, and defects and/or Analyze
Functional
Area
Benefits
problems
Frequency or
Financial Measure
Control
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
IMPROVE Close Down Project
Copyright Air Academy Associates 7-3
GAP BETWEEN CURRENT
PERFORMANCE AND GOAL
• Examples
70
60
Cycle Time (minutes)
50
40
30
20
Goal: 20 minutes
10
0
6-Aug
13-Aug
20-Aug
27-Aug
3-Sep
10-Sep
17-Sep
24-Sep
1-Oct
8-Oct
15-Oct
22-Oct
2-Jul
9-Jul
16-Jul
23-Jul
30-Jul
Week of
• Sources of information
– brainstorming
– observation
– process map(s)
– discussion with operators, team, etc.
– data
Causes or
Sources of
Variation
Effect or
Response
(CTC)
or problem
statement
Miscellaneous Methods Machines involving the
CTC
MEASUREMENT
MEASUREMENT METHOD
METHOD MACHINE
MACHINE
Carburetor Transmission
M
Gage for
an
Ride Brakes Tire Rotation
Au
Tire Press.
ua
to
Tire Press.
l
Estimate of
Full Tank Maint. Sched. Tune-Ups Weight Tire Type
R
R
Al
eg
ad
Front or
ul
i
W
al
Rear Drive
ar
Speedometer
X
Engine Warm-
Type Fast Start Up Time Radio
Air Conditioner
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Oil Type Ice
Passengers
Pr
Terrain Snow
ec
H
A
Te
B
Rain
um
ip
m
it a
W
id
pe
ns
in
t io
ity
at
ra
d
tai
ud
n
Fl
tu
Driver
un
re
tit
Mo
Al
Fuel Additive
PEOPLE
PEOPLE MATERIALS
MATERIALS ENVIRONMENT
ENVIRONMENT Gary Greenbelt
Update: 2/4/2004
C = Constant
• To hold a variable Constant requires Standard Operating
Procedures (SOP’s) and Mistake Proofing to eliminate
errors and reduce variation in the CTCs.
• Holding a variable constant doesn't just happen. It must
be engineered into the process.
N = Noise
• Noise variables are those that are not being controlled or
held constant.
• Mistake Proofing is needed to change an "N" variable to
a "C" variable.
X = Experimental
• These are key variables that can be controlled and held
constant at different levels or settings for the purpose of
determining the effect of this variable on the CTC.
Method 1
1. Describe the problem or CTC (output) being discussed and place
it at the “head” of the cause and effect (or fishbone) diagram
2. Determine appropriate* cause categories (the major “bones”)
3. Brainstorm potential causes (or sources of variation) within each
category, which are believed to affect the problem or CTC, and
place on the diagram
Method 2
1. Describe the problem or CTC (output) being discussed and place
it at the “head” of the cause and effect (or fishbone) diagram
2. Brainstorm potential causes (or sources of variation) and place
on sticky notes
3. Cluster the causes into related groups and name the groups.
Use the group names as the major “bones” on the cause and
effect diagram and place the causes within each group
• Structured:
– Go around the room and have each team member,
in turn, give an idea
– Accept each idea without analysis or judgment
– If a person is out of ideas, they may pass their turn
onto the next person
– Continue, around the room, until everyone is out of
ideas
– As a group, review all the ideas. This provides an
opportunity for clarification and to discard duplicate
ideas. Often, the review may stimulate the
generation of new ideas.
• Unstructured:
– Identical to the previous process except that ideas
can be given at any time by anyone without the
need to follow a structured rotation of the team
members
• Overproduction
• Waiting
• Transportation
• Inventory
• Processing
• Motion
• Defects
Actual
Value
Transportation Added Wasted
Motion
Processing
Product
Defects
Excess
Inventory
$$$
$$$hares
Extra Handling
Extra Defects
Examples
– An aircraft waits for de-icing before it can take off
– A waiter must wait before he can process a credit card
– An underwriter must wait for a credit check before
completing a loan application
– Items are processed monthly, not as work comes in
Examples
Hiring Process – Applicants folder is
hand carried from one department
to another for approval.
DOCK
Functional Benefits
Area
After
Warehouse
HR Legal
Functional Benefits
Area
Examples
– Procurement orders 250 lbs of steak when only 100 lbs is
required
– Hospital orders enough gauze pads to last for 17 months
– Multiple copies of a document are stored in every desk
Before After
Change Request Change Request
Final Document
Created
Final Approval
Examples
– Bank teller who must walk from her primary location to an
alternate location in order to deposit payroll taxes
– A waiter who must go to two different locations to place food
and drink orders
COST $28,000
ROI
3.2 months
Kaizen Team Changes And $95,000
More Sellable
Added Rotating Table Goods/Yr.
This is GREAT!!
$3,000/once
Reduced time by
62%
Increased Production
$95,000/yr
Examples by Area:
Waiting • Waiting for information • Waiting for test • Waiting for people to • Waiting for purchase
from a supplier scripts to be complete reviews requisition approval
• Waiting for available test developed • Waiting for information
equipment
Transportation • Delivery of test parts from • documentation loops • Paperwork and • Delivery of paper and other
a supplier documentation loops materials to a copy center
Inventory • Stockpiling more test • Pending requests for • Marketing materials • Multiple copies of a
parts than required software tests not mailed document stored in every
• Excess test equipment desk
Motion • Extra walking and • Having to perform • Searching for people • Walking back and forth
reaching into bins to extra keystrokes to to get feedback from stockroom to get
retrieve test parts accomplish a task paper for a copy machine
Defects • Incorrect tolerances • A logic error in coding • Not supplying the • An inaccurate report which
• Parts which do not meet correct information to must be corrected
specification the customer
Appraisal Costs
Prevention Costs
Some examples:
• Over-Production
• Waiting
• Transportation The 7 Classic Wastes
• Inventory
• Processing
• Movement
• Defects
• Design Corrective Action
• Rework-Design Change
• Scrap-Design Change
• Production Liaison Costs
• Purchasing Failure Costs
• Purchased Material Replacement Costs
• Supplier Corrective Action
• Rework of Supplier Rejects
• Uncontrolled Material Losses
• Material Review/Corrective Action
• Disposition Costs
• Troubleshooting/Failure Analysis
• Investigation Support Costs
• Operations Corrective Action
• Failure to Transfer from Design to Manufacturing
• Rework
• Repair
• Reinspection/Retest Costs
Some examples:
• Customer Complaint
Investigation
• Returned Goods
• Retrofit Costs
• Recall Costs
• Warranty Claims
• Liability Costs
• Penalties
• Customer/User Goodwill
• Premium Freight (overnight)
• Pricing Errors
Some examples:
• Document Checking
• Purchasing Appraisal Costs
• Receiving/Incoming Inspection and
Test
• Qualification of Supplier Product
• Source Inspection and Control
Programs
• Manufacturing Appraisal Costs
• Planned Inspections, Tests, Audits
• Financial Audits
• Checking Labor
• Inspection and Test Materials
• Set-Up Inspections and Tests
• Depreciation Allowances
• Measurement Equipment Expense
• Maintenance and Calibration Labor
• Outside Endorsements and
Certifications
• External Appraisal Costs
• Field Performance Evaluations
• Special Product Evaluations
• Review of Test and Inspection Data
• Other Quality Evaluations
• Special Tests
• Process Control Measurement
• Laboratory Support
Copyright Air Academy Associates 7 - 28
THE COST OF POOR QUALITY (cont.)
PREVENTION COSTS
The costs of all activities specifically designed to prevent
defects (poor quality) in products or services. These occur
prior to production and are aimed at preventing defects
before they occur.
Some examples:
• Marketing Research
• Customer Perception Surveys
• Contract/Document Review
• Design Quality Progress Review
• Design Support Activities
• Product/Service Design Qualification
Test
• Field Trials
• Supplier Quality Reviews/Ratings
• Purchase Order Technical Data Review
• Supplier Quality Planning
• Operations Prevention Costs
• Operations Process Validation
• Operations Quality Planning
• Operational Support Quality Planning
• Operator Quality Education
• Operator SPC/Process Control
• Quality Administrative Salaries and
Expenses
• Quality Program Planning
• Quality Performance Planning and
Reporting
• Quality Education
• Quality Improvement
• Quality System Audits
• Training without ROI
Copyright Air Academy Associates 7 - 29
THE COST OF POOR QUALITY (cont.)
LOST OPPORTUNITY COSTS
Some examples:
• Lost Sales
• Lost Customers
• Reliability Issues
• Safety Issues
• Ergonomic Issues
• Legal Issues
• Review each step and identify the value added (VA) steps. (all
others are non-value added (NVA) or non-value added but
required (NVA-R).
VA time
• Calculate the process cycle efficiency: =
Total Lead Time
1 minute
0.5 minutes 0.1 minutes 0.05 minutes
move any
park the car get out of car lock the car shopping carts
in my way
1 minute
0.5 minutes
1 minute 1 minute
talk with kids
exit the
checkout and selling items at walk back to
grocery
pay for coffee the door of the car
store
store
1 minute
move 0.1 minutes
someones 0.05 minutes 0.1 minutes
put coffee bag
shopping cart unlock car get into car
in the car
away from my
car
0.5 minutes
0.5 minutes 10 minutes 10:36 am
wait for other
cars to move drive out of the drive back
from behind parking lot home
finish
me
– brainstorming
– Takt time
• TAKT time is
• determined by the customer
• important for understanding work balance and flow
• is NOT cycle time!
Assume:
One 8 hour shift
One 1 hour lunch break
Average customer demand 400 units per day
3,600 sec
Total Work Time = 8 hours • = 28,800 sec
hour
3,600 sec
Non-Working Time = 1 hour • = 3,600 sec
hour
Total Available
28,800 sec – 3,600 sec = 25,200 sec
Production Time =
25,200 sec
TAKT Time = = 63 sec/unit
400 units
Assume:
Two 8 hour shifts
Each shift is given 1 hour for meals
Each shift is given 3 breaks @ 10 minutes each
Average customer demand 200 units per day
3,600 sec
Total Work Time = 2 • 8 hours • = 57,600 sec
hour
Non-Working Time = 1 hour for meals
+ 30 minutes in breaks
= 90 minutes per shift
• 2 shifts
= 180 minutes per day • 60 sec = 10,800 sec
minute
Total Available
Production Time = 57,600 sec – 10,800 sec = 46,800 sec
46,800 sec
TAKT Time = = 234 sec/unit
200 units
Op 1 Op 2 Op 3 Op 4
• Three types
250 sec
Total Operators Required = = 3.09
81 sec
Seconds
100 90 sec
82 sec 78 sec
80 TAKT Time = 81 sec
60
40
20
80
60
40
20
80
60
40
20
80
60
40
20
• Load Chart
Avg.
Step No. Description 1 2 3 4 5 6 7 8 Max. Min. Comments
Time
1 Pick up part
3 Attach item #2
7 - 47
TOTAL
HOW TO COMPUTE
OPERATOR CYCLE TIME
Avg.
Step No. Description 1 2 3 4 5 6 7 8 Max. Min. Comments
Time
7 - 49
TOTAL
TIME OBSERVATION FORM
Cumulative time
Process: Date: Plant:
Avg.
Step No. Description 1 2 3 4 5 6 7 8 Max. Min. Comments
Time
1 27 1:00 1:29 2:06 2:34 3:03
1 Pick up a part 3 1 1.86
1 2 1 3 2 2 2
11 40 1:10 1:47 2:19 2:45 3:13
7 - 50
TOTAL 25 34 27 38 28 29 30 43 24 30.14
• Select the operator cycle time based upon where you are
in process improvement.
• Usually, the time required for one cycle has low variability.
However, if there is a great deal of variability in the
machine cycle time, take the average of seven readings.
• The machine cycle time plus the time required for the
operator to load, unload, start, and perform any tool
change time
100
83 sec
TAKT Time = 81 sec
80
60
Machine Time
40 Operator Time
20
TAKT Time
Actual Capacity = • Demand
Processing Cycle Time
81
= • 200 = 195
83
TECHNIQUES FOR
NARROWING THE FOCUS
TECHNIQUES FOR
NARROWING THE FOCUS
Measure
MEASUREMENT METHOD MACHINE
Carburetor Transmission
Gage for
Ma
nu
Tire Press.
Tire Press.
to
al
Estimate of
Full Tank Maint. Sched. Tune-Ups Weight Tire Type
Ra
Re
gu
WX
Engine Warm-
Type Fast Start Up Time Radio
Air Conditioner
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Ice
Oil Type
Passengers Snow
Terrain
Pr
A
Rain
Hu
ec
Te
ipi
mi
s
Wi
mp
ain
tat
dit
e
t
Fla
nd
tud
era
ion
y
unt
Driver
tur
Alti
Mo
Fuel Additive
Acctng
Identify the non-value added activities,
HR
Legal
sources of waste, and defects and/or Analyze
Functional
Area
Benefits
problems
Frequency or
Financial Measure
Control
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
IMPROVE Close Down Project
Copyright Air Academy Associates 8-2
PRIORITIZING AND
NARROWING THE FOCUS
• As the team measures and analyzes the
process, the goal is to identify the key input
variables and process steps to focus on
Radio
M
ut
an
o
ua
Estimate of
l
Full Tank Maint Sched
. Tune-Ups Weight Tire Type
Front or
Speedometer Engine Warm- Rear Drive
R
A
ad
eg
ll
Type Fast Start Up Time
ia
ul
l
ar
X
Air Conditioner
GAS
Driver
B
Gas Type MILEAGE
A
Highway or City
Training
Weather Conditions
ns
# of Oil Type Ice
tai
e
Passengers Snow
ud
t
un
Rain
Fla
tit
Mo
Pr ity re
Al
H
um
W
ec
Te
Driver
in
ip
id
m
d
ita
pe
tio
Fuel Additive
ra
n
tu
PEOPLE MATERIALS ENVIRONMENT
-- Data
- Graphical analysis
- Statistical tools
Key inputs
and Process steps
(causing problems, affecting cost
and/or performance,
and generating waste)
B
Marilyn Vos Savant
Columnist in Parade Magazine and
Guinness World Book record holder for
highest IQ
Address 11 .11
Ship Early 9 .09
Ship Late 37 .37
Method 6 .06
Contents 8 .08
Packing 29
100
.29
1.00 A Pareto chart
Types of Errors and Frequency of Occurrence
turns data into
information
40
35 focus on the
30 causes that occur
most frequently!
Frequency
25
20
15
10
0
Ship Late Packing Address Ship Early Contents Method
– systematic sampling
Population Sample
Parameters Statistics
(μ, σ) ( x , s)
Confidence Interval =
Sample Estimate ± Margin of Error
Sample 1
POPULATION
Sample 2
Sample 3
σx
n
X
μx μx
X
⎛U⎞ ⎛ s ⎞
⎜⎜ ⎟⎟ = x ± Z ⎜ ⎟
⎝ ⎠
L ⎝ n ⎠
Where
U = upper confidence limit
L = lower confidence limit
x = sample average
Z = 2 (for 95% confidence) or
3 (for 99% confidence)
s = sample standard deviation
n = sample size
EXAMPLE:
Over the course of a week, we randomly select and time 16
customer service calls. We find that the average time is 15.6
minutes with a standard deviation of 2.1 minutes. What is a
99% confidence interval for the true average service time?
Confidence Interval
Lower Limit Upper Limit
14.05296861 17.14703139
Results
Estimated Sample Size (n) 29.3
⎛U⎞ pq
⎜⎜ ⎟⎟ = p ± Z
⎝L⎠ n
Where
U = upper confidence limit
L = lower confidence limit
p = proportion of “defectives”
(or category of interest) in the sample
q = 1 – p (q is the proportion of “non-
defectives”)
Z = 2 (for 95% confidence) or
3 (for 99% confidence)
n = sample size
EXAMPLE:
You work in a finance office and are in charge of processing
travel vouchers submitted by several different organizations.
You sample 100 vouchers and find 8 to have discrepancies
or errors. Find a 95% confidence interval for the true but
unknown proportion of vouchers containing errors.
Confidence Interval
Lower Limit <p< Upper Limit
0.026827589 0.08 0.133172411
Results
Estimated Sample Size (n) 752.9
Copyright Air Academy Associates 8 - 14
EXERCISE
X
p=
n
Lower
Bound
Upper
Bound
Gage for
METHOD
Tire Press.
Transmission
M
ut
an
o
ua
Estimate of
Weight Tire Type
l
Full Tank Maint Sched
. Tune-Ups
Front or
Speedometer Rear Drive
Engine Warm-
R
ad
Type
ll
Up Time
eg
Fast Start
ia
ul
l
– unavailable
ar
Air Conditioner
GAS
Gas Type MILEAGE
B
Driver
A
Highway or City
Training
Weather Conditions
Al tains
# of Oil Type Ice
e
ud
Passengers Snow
un
tit
t
Rain
Fla
Mo
Pr
Driver
um
ec
in
id
Te
d
ip
ity
m
ita
Fuel Additive
pe
tio
ra
n
B
tu
re
PEOPLE MATERIALS ENVIRONMENT
– very limited
– difficult to collect
Person
Cause Total Ranking
1 2 3 4 5
A: Address Incorrect 4 2 1 2 3 12 #2
B: Quality Issues 1 1 3 1 1 7 #1
C: Damage, Supplier 2 5 4 5 2 18 #4
D: Shipping Company 3 3 2 3 4 15 #3
E: Database errors 5 4 5 4 5 23 #5
Low 2 4
Effort
Required
• Example: High 1 3
1
2
Damage from
Quality suppliers shipping
3 issues company
(defects)
Late
Shipments
Address
4 incorrect
Database
errors 3
A: Address Incorrect
B: Quality Issues B
C: Damage, Supplier A B
D: Shipping Company A B D
E: Database errors A B E D
4
3
2
1
0
B: Quality A: Address D: Shipping E: Database C: Damage,
Issues Incorrect Company errors Supplier
Database Accuracy
Shipment Cost (9)
Address information
Shipping
Process Number of Damaged Shipments (8)
Internal Quality issues
(Customer Priority)
Process "Inputs"
Supplier Quality 1* 1 16
Database Accuracy 9 1 3 117
Address Information 9 1 3 117
Internal Quality Issues 9 1 1 1 113
Shipping Company used 3 9 9 1 189
Total: 552
Colonial Williamsburg
Uncle Joe’s
Total:
Copyright Air Academy Associates 8 - 22
OTHER USES FOR THE IPO
(PRIORITIZATION) MATRIX
Project #1
Project #2
Project #3
Project #4
Total:
Solution #1
Solution #2
Solution #3
Solution #4
Total:
• Example:
Problem: The cylinder is moving too slowly.
Fu i l u
connection
cl
og
el r e
fa
Below -200F
ge
pu
d
m
p
Water in fuel
My car
Clutch not won’t start
Bad starter Dead battery depressed Car not in
Loose Park/neutral
connection
Wrong key
Dead
No electrical Battery Never Check; Too busy
power No routine
maintenance
Lazy
Car
won’t
start
Never learned
Clutch not Not familiar or had access
Operator depressed with manual to car with
Error (manual) transmission manual
transmission
Car not in
park or Forgot
Other neutral
branches
• You don’t have any historical data on the cycle times for
individual steps in the launch process. Your team has
brainstormed reasons (causes) for why it takes so long
to launch the statapult® ball (shown on the next page)
and you’re not sure where to focus or where to start your
data collection efforts.
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 10 - 2
ANALYZE
Define
Measure
MEASUREMENT METHOD MACHINE
Carburetor Transmission
Gage for
Ma
nu
Tire Press.
Tire Press.
to
al
Estimate of
Full Tank Maint. Sched. Tune-Ups Weight Tire Type
Ra
Re
gu
WX
Engine Warm-
Type Fast Start Up Time Radio
Air Conditioner
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Ice
Oil Type
Passengers Snow
Terrain
Pr
A
Rain
Hu
ec
Te
ipi
mi
s
Wi
mp
ain
tat
dit
e
t
Fla
nd
tud
era
ion
y
unt
Driver
tur
Alti
Mo
Fuel Additive
Acctng
Identify the non-value added activities,
HR
Legal
sources of waste, and defects and/or Analyze
Functional
Area
Benefits
problems
Frequency or
Financial Measure
Control
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
IMPROVE Close Down Project
• We set up 2 hypotheses
– H0 is called the null hypothesis (no change, no
difference)
– H1 is called the alternate hypothesis
– Example: H0 : μ1 = μ2 vs. H1 : μ1 ≠ μ2
20 22 24 26 28 30 32 34 36 38 40 42 44
H0 H1
Note: All times include wait time and approval times in addition to actual
value added time spent completing the form
H H H H H HH H H
L L L L L L L L L
2 3 4 5 6 7 8
H0: μL = μH
versus
H H H H H HH H H
L L L L L L L LL
2 3 4 5 6 7 8
H1: μL ≠ μH
Note: If the leftmost and rightmost symbols are the same, then the End
Count is zero. The significance associated with a given End Count can be
found using the following table:
The results below represent the p-values from a 2 sample t-test. This
means that the probability of falsely concluding the alternative
hypothesis is the value shown (where the alternate hypothesis is that the
means are not equal). Another way of interpreting this result is that you
can have (1-pvalue)*100% confidence that the means are not equal.
• Rule of Thumb:
– If p-value < .05, highly significant difference
– If .05 < p-value < .10, moderately significant
difference
– (1 – p-value) • 100% is our percent confidence that
there is a significant difference in the averages
Copyright Air Academy Associates 9 - 10
RULES OF THUMB
Quick Test for Detecting a Significant
Shift in Standard Deviation
smax
1. If > 2.72 then there is a significant shift in standard
smin
deviation.
smax
A “gray zone” occurs when 1.5 ≤ ≤ 2.72 . If this is the
smin
s2max n1 + n2
2. If > 10, then there is a significant shift in
s2min 2
Note: If the sample sizes are the same, s (standard deviation) can be
replaced by R (range).
The results below represent the p-values from a 2 sample F-test. This
means the probability of falsely concluding the alternative hypothesis is
the value shown (where the alternate hypothesis is that the variances are
NOT equal). Another way of interpreting this result is that you can have
(1-pvalue)*100% confidence that the variances are not equal.
• Rule of Thumb:
– If p-value < .05, highly significant difference
– If .05 < p-value < .10, moderately significant
difference
– (1 – p-value) • 100% is our percent confidence that
there is a significant difference in the standard
deviations
Copyright Air Academy Associates 9 - 12
EXERCISE
L 201 209 215 221 211 213 217 205 218 208 203 214 212 215 211.6 5.8
H 218 225 217 222 223 220 222 216 221 224 224 221 220 219 220.9 2.7
• In Teams
What we learned:
Test of Proportions
Results
• Rule of Thumb:
– If p-value < .05, highly significant difference
– If .05 < p-value < .10, moderately significant
difference
– (1 – p-value) • 100% is our percent confidence that
there is a significant difference in the two
proportions
• Individual or Team
Special Notes:
a. For a one-tailed test (e.g., H0: μ1 < μ2 vs. H1: μ1 > μ2 ) then divide the P-value by 2 prior to Step 6.
b. You cannot use this test if the data values are “paired” (e.g., measuring the same part using two
different instruments).
Special Notes:
a. For a one-tailed test (e.g., H0: σ1 < σ2 vs. H1: σ1 > σ2 ) then divide the P-value by 2 prior to Step 6.
b. You cannot use this test if the data values are “paired” (e.g., measuring the same part using two
different instruments).
Special Notes:
a. For a one-tailed test (e.g., H0: π1 < π2 vs. H1: π1 > π2 ) then divide the P-value by 2 prior to Step 6.
b. Do not use this test if either number of “defectives” observed (i.e., X1 or X2) is too small.
Rule of Thumb: The number of defectives in each group must both be 5 or greater.
Reason
for
Return Cosmetic Hardware Software
Total
Defect Failure Incompatible
Product
Line
1 125 43 12 180
2 136 55 8 199
3 186 75 20 281
4 78 33 10 121
Step 2: α = 0.05
Total
125 43 12
1 180
(121.00) (47.48) (11.52)
136 55 8
2 199
(133.77) (52.49) (12.74)
186 75 20
3 281
(188.89) (74.12) (17.99)
78 33 10
4 121
(81.34) (31.92) (7.75)
781
Total 525 206 50
(N)
χ = ∑∑ = 3.604
2 ij
i =1 j =1 Fij
χ 26
.95 .7301
.05 .2699
Color
– CELLULAR MANUFACTURING
– Principles of Cell Design / Layout
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 10 - 2
IMPROVE
Define
170
150
140
130
120
110
100
90
80
validate improvement
1
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
Month Number
Analyze
Control
Update Knowledge/Project Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
CONTROL Close Down Project
VIDEO
Time Time
1st
EXT.
2nd
Changeover
Time
EXT. 3rd
INT.
EXT.
INT. INT.
10 - 12
BUILDING A SMED CHECKLIST
Equipment
Molds/Dies _____ _____ _____ _____
Fixtures/Jigs _____ _____ _____ _____
Transport trays/racks _____ _____ _____ _____
Carts/dollies _____ _____ _____ _____
Hand tools/accessories _____ _____ _____ _____
Ancillary machines _____ _____ _____ _____
Supplies/Apparel
Pens/pencils/paper _____ _____ _____ _____
Gloves or Masks _____ _____ _____ _____
Helmets/shoes _____ _____ _____ _____
Coats/aprons _____ _____ _____ _____
Cleaning supplies _____ _____ _____ _____
(Physical Layout)
Storage End
Assembly
Machining
Packing
Storage
Storage
Start Storage
Storage Deburring
Inspection
Storage
Assembly
Inspection Deburring
Workers
Packing Machining
Customer Raw
Material
Team 1 Team 2
Total Elapsed time
to obtain 12 completed
sheets
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Machining x xx xx x
Deburring xx xxx x
Assembly x xx x xx
Testing xxxxxx
Packing x x xx xx
Machining x xx xxx
Deburring xx xxxx
Assembly x xxxxx
Testing xxxxxx
Packing xxxxxx
66% Reduction in Lead Time
Material
Out
In
Material Material
14-18 inches
18-24 inches
28 – 33 in.
• Current Layout
Mtl Mover
Supply
Customer
Op #2 Op #3
QC
Op #4 Op #1
Total # of
Nameplates
Shipped
Time till
first unit
received by
customer
Total WIP
Total
Quality
Rejects
Profit
(Loss)
Net Profit/Loss
• Goals
– improve responsiveness to customer
– make more money
– use some of the tools we have learned
• Constraints
– must still use batch size of 6
– each operator must remain in their department
Net Profit/Loss
• Goals
– improve responsiveness to customer
– make more money
– use some of the tools we have learned
• In teams,
– complete further analysis and make additional
improvements to the process
– you may conduct any time studies, as needed
– ensure you understand the customer requirements
– brainstorm ideas for improvement
– select and implement 3 improvement ideas
• Additional rules
– no requirements for batch size
– operators may be cross trained
– work may be redistributed
Net Profit/Loss
– Measures (Statistical)
• Compare process average, standard deviation, Cp,
Cpk, etc.
• Hypothesis testing
μ1 μ2
2 std deviations
σ1
σ2
2 proportions
%1 vs %2
Count/classified
data
• Graphical Evidence
5
Cycle Time
before
3
2
after
1
0
1 3 5 7 9 11 13 15 17
X Axis
Shift in process,
with lower variation
(i.e., “smaller” box)
5
Q1
4 Min
Median
3 Max
Q3
2
0
Shift 1 (Before) Shift 1 (After)
• Numerical Evidence
Mode 0.8
Max 6.3 1.5
Min 2.8 0.5
Range 3.5 1
Std Dev (Pop) 1.10598819 0.314269681
Standard Deviation Standard Deviation
Std Dev (Sample) 1.173078 0.33333 = 1.17 hours = 0.33 hours
Variance (Pop) 1.223209877 0.098765432
Variance (Sample) 1.376111111 0.111111111
Dot Plot of Cycle Times
0 1 2 3 4 5 6 7
t-test shows a
statistically
significant
difference in the
t Test Analysis (Mean) average cycle
P-value = 0.0000001 times
F-test shows a
statistically
significant difference
in the standard
deviations
F Test Analysis (Std Dev)
P-value = 0.001814
• SOPs
– Should document consistent and reliable methods
for performing tasks and/or holding variables
“constant”
Step
Work Steps Operator Time Units
Work Sequence Layout
checklists
#
9 4 1
Op #2
10
Completed Blank
Components Boards
Cart Cart
What is it?
FMEA is a systematic method for identifying, analyzing,
prioritizing, and documenting potential failure modes,
their effects on a system, product or process
performance, and the possible causes of failure.
Types of FMEAs
CHARLESTON, S.C. — A jury Wednesday The $12.5 million in actual damages and
ordered Chrysler Corp. to pay $262.5 million $250 million in punitive damages awarded
to the parents of a 6-year-old boy killed when to the parents of Sergio Jimenez II was the
he was thrown from the family minivan in an largest award against the automaker, a
accident because a rear latch was defective. company spokesman said.
• Failure Modes
• Recommended Actions
9 KISS (“Texas-Style”)
9 Formal (spreadsheet)
• Brainstorm:
– How the product or process can fail?
– What actions are needed to prevent these failures
from happening (i.e., mistake proofing)?
– Who is accountable for these actions and when they
will be completed?
Product Failure Failure Effects S Causes O Controls D R Actions Plans pSEV pOCC pDET p
Component Mode (Consequence of E (of Failure C E P R
or Process Failure Mode) V Mode) C T N P
Step N
10 - 46
“customer” “customer” “customer”
SAMPLE PROCESS FMEA: SELF-SERVE GAS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Process Failure Failure Effects S Causes O Controls D R Actions Plans pSEV pOCC pDET p
Step Mode (Consequence of E (of Failure C E P R
Failure Mode) V Mode) C T N P
N
Find gas All pumps Must wait for pump 1 Friday / pm 3 Fill on Thursday 1 3
pump busy rush hour
Find another 3 Impatient / 3 More exercise 2 18
station hurry
10 - 47
Pay for gas on 2 Auto tank 5 None 5 50
ground design
SAMPLE PRODUCT FMEA:
AUTOMOBILE GLASS
p
S O Current D R p p p
Product Potential Failure Potential S O D R
Failure Mode E C Design E P Actions Plans
Components Effects Causes E C E P
V C Controls T N
V C T N
Item on Engineering
Does not squirt Obscured vision Add fluid
Windshield No fluid in periodic to review
washer fluid on for driver and 4 5 3 60 low level 4 2 1 8
Washer reservoir maintenance
windows passengers light by 7/1/01
checklists
Screen
4 Blocked line 2 1 8 None
installed
Review Design
Pump doesn't eng'r'g
4 1 None 5 20 historical 4 1 3 12
work review by
failure data
6/1/01
10 - 48
Loose
Crimp
3 connection to 2 2 12 None
connector
electric motor
SOME MISTAKE PROOFING
(POKA YOKE) IDEAS
(and other ways to simplify and prevent defects)
• Checklists
• Templates/Pictures
• SOP’s
• Others?
• In small groups
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Product Failure Failure Effects S Causes O Controls D R Actions Plans pSEV pOCC pDET p
Component Mode (Consequence of E (of Failure C E P R
or Process Failure Mode) V Mode) C T N P
Step N
CONTROLLING PROCESS
PERFORMANCE
(REALIZING AND HOLDING THE GAINS)
CONTROLLING PROCESS
PERFORMANCE
– PROJECT DOCUMENTATION
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 11 - 2
CONTROL
Define
UCL
CENTERLINE
Monitor performance and measure
success
LCL
Measure
Performance No
confirmed and Gap Analysis
project goals Analyze
achieved?
Yes
“LISTENING TO THE
VOICE OF THE PROCESS”
CONTROL CHART
Lean Six Sigma: A Tools Guide
Q
U Control limits extended
A
L
I Control Region
T captures variation
Y natural to the
original process
M
E
A Control limits Process trend
S established
U
Process has shifted
R
Special cause identified
E
S
TIME
• Many Uses
– In the Measure phase
Checking to see if a process is “in control” and
stable before relying on metrics like Cp and Cpk
Must be some
special cause
of variation
INPUTS OUTPUTS
Processing
MEASUREMENT (days, dollars,, …)
Number rejected
11 - 7
EXAMPLES OF WHERE SPC HAS
PROVEN SUCCESSFUL
Work Area Process Output(s)
hospital health care operating room delays, surgery time,
recovery time, number of patients,
surgery costs, lab costs, time to
process lab results
secretarial office typing cycle time, number of errors, number
of pages per day, number of phone
calls, time to answer a call
customer response to time to return phone calls, number of
service complaints complaints, time to repair, number of
repeat complaints
barber shop hair styling number of clients, time to wash, time
to dry, time to cut, cycle time, waiting
time, number of re-cuts
assembly shop assemble parts number of completed parts per day,
number of errors, cycle time
pharmaceutical manufacture weight of tablet, number of defects,
firm drugs cycle time, dimensions of tablets
PROCESS CONTROL
IN CONTROL
(SPECIAL CAUSES ELIMINATED)
TIME
SIZE
OUT OF CONTROL
(SPECIAL CAUSES PRESENT)
PROCESS CAPABILITY
IN CONTROL?
YES
CAPABLE?
NO
22
Gas Mileage*
20
18
1 5 10 15 20 Week
•To keep this example as simple as possible, gas mileage values
are rounded to integer values.
20
Gas Mileage
y
22
20
18
1 5 10 15 20 Week
∑ (y − y )
2
32
s 2
= = = 1.684
n −1 19
Freq. of Weeks
18 19 20 21 22 Gas
Mileage
y
μ ± 1σ ≅ 68.27%
μ ± 2σ ≅ 95.45%
μ ± 3σ ≅ 99.73%
Gas Mileage
(mpg)
UCL = 23.9
24
22
20 ȳ = 20
18
LCL = 16.1
16
1 5 10 15 20 Week
Carburetor Transmission
M
Gage for Au an
Ride Brakes Tire Rotation t o
ua
Tire Press. l
Tire Press.
Estimate of
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Ice
Oil Type Pr
Passengers H ec Snow
Terrain
A
Te
B
m
um ip Rain
W id ita
s
in pe it y tio
d
e
ra
at
n
a in
tu
Fl
ud
nt
Driver re
tit
u
Al
Mo
Fuel Additive
PEOPLE
PEOPLE MATERIALS
MATERIALS ENVIRONMENT
ENVIRONMENT
11 - 14
CONTROL CHARTS
• VARIABLES DATA
– Most powerful
• ATTRIBUTE DATA
– Provides a record of quality history with data that is
often already being collected
Need to
construct a
control chart
n = sample size or
subgroup size
Is n fixed Variables
x − S chart
No Variables (measurements) Attributes
and or attributes?
n < 10? (go-nogo)
Yes
n=1 2≤n≤9
IMR
x − R chart
chart
Constant Constant
Yes No Yes No
sample area of
size? opportunity?
14
13
12
11
10
9
8
X
R
11 - 17
SUMMARY OF OUT-OF-CONTROL
SYMPTOMS
UCL
ZONE A
ZONE B
ZONE C
X ZONE C
ZONE B
LCL ZONE A
one or more pts are outside the control limits 7 cons. pts are on one side of the centerline
UCL UCL
A A
B B
C C
X X
C C
B B
A A
LCL LCL
7 cons. increasing or decreasing intervals 2 out of 3 cons. pts are in same Zone A or beyond
UCL UCL
A A
B B
C C
X X
C C
B B
A A
LCL LCL
4 out of 5 cons. pts are in same Zone B or beyond 14 cons. pts that alternate up and down
UCL
A
B
C
X
C
B
A
LCL
• EXAMPLE:
– Create a template for an XbarR Chart (subgroup size = 3)
– Open the data file XbarR data.xls in the SPC XL directory
(or go to SPC XL > About SPC XL (Help) > Tutorial > Control
Chart Tutorials > XbarR and XbarS Charts > Open data file)
– Copy and Paste the Sample Data into the template you
created
Enter the data in the column provided below. XAxis labels are optional. When you
are ready to create the control chart select "SPC XL" - "Control Charts" - "XBarR
Chart..." from the menu.
• Rational Subgroups
• Class as a whole
11
10.5
10
8.5
7.5
7 Series1
Center = 8.044
5.5
5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
11 - 26
MOVING RANGE CHART FOR pH READINGS
3.5
3
Series1
1.5
0.5
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
15-Minute Time Intervals
11 - 27
EXERCISE (IMR CHART)
0.12 Series1
Center = 0.0531
UCL = 0.0846
LCL = 0.0216
0.1
0.06
proportion
0.04
0.02
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
11 - 30
Week Number
PARETO CHART
80 REASONS FOR SUBSTITUTIONS
70
60
40
Percentage
30
20
10
0
TDY Meetings Leave Obs/Trng Medical Misc Retreats
11 - 31
p-CHART FOR SUBSTITUTIONS
0.12
Fall Semester
0.1 Spring Semester
0.06
UCL = .0497
0.04
Proportion of Substitutions
Center = .0248
0.02 New Policy
0 LCL = 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Week Number
11 - 32
EXERCISE
p (proportions)-chart
c (count)-chart
Day 1 2 3 4 5 6 7 8 9 10 11 12 13
Errors 25 24 23 23 27 25 21 27 39 23 22 23 27
Day 14 15 16 17 18 19 20 21 22 23 24 25
Errors 21 24 20 24 26 20 18 30 21 22 23 24
Collect Data
Is Identify and
Process N remove special
Stable cause variation
Systematically
Is N investigate
Process common cause
Capable and improve the
process (DMAIC)
Is N Innovate or
Process totally
Flexible redesign
Standardize (SOP)
and Replicate
• Major Components
• Critical Process Operations and Steps
• Key Process Output Variables (primary and
secondary CTCs) with Targets and Specs
• Key Process Input Variables (KPIVs)
• Measurement System Methodology and Analysis
• Monitoring Methods and Responsibilities
Process Flowchart
Position
Mail
Step Coordi
Underwriting Corresp. Cap Mkts -nator
/Time
Obtain Loan
File from
B
U/W Queue
X2
Receives Docs
NT Obtain NT No for Suspended
Yes
Required? from Cap Approved? C /Conditioned
Markets Loan
X7
Underwrite Yes
Loan Resubmit
to U/W
X1 X3 X4 X5
No Suspend Fax Rec. Needs
Decisionable? Needs to Letter &
Loan seller Submit Docs
C
Yes
B
No Deny Lead Yes Counter No
Approved?
Loan Agrees? Offer?
X9
Update
System
Yes
No
Yes
Notify Retain
Submit Offer
Seller of File Copy
to Seller
Declination
Yes Receive
Fax
Conditions? Condition Send Original
Conditions Request & X6
to seller
Provide Docs
to Seller
X8
No Acceptable?
No B
Input in Yes
System
Fax
Receive Y1
Approval to
Approval
seller
X8 = Counter offer cycle time = date Every Loan with Weekly U / W Lead
counter offer accepted / rejected – Counter Offer
date offer made.
IMR Chart
VIDEO
SORT
Red-tag
Do you see….
• Unnecessary things cluttering up
your workspace
• Tools or materials left around
• Paperwork lying around that isn’t
required
• Unnecessary duplicates of
anything
• Tools or office equipment that isn’t
properly classified and stored
5S SORT Conditions
1. Record the date found
2. Record the condition from
Obsolete (Unneeded)
choices at right
3. Record location found
Defective (Fixable)
4. List name or description of item
5. Record your name or initials
(Unfixable)
6. Tag items to be moved
7. Place items, other than trash, in
Scrap (Reclamation)
Red Tag Holding Area
8. Set specific date for
final disposition Trash
(Used monthly)
Date_______________
Seldom used here
Condition___________
Location____________
Item_______________
Name______________
SET IN ORDER
In your workplace….
• Are aisles and storage
places clearly marked?
• Are workstations and
equipment clearly identified?
• Do current reports have a
place that is identified?
• Is anything stored around
fire extinguishers?
5S Set-in-order
1. Determine need first
2. Determine good home location. Ask workers
for input to determine safe, easy to use,
location.
3. Mark location, with visual factory principles in
mind.
4. Brief all operators, all shifts, to aid understanding
and compliance.
5. Include check of these locations into the work
area checklist.
Equipment
Molds/Dies _____ _____ _____ _____
Fixtures/Jigs _____ _____ _____ _____
Transport trays/racks _____ _____ _____ _____
Carts/dollies _____ _____ _____ _____
Hand tools/accessories _____ _____ _____ _____
Ancillary machines _____ _____ _____ _____
Supplies/Apparel
Pens/pencils/paper _____ _____ _____ _____
Documents/charts _____ _____ _____ _____
Helmets/shoes _____ _____ _____ _____
Coats/aprons _____ _____ _____ _____
Cleaning supplies _____ _____ _____ _____
SHINE
In your workplace….
• Look carefully at the floor,
aisles and around
machines. How much dirt,
oil, and dust can you find?
• Is there paperwork lying
around?
• Do the computers, desks, or
chairs require cleaning?
• Is there food or garbage just
lying around?
5S Shine
1. Use Shine Checklist
2. Determine items or areas to be cleaned
3. Mark location, with visual factory principles
in mind.
4. Brief all operators, all shifts, to aid
understanding and compliance.
5. Include check of these locations into the
work area checklist.
Equipment
Machine surfaces _____ _____ _____ _____
Fixtures/Jigs _____ _____ _____ _____
Transport trays/racks _____ _____ _____ _____
Carts/dollies _____ _____ _____ _____
Hand tools/accessories _____ _____ _____ _____
Ancillary machines _____ _____ _____ _____
Supplies/Apparel
Pens/pencils/paper _____ _____ _____ _____
Tool boxes/Tools _____ _____ _____ _____
Drawers/Cabinets _____ _____ _____ _____
Trash/Wastes bins _____ _____ _____ _____
Cleaning Equipment _____ _____ _____ _____
STANDARDIZE
SUSTAIN
Definition:
Innovative method of providing
valuable information to everyone!
• Improved Morale
20 30
10 FULL
40
0
HARD TODAY’S
HAT GOAL
AREA 6 0 0
PRODUCTION RESULTS
TODAY’S
SCHEDULE SOPs
RUN
1 3 7
QUALITY RESULTS
LOW
STATION 3
OPERATOR
FIRE SLEDGE
EXSTINGUISHER HAMMER CLEAN
GLOVES
DIRTY
GLOVES
WARNING
HOT STEEL
• COMMUNICATION OF RESULTS
– Company intranet and/or project database
– Newsletters
– Team meetings
– Department meetings
– Emails
– SOP’s
– Word of mouth
• To create a storyboard
– Use the DMAIC structure
– Describe each of the main steps (Define, Measure,
Analyze, Improve, and Control) and include 2-4 bullet
points for what you did/learned during that phase
– Include several graphics which will communicate the
key messages
Example Format
De A
De t. B
C
De .
case, etc.
pt.
pt.
tg
p
Mk
TEAM
IPO
Person 1 *
2 *
3 *
4 *
Current State
Measure • Discuss current Pareto of
Problems
state (baseline VALUE ADDED NON-VALUE ADDED
analysis)
16
12
histogram
Number of Cars
10
0
6 to <12 12 to <18 18 to <24 24 to <30 30 to <36 36 to <42 42 to <=48
Gas Mileage (mpg)
1
Analyze • Describe problems, Damage from
suppliers
2 scatterplot
Quality shipping
3
causes, and how issues
(defects)
company
Late
Shipments
• Analysis and 40
Information about Shipping Errors
focus on the
interpretation of data
35
25
20
15
10
0
Ship Late Packing Address Ship Early Contents Method
Implementation plan
Improve • Analysis of results ID TASK 1 2 3
DAY #
4 5 6 7
Before After
1.1
1.2
Assess the measurement
system capability
Select items to be
included in the study
Determine the sample
1.3
size
Take data
• Improvements made
1.4
(measurem ents)
Determine actions to be
1.6
taken
. .
. .
. .
• Implementation plan ( ) ( ) ( ) ( )
Solution #1
Solution #2
Solution #3
Solution #4
Total:
steps Future
150
140
130
improvement
• Updated SOPs, 120
110
100
PFs, etc. 90
80
17
41
13
21
23
25
27
31
33
37
45
1
11
15
19
29
35
39
43
47
Month Number
DMAIC SUMMARY
LEAN SIX SIGMA PROJECT MASTER
STRATEGY (DMAIC)
… the Roadmap
What is important to the
customer and the
business? Clearly
define the problem,
project, and team.
Define
Pr
oj thi
ec s
t s wa
uc y
ce
ss
…
Measure
I mprove
Control
Measure success. Plan for
holding the gains. Monitor
critical outputs to prevent
repeat problems.
Define
Define Team, Stakeholders,
Voice of the Customer
Tollgate Review
Understand Current
Process (baseline)
Measure
Tollgate Review
Analyze Causes of
Poor Performance
Tollgate Review
Improve
Test and Confirm
Solutions
Tollgate Review
No
OK
Yes
Monitor Performance;
Control
Measure Success
No
OK
Yes
Celebrate
Copyright Air Academy Associates 12 - 2
DEFINE
Define
Control
Build Knowledge /Project Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go
Yes
MEASURE
Close Down Project
Measure
Plan for data collection; Establish or
refine metrics if necessary
Update Knowledge/Project
Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
ANALYZE Close Down Project
Measure
MEASUREMENT METHOD MACHINE
Carburetor Transmission
Gage for
Ma
nu
Tire Press.
Tire Press.
to
al
Estimate of
Full Tank Maint. Sched. Tune-Ups Weight Tire Type
Ra
Re
gu
WX
Engine Warm-
Type Fast Start Up Time Radio
Air Conditioner
GAS
MILEAGE
Driver Gas Type Highway or City
Training
A
Weather Conditions
# of Ice
Oil Type
Passengers Snow
Terrain
Pr
A
Rain
Hu
ec
Te
ipi
mi
s
Wi
mp
ain
tat
dit
e
t
Fla
nd
tud
era
ion
y
unt
Driver
tur
Alti
Mo
Fuel Additive
Acctng
Identify the non-value added activities,
HR
Legal
sources of waste, and defects and/or Analyze
Functional
Area
Benefits
problems
Frequency or
Financial Measure
Control
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
IMPROVE Close Down Project
Copyright Air Academy Associates 12 - 5
IMPROVE
Define
170
150
140
130
120
110
100
90
80
validate improvement
1
11
13
15
17
19
21
23
25
27
29
31
33
35
37
39
41
43
45
47
Month Number
Analyze
Control
Update Knowledge/Project Notebook
No
Toll
Gate Review No Go
Kill Project?
Go/No Go
?
Go Yes
CONTROL Close Down Project
CENTERLINE
Monitor performance and measure
success
LCL
Measure
Performance No
confirmed and Gap Analysis
project goals
achieved? Analyze
Yes
Analyze
Improve
Measure
Analyze
Improve
Control
Measure Phase Completion Checklist
Measure
Analyze
Improve
Control
Analyze Phase Completion Checklist
Measure
Analyze
Improve
Control
Improve Phase Completion Checklist
Measure
Analyze
Improve
Control Phase Completion Checklist
Control
MEASURE
9 How well are we doing? What does the
current process look like and how is it
performing? (baseline) (current state)
9 What about our measurement process? Is it
accurate, repeatable, and reproducible?
9 What data do we need to collect?
9 What is our current process capability?
ANALYZE
9 What are the problems? Bottlenecks?
9 What are the sources of waste? non-value
added tasks?
9 What are the key input variables and output
variables in the process?
9 Where should we focus? What are the root
causes of the problems?
IMPROVE
9 Identify ways to optimize performance and
minimize variability / waste
9 Investigate and identify candidate solutions
9 Test out solutions and mistake proof the
process
9 Re-assess the capability of the new process
9 Define new process flows, SOP’s, etc.
CONTROL
9 Measure Success. Quantify the gains. ($,
cycle time, performance)
9 Put a plan in place to hold the gains; determine
accountability; monitor the process
9 Train team on updated process flow and SOP’s
9 Communicate results, share lessons learned,
document future plans and recommendations
9 Create a storyboard/summary of the project
Understand the Lean Six Sigma Strategy and DMAIC, Knowledge Notebook 1
• • • • •
Roadmap
Plan for documentation Knowledge Notebook 1, 3 • •
Lead effective meetings and enhance teamwork “PACER”; team building 3 • • • • •
Capture the Voice of the Customer QFD, House of Quality 4 • •
Define and analyze key project stakeholders Stakeholder analysis 3 •
Define project goals and work on problem SMART goals, problem 3
•
statements and scoping statements, project charter
Define and quantify the Cost of Poor Quality COPQ, 7 wastes 2, 7 • •
Define and understand your process IPO diagram, SIPOC, value 2, 4, 5
stream map, time value and
• • •
physical flow maps,
PF/CE/CNX/SOP
Describe variables data graphically Histogram, box plot, run chart 5 • • •
Describe categorical data graphically Pareto diagram 5 • • •
Summarize variables data numerically Mean, std dev., sigma level, 2, 5, 8
• • •
Cp, Cpk, confidence interval
Summarize attribute data numerically FPY, dpu, dpmo, sigma 2, 5, 8
• • •
capability, confidence interval
Prioritize data IPO Matrix, FMEA, Pareto, 3, 8
Effort Impact analysis, voting, • • • •
pairwise comparison
Look for relationships with variables data Correlation coefficient, scatter 5
• •
diagram, regression
Look for relationships with count data Chi Square Test for 9
• •
independence
Test for significant shifts in the mean t-test 9 • •
Test for significant shifts in the standard deviation F-test 9 • •
Test for significant shifts with proportion data Test for proportions 9 • •
Assess the capability of a process Cp, Cpk 5 • • •
Evaluate your measurement system to see if you MSA 6
• •
have integrity in the data
Look for trends in the data over time Run chart 5, 11
• • •
Control chart
Reduce changeover time SMED 10 • •
Reduce extraneous variation and waste PF/CE/CNX/SOP 2, 7 • • • •
Organize the workplace and set up visual controls 5S, work cell design, visual 10, 11
• •
controls
Determine sample size for estimating means and Sample size (normal and 8
• • •
proportions binomial distributions)
Develop a plan for sustaining the gains made on Control plans 11
•
your project
Mistake proof your process FMEA; poka yoke 10 •
Determine if your process is stable and repeatable Control charts 11 • • •
DESIGN OF EXPERIMENTS
¾ Barker, Thomas B., Quality By Experimental Design, 2nd ed., Marcel
Dekker Inc., 1994.
¾ Schmidt, Stephen R. and Launsby, Robert G., Understanding Industrial
Designed Experiments, 4th ed., Air Academy Press, 1998.
¾ Taguchi, G. and Konishi, S., Taguchi Methods: Orthogonal Arrays and
Linear Graphs, American Supplier Institute, Inc., 1987.
¾ Advanced DOE:
¾ Box, George E. P., Hunter, William G., and Hunter, J. Stuart, Statistics for
Experimenters, John Wiley & Sons, 1978.
¾ Box, George E. P. and Draper, Norman R., Empirical Model-Building and
Response Surfaces, John Wiley & Sons, 1987.
¾ Hicks, Charles R., Fundamental Concepts in the Design of Experiments, 5th
ed., CBS College Publishing, Holt, Rinehart, and Winston, 1999.
¾ Montgomery, Douglas C., Design and Analysis of Experiments, 5th ed.,
John Wiley & Sons, 2000.
BENCHMARKING
¾ Camp, Robert C., Benchmarking, ASQ Quality Press, 1989.
QUALITY (GENERAL)
¾ Brassard, Michael and Ritter, Diane, The Memory Jogger II (Pocket Guide of
Tools for Continuous Improvement and Effective Planning), Goal/QPC, 1994.
¾ Deming, W. Edwards, Out of the Crisis, MIT Center for Advanced Engineering
Study, 1986.
¾ Ishikawa, Kaoru, Guide to Quality Control, Asian Productivity Organization,
1982.
¾ Juran, J. M., Juran on Planning for Quality, The Free Press, 1988.
¾ Lam, K. D., Watson, Frank D., and Schmidt, Stephen R., Total Quality: A
Textbook of Strategic Quality, Leadership & Planning, Air Academy Press,
1991.
¾ Walton, Mary, The Deming Management Method, The Putnam Publishing
Group, 1986.
SPC
¾ Grant, Eugene L. and Leavenworth, Richard S., Statistical Quality Control,
McGraw Hill, 1988.
¾ Wheeler, Donald J. and Chambers, David S., Understanding Statistical
Process Control, 2nd ed., SPC Press, 1992.
FMEA
¾ Automotive Industry Action Group (AIAG), Potential Failure Mode and
Effects Analysis – Reference Manual, 3rd ed., 2001.
¾ IEC Standard Publication 812, Analysis Techniques for System Reliability –
Procedure for Failure Mode and Effects Analysis, 1985.
¾ Stamatis, D. H., Failure Mode and Effect Analysis, ASQC Quality Press,
1995.
ADVANCED STATISTICS/TOOLS
¾ Anderson, T. W., An Introduction to Multivariate Statistical Analysis, 3rd ed.,
John Wiley & Sons, 2003.
¾ Draper, Norman R. and Smith, H., Applied Regression Analysis, 3rd ed.,
John Wiley & Sons, 1998.
¾ Johnson, Richard A. and Wichern, Dean W., Applied Multivariate Statistical
Analysis, 4th ed., Prentice-Hall, Inc., 1998.
¾ Hocking, Ronald R., The Analysis of Linear Models, Brooks/Cole Publishing
Co., 1985.
¾ Myers, Raymond H., Classical and Modern Regression with Applications,
2nd ed., Duxbury Press, 1990.
¾ Neter, John, Wasserman, William, and Kutner, Michael, Applied Linear
Statistical Models, 4th ed., Richard D. Irwin, Inc., 1996.
¾ Ostle, Bernard and Malone, Linda C., Statistics in Research, 4th ed., Iowa
State University Press, 1987.
¾ Seber, G. A. F., Linear Regression Analysis, 2nd ed., John Wiley & Sons,
2003.
Reference
Term/Tool What It Stands For Description
(module)
Data which is not measured on a continuous
attribute data 2 good/bad, pass/fail data
scale. (Qualitative).
Reference
Term/Tool What It Stands For Description
(module)
Individual Moving Range A control chart for variables data with individual
IMR 11
Chart measurements (no subgroups)
Reference
Term/Tool What It Stands For Description
(module)
Method for analyzing the variation present in a
Measurement System
MSA 6 measurement system and assessing its
Analysis
capability to make good measurements.
Reference
Term/Tool What It Stands For Description
(module)
Reference
Term/Tool What It Stands For Description
(module)
The square of the correlation coefficient. It
Also called coefficient of represents the strength of a regression model
R-squared metric
determination by measuring the proportion of variation in the
data which is accounted for by the model.
Reference
Term/Tool What It Stands For Description
(module)
A measure of variability or spread in a
standard deviation standard deviation
2, 5 distribution. We try to reduce variation using
(σ) (pronounced "sigma")
PF/CE/CNX/SOP, statistical analysis, etc.
The average time to produce a unit in order to
Takt time 7
meet the current rate of customer demand.
A comprehensive and coordinated maintenance
program designed to maximize equipment
effectiveness by minimizing downtime and
optimizing output in terms of speed and quality.
Total Productive
TPM This approach is founded on the ability of well
Maintenance
trained equipment operators to proactively
identify and correct small maintenance issues
before they become significant and lead to
breakdowns.
A sequence of processes or activities to include
all of the actions (both value added and non-
value stream 2, 5 value added) required to bring a product or
service from concept to raw materials all the
way to the customer.
Features, steps, etc. that the customer
considers desirable. Value is measured in
VA 2, 7 value-added
terms of what an external customer is willing to
pay for.
A dimensional measurement can be obtained,
variables data 2 measurement data and is only limited in value by the resolution of
the measurement system.
Any activity or process that does not add value
to the product or service from the perspective of
the customer. The seven “classic wastes”,
waste 2, 7 originally identified by Taiichi Ohno, include:
defects, over processing, overproduction,
unnecessary movement, unnecessary
transportation, inventory, and waiting.
sigma sigma
dpmo dpmo
capability* capability*
500,000 1.5 6,210 4.0
460,172 1.6 4,661 4.1
420,740 1.7 3,467 4.2
382,089 1.8 2,555 4.3
344,578 1.9 1,866 4.4
308,538 2.0 1,350 4.5
274,253 2.1 968 4.6
241,964 2.2 687 4.7
211,855 2.3 483 4.8
184,060 2.4 337 4.9
158,655 2.5 233 5.0
135,666 2.6 159 5.1
115,070 2.7 108 5.2
96,801 2.8 72 5.3
80,757 2.9 48 5.4
66,807 3.0 32 5.5
54,799 3.1 21 5.6
44,565 3.2 13 5.7
35,930 3.3 9 5.8
28,716 3.4 5 5.9
22,750 3.5 3.4 6.0
17,864 3.6
13,903 3.7
10,724 3.8
8,198 3.9
WORKSHOP NAME: LEAN SIX SIGMA BLACK BELT (WEEK 2) DATE: _______________________
POOR EXCELLENT
A. COURSE OBJECTIVES 1 2 3 4 5
B. INSTRUCTOR 1 2 3 4 5
C. INTERACTION AND DISCUSSION 1 2 3 4 5
D. TOPICS COVERED 1 2 3 4 5
E. PARTICIPANT'S GUIDE 1 2 3 4 5
F. GROUP EXERCISES 1 2 3 4 5
G. MEETING ROOM ACCOMMODATIONS 1 2 3 4 5
1 2 3 4 5
H. WORKSHOP DURATION
TOO LONG OK TOO SHORT
4. DO YOU HAVE SUGGESTIONS THAT WOULD MAKE THE COURSE MORE EFFECTIVE?
5. TO WHOM WOULD YOU RECOMMEND THIS WORKSHOP? (PLEASE INCLUDE NAME, ADDRESS,
AND PHONE NUMBER. WE WILL NOTIFY THESE PEOPLE OF FUTURE OFFERINGS.)
WORKSHOP NAME: LEAN SIX SIGMA BLACK BELT (WEEK 1) DATE: _______________________
POOR EXCELLENT
A. COURSE OBJECTIVES 1 2 3 4 5
B. INSTRUCTOR 1 2 3 4 5
C. INTERACTION AND DISCUSSION 1 2 3 4 5
D. TOPICS COVERED 1 2 3 4 5
E. PARTICIPANT'S GUIDE 1 2 3 4 5
F. GROUP EXERCISES 1 2 3 4 5
G. MEETING ROOM ACCOMMODATIONS 1 2 3 4 5
1 2 3 4 5
H. WORKSHOP DURATION
TOO LONG OK TOO SHORT
4. DO YOU HAVE SUGGESTIONS THAT WOULD MAKE THE COURSE MORE EFFECTIVE?
5. TO WHOM WOULD YOU RECOMMEND THIS WORKSHOP? (PLEASE INCLUDE NAME, ADDRESS,
AND PHONE NUMBER. WE WILL NOTIFY THESE PEOPLE OF FUTURE OFFERINGS.)
3. Is the pace of the course too fast, too slow, or about right?