Sie sind auf Seite 1von 64

Introduction to Six Sigma

Topics (Session 1)
Understanding Six Sigma
History of Six Sigma
Six Sigma Methodologies & Tools
Roles & Responsibilities
How YOU can use Six Sigma

Six Sigma is. . .


A performance goal, representing 3.4 defects for every million

opportunities to make one.

A series of tools and methods used to improve or design products,

processes, and/or services.

A statistical measure indicating the number of standard deviations within

customer expectations.

A disciplined, fact-based approach to managing a business and its

processes.

A means to promote greater awareness of customer needs, performance

measurement, and business improvement.

Whats in a name?
Sigma is the Greek letter representing the standard

deviation of a population of data.


Sigma is a measure

of variation
(the data spread)

What does variation mean?


20

Variation means that a

process does not produce


the same result (the Y)
every time.
Some variation will exist in

all processes.

15

10

-5

-10

Variation directly affects customer experiences.

Customers do not feel averages!

Measuring Process Performance


The pizza delivery example. . .
Customers want their pizza

delivered fast!

Guarantee = 30 minutes or less

What if we measured performance and found an average

delivery time of 23.5 minutes?

On-time performance is great, right?


Our customers must be happy with us, right?

How often are we delivering on


time?
Answer: Look at
the variation!

30 min. or less

x
0
40 The average
50 and
Managing
by 10
the average20
doesnt tell 30
the whole story.
the variation together show whats happening.

Reduce Variation to Improve


Performance
How many standard
deviations can you
fit within
customer
expectations?

10

20

30 min. or less

30

40

50

Sigma level measures how often we meet (or fail to meet) the

requirement(s) of our customer(s).

Managing Up the Sigma Scale


Sigma
1
2
3
4
5
6

% Good % Bad

DPMO

30.9%

69.1%

691,462

69.1%

30.9%

308,538

93.3%

6.7%

66,807

99.38%

0.62%

6,210

99.977%

0.023%

233

99.9997% 0.00034%

3.4

Examples of the Sigma Scale


In a world at 3 sigma. . .

In a world at 6 sigma. . .

There are 964 U.S. flight

1 U.S. flight is cancelled every 3

The police make 7 false arrests

There are fewer than 4 false

In MA, 5,390 newborns are

1 newborn is dropped every 4

In one hour, 47,283 international

It would take more than

cancellations per day.

every 4 minutes.

dropped each year.

long distance calls are


accidentally disconnected.

weeks.

arrests per month.

years in MA.

2 years to see the same number


of dropped international calls.

Topics
Understanding Six Sigma
History of Six Sigma
Six Sigma Methodologies & Tools
Roles & Responsibilities
How YOU can use Six Sigma

The Six Sigma Evolutionary Timeline


1818: Gauss uses the normal
curve to explore the
mathematics of error analysis for
measurement, probability
analysis, and hypothesis testing.
1736: French
mathematician
Abraham de
Moivre
publishes an
article
introducing the
normal curve.

1924: Walter A. Shewhart


introduces the control chart and
the distinction of special vs.
common cause variation as
contributors to process problems.
1896: Italian sociologist Vilfredo
Alfredo Pareto introduces the
80/20 rule and the Pareto
distribution in Cours dEconomie
Politique.

1949: U. S. DOD issues Military


Procedure MIL-P-1629,
Procedures for Performing a
Failure Mode Effects and
Criticality Analysis.

1960: Kaoru Ishikawa


introduces his now
famous cause-andeffect diagram.

1941: Alex Osborn, head


of BBDO Advertising,
fathers a widely-adopted
set of rules for
brainstorming.

1970s: Dr. Noriaki Kano


introduces his twodimensional quality model
and the three types of
quality.

1986: Bill Smith, a senior


engineer and scientist
introduces the concept of Six
Sigma at Motorola

1995: Jack Welch


launches Six Sigma at
GE.
1994: Larry Bossidy
launches Six Sigma at
Allied Signal.

Six Sigma Companies

Six Sigma and Financial Services

Topics
Understanding Six Sigma
History of Six Sigma
Six Sigma Methodologies & Tools
Roles & Responsibilities
How YOU can use Six Sigma

DMAIC The Improvement


Methodology
Define

Measure Analyze Improve Control

Objective:
DEFINE the
opportunity

Objective:
Objective:
MEASURE current ANALYZE the
performance
root causes of
problems

Objective:
IMPROVE the
process to
eliminate root
causes

Objective:
CONTROL the
process
to sustain the gains.

Key Define Tools:


Cost of Poor
Quality (COPQ)
Voice of the
Stakeholder
(VOS)
Project Charter
As-Is Process
Map(s)
Primary Metric
(Y)

Key Measure
Tools:
Critical to Quality
Requirements
(CTQs)
Sample Plan
Capability
Analysis
Failure Modes
and Effect
Analysis (FMEA)

Key Improve
Tools:
Solution
Selection Matrix
To-Be Process
Map(s)

Key Control
Tools:
Control Charts
Contingency
and/or Action
Plan(s)

Key Analyze
Tools:
Histograms,
Boxplots, MultiVari Charts, etc.
Hypothesis Tests
Regression
Analysis

Define DMAIC Project


What is the project?
$

Project
Charter

Cost of
Poor
Quality

S ta k e h o ld e r s

Voice of
the
Stakeholde
r

Six Sigma
What is the problem? The problem is the Output (a Y

in a math equation Y=f(x1,x2,x3) etc).


What is the cost of this problem
Who are the stake holders / decision makers
Align resources and expectations

Define As-Is Process


How does our existing process work?
Move-It! Courier Package Handling
Process
Courier

Mail Clerk

In-SortClerk

In-SortSupervisor DistanceFeeClerk

WeightFeeClerk

Accounts
ReceivableClerk

Accounts
Supervisor

Out-SortClerk

Out-SortSupervisor

Observ e package
weight (1 or 2) on
back of package

Look up
appropriate
Weight Fee and
write in top middle
box on package
back
Add Distance &
Weight Fees
together and write
in top right box on
package back

Accounting

Take packages
f rom Weight Fee
Clerk Outbox to
A/R Clerk Inbox.

Take packages
f rom A/R Clerk
Outbox to
Accounts
Superv isorInbox.

Does EVERYONE
agree how the current
process works?

Circle Total Fee


and Draw Arrow
f rom total to
sender code

Write Total Fee


f rom package in
appropriate
Sender column on
Accts. Supv .s log

Take packages
f rom Accounts
Superv isor
Outbox to OutSort Clerk Inbox.

Delivery

Finalizing

Take packages
f rom Out-Sort
Clerk Outbox to
Out-Sort
Superv isorInbox.

Draw 5-point Star


in upper right
corner of package
f ront

Define the Non Value


Add steps

Sort packages in
order of Sender
Code bef ore
placing in outbox

Add up Total # of
Packages and
Total Fees f rom
log and create
client inv oice

Observ e sender
and receiv er
codes and make
entry in Out-Sort
Superv isors log

Deliv erPackages
to customers
according to N, S,
E, W route

Deliv er inv oiceto


client

Submit log to
General Manager
at end of round

Submit log to
General Manager
at conclusion of
round.

Submit log to
General Manager
at end of round

Define Customer Requirements


What are the CTQs? What motivates the customer?

Listening
Posts

Industry
Intel

SECONDARY RESEARCH
Market
Data

Customer
Service

Industry
Benchmarking

Customer
Correspondence

PRIMARY RESEARCH
Surve
ysSurve
ys

OTM

Focus Groups

Observations

Voice of the Customer

Key Customer Issue

Critical to Quality

Measure Baselines and


Capability
our
What
Sample is
some
data /current
not all data level of performance?
Descriptive Statistics

Current Process actuals measured against


the Customer expectation
What is the chance that we will succeed
at this level every time?

Variable: 2003 Output


Anderson-Darling Normality Test
A-Squared:
P-Value:

10

Pareto Chart for Txfr Defects

20

30

40

50

95% Confidence Interval for Mu

Mean
StDev
Variance
Skewness
Kurtosis
N

23.1692
10.2152
104.349
0.238483
0.240771
100

Minimum
1st Quartile
Median
3rd Quartile
Maximum

0.2156
16.4134
23.1475
29.6100
55.2907

95% Confidence Interval for Mu


21.1423
19.5

100

22.5

23.5

24.5

60
50
40
20
0

Defect
79
79.0
79.0

17
17.0
96.0

4
4.0
100.0

25.5

26.5

25.1961

95% Confidence Interval for Sigma


8.9690

Percent

Count

21.5

95% Confidence Interval for Median

80

Count
Percent
Cum %

20.5

100

0.211
0.854

11.8667

95% Confidence Interval for Median


19.7313

26.0572

Measure Failures and Risks


Where does our process fail and why?
Subjective opinion mapped into an objective risk profile number

X1
X2
X3
X4
etc

Analyze Potential Root Causes


What affects our process?
Ishikawa
Diagram
(Fishbone)

Six Sigma

y = f (x1, x2, x3 . . . xn)

Analyze Validated Root Causes


What are the key root causes?
Pareto Chart for Txfr Defects

100

100

60
40

Percent

Count

80

50

20
0

E x p e r im e n ta l D e s ig n

Defect
Count
Percent
Cum %

79
79.0
79.0

Data
Stratificatio
n

17
17.0
96.0

4
4.0
100.0

Regression
Analysis

Pareto Chart for Amt Defects


100

15

60

10

40
5
20
0

Percent

Count

80

Process
Simulatio
n

Defect
Count
Percent
Cum %

12
70.6
70.6

3
17.6
88.2

2
11.8
100.0

Six Sigma

y = f (x1, x2, x3 . . . xn)


Critical Xs

Improve Potential Solutions


How can we address the root causes we identified?
Address the causes, not the symptoms.

Evaluate

Clarify

Generate

y = f (x1, x2, x3 . . . xn)


Critical Xs

Divergent | Convergent

Decision

Improve Solution Selection


How do we choose the best solution?
Solution Selection Matrix
Qualit
y

Time

Solution

Sigma

Cost

Six Sigma
Implementation
Bad
Good

Solution
Right
Wrong

Nice
Try

Nice
Idea

Solution
Implementatio
n Plan

Time

CBA

Other

Score

Control Sustainable Benefits


How do we hold the gains of our new process?
Some variation is normal and OK
How High and Low can an X go yet not materially impact the Y
Pre-plan approach for control exceptions

35

Individual Value

UCL=33.48

25

Mean=24.35

LCL=15.21

15
0

10

20

Observation Number

30

DFSS The Design Methodology


Design for Six Sigma
Define

Measure Analyze Develop

Verify

Uses
Design new processes, products, and/or services from scratch
Replace old processes where improvement will not suffice

Differences between DFSS and DMAIC


Projects typically longer than 4-6 months
Extensive definition of Customer Requirements (CTQs)
Heavy emphasis on benchmarking and simulation; less emphasis on baselining

Key Tools
Multi-Generational Planning (MGP)
Quality Function Deployment (QFD)

Topics
Understanding Six Sigma
History of Six Sigma
Six Sigma Methodologies & Tools
Roles & Responsibilities
How YOU can use Six Sigma

Champions
Promote awareness and execution of Six Sigma

within lines of business and/or functions

Identify potential Six Sigma projects to be executed

by Black Belts and Green Belts

Identify, select, and support Black Belt and

Green Belt candidates

Participate in 2-3 days of workshop training

Black Belts
Use Six Sigma methodologies and advanced tools

(to execute business improvement projects


Are dedicated full-time (100%) to Six Sigma
Serve as Six Sigma knowledge leaders within

Business Unit(s)
Undergo 5 weeks of training over 5-10 months

Green Belts
Use Six Sigma DMAIC methodology and basic tools to

execute improvements within their existing job


function(s)

May lead smaller improvement projects within Business

Unit(s)

Bring knowledge of Six Sigma concepts & tools to their

respective job function(s)

Undergo 8-11 days of training over 3-6 months

Other Roles
Subject Matter Experts
Provide specific process knowledge to Six Sigma teams
Ad hoc members of Six Sigma project teams
Financial Controllers
Ensure validity and reliability of financial figures used
by Six Sigma project teams
Assist in development of financial components of
initial business case and final cost-benefit analysis

Topics
Understanding Six Sigma
History of Six Sigma
Six Sigma Methodologies & Tools
Roles & Responsibilities
How YOU can use Six Sigma

Questions?

Topics for Detailed Discussion

Problem Identification
Cost of Poor Quality
Problem Refinement
Process Understanding
Potential X to Critical X
Improvement

Problem Identification
If it aint broke, why fix it
This is the way weve always done it

Problem Identification
First Pass Yield
Roll Throughput Yield
Histogram
Pareto

Problem Identification
First Pass Yield (FPY):
The probability that
any given unit can go
through a system
defect-free without
rework.

100 Units
Step 1

Outputs / Inputs
100 / 100 = 1

100

Scrap 10 Units

Step 2

90 / 100 = .90

90
Scrap 3 Units

Step 3

87 / 90 = .96

87
Scrap 2 Units

At first glance, the yield would seem to be


85% (85/100 but.)

Step 4

85

85 / 87 = .97

When in fact the FPY is (1 x .90 x .96 x .97 = .


838)

Problem Identification
Rolled
Throughput
Yield (RTY):
The yield of
individual
process steps
multiplied
together.
Reflects the
hidden factory
rework issues
associated with
a process.

Re-Work
10 Units

100 Units

Outputs / Inputs

Step 1

90 / 100 = .90

100 Units

97 / 100 = .97

Step 2
Re-Work
3 Units

100 Units

98 / 100 = .98

Step 3
Re-Work
2 Units

100 Units

Step 4

100 Units

.90 x .97 x .98 = .855

Problem Identification
RTY Examples - Widgets
50

Roll Throughput Yield


Function 1

50/50 = 1
(50-5)/50 = .90

50
Function 2

(50-10)/50 = .80
5

(50-5)/50 = .90

50
Function 3
10

1 x .90 x .80 x .90 = .65

50
Function 4
5
50

Put another way, this process is operating


a 65% efficiency

Problem Identification
RTY Example - Loan Underwriting
50

Roll Throughput Yield


50/50 = 1

Application

50
Underwrite

7
Fails
Underwriting

(50-7-2)/50 = .82
(43-6)/43 = .86
(43-1-2)/43 = .93

43
Complete Full
Paperwork

43
Close

42

1 x .82 x .86 x .93 = .66


1
Decide not to
borrow

Put another way, this process is operating


a 66% efficiency

Problem Identification
Histogram A histogram is a basic graphing tool that displays the
relative frequency or occurrence of continuous data values showing
which values occur most and least frequently. A histogram illustrates the
shape, centering, and spread of data distribution and indicates whether
there are any outliers.
Histogram of Cycle Time
40

Frequency

30

20

10

0
0

100

200

300

C8

400

500

Problem Identification
Histogram Can also help us graphically understand the data

Descriptive Statistics
Variable: CT
Anderson-Darling Normality Test
A-Squared:
P-Value:

25

100

175

250

325

400

95% Confidence Interval for Mu

6.261
0.000

Mean
StDev
Variance
Skewness
Kurtosis
N

80.1824
67.6003
4569.81
2.31712
8.26356
170

Minimum
1st Quartile
Median
3rd Quartile
Maximum

1.000
31.000
66.000
105.000
444.000

95% Confidence Interval for Mu


69.947
54

64

74

84

94

95% Confidence Interval for Sigma


61.098

95% Confidence Interval for Median

90.417
75.664

95% Confidence Interval for Median


55.753

84.494

Problem Identification
Pareto The Pareto principle states that 80% of the impact of the
problem will show up in 20% of the causes. A bar chart that displays by
frequency, in descending order, the most important defects.
Pareto Chart for WEB
100
100

60
50

40
20

Defect
Count
Percent
Cum %

96
86.5
86.5

15
13.5
100.0

Percent

Count

80

Topics (Session 2)

Problem Identification
Cost of Poor Quality
Problem Refinement
Process Understanding
Potential X to Critical X
Improvement

Cost of Poor Quality


COPQ - The cost involved in fulfilling the gap between the desired and
actual product/service quality. It also includes the cost of lost opportunity
due to the loss of resources used in rectifying the defect.
Hard Savings - Six Sigma project benefits that allow you to do the same
amount of business with less employees (cost savings) or handle more
business without adding people (cost avoidance).
Soft Savings - Six Sigma project benefits such as reduced time to market,
cost avoidance, lost profit avoidance, improved employee morale,
enhanced image for the organization and other intangibles may result in
additional savings to your organization, but are harder to quantify.
Examples / Buckets
Roll Throughput Yield Inefficiencies (GAP between desired result and
current result multiplied by direct costs AND indirect costs in the process).
Cycle Time GAP (stated as a percentage between current results and
desired results) multiplied by direct and indirect costs in the process.
Square Footage opportunity cost, advertising costs, overhead costs, etc

Topics (Session 2)

Problem Identification
Cost of Poor Quality
Problem Refinement
Process Understanding
Potential X to Critical X
Improvement

Problem Refinement
Multi Level Pareto Logically Break down initial Pareto data into subsets (to help refine area of focus)

Pareto Chart for WEB


100
100

60
50

40

Percent

Count

80

20
0

Pareto Chart for Type

Defect

100
100

15
13.5
100.0

80
60
50

40
20

Defect
Count
Percent
Cum %

45
41.3
41.3

35
32.1
73.4

13
11.9
85.3

16
14.7
100.0

Percent

96
86.5
86.5

Count

Count
Percent
Cum %

Problem Refinement
Problem Statement A crisp description of what we are trying to solve.
Primary Metric An objective measurement of what we are attempting
to solve (the y in the y = f(x1, x2, x3.) calculation).
Secondary Metric An objective measurement that ensures that a Six
Sigma Project does not create a new problem as it fixes the primary
problem. For example, a quality metric would be a good secondary
metric for an improve cycle time primary metric.

Problem Refinement
Fish Bone Diagram - A tool used to solve quality problems by
brainstorming causes and logically organizing them by branches. Also
called the Cause & Effect diagram and Ishikawa diagram

Provides tool for exploring cause / effect and 5 whys

Topics (Session 2)

Problem Identification
Cost of Poor Quality
Problem Refinement
Process Understanding
Potential X to Critical X
Improvement

Process Understanding
SIPOC Suppliers, Inputs, Process, Outputs, Customers
You obtain inputs from suppliers, add value through your process, and provide
an output that meets or exceeds your customer's requirements.

Process Understanding
Process Map should allow people unfamiliar with the process to understand
the interaction of causes during the work-flow. Should outline Value Added
(VA) steps and non-value add (NVA) steps.

Receipt /
Extract

Start

Size Sorts

Control
Docs

Full Form

Open

Pull & Sort

Ck / Vouch
Verify

Perfection

Requal Group

No
Remit

Yes
Pass 1

Pass 2

Rulrs

Prep cks,
route
vouch

Prep cks

Ship to IP

Vouchers

Data Cap

Key from
image

Balance

Vouch
OK
Inventory

No

Yes
Prep
Folders /
Box

Full Form
QCReview

Ship to
Cust

Process Understanding
Action
Plan

Create daily peak


staff need plan

No
Add 30% to
the required
no.

Operations

Manually
Update HR
Billet Request

Create
Staff
Billet

To Floor

HR /
Recruit

HR sends
req for
staffing
nos.

Call employee
(3x)

Can they
make it?

No

Review
Staff
Billet

Compare to
original Billet rpt

Check off
desired
returnee
staff & "need
to retrain"
list

Add 40% to
staff needed

What if the
returnee is
already
working here
on another
program?
Currently
send the ltr
anyways

Need OJT
Re-Tng

Yes

Make
it?

OJT

No

Yes
No

Need re
-train

Yes

To Floor

Call (3x)

Stop!
Update
IPS

No
Send Letters
to desired
staff

Start

Yes

Do they
respond?

Yes

Do they
want to
work this
peak?

No

No

Do they
want to
stay on the
list

Yes

Rev
original
billet &
call
uncheck
ed

Have we No
hired
enough?

Yes

Do they
want to
work this
peak?

No

No

Stop!

Do they
want to
stay on the
list

Yes

Have we No
Call Wait
hired
List
enough?

No

No
New

Yes Interview /

Meet Fleet
hiring
criteria

Rank as
"1 2 3"

pre-hire

Hire in 12 order
(3's are
not
placed)

Yes

Stop!

Yes
Place into
dept

Set 14
month
flag (on
IPS?)

Call
3X

Yes

Wait List

Yes

Take off
IPS
system

No

show up
orienta
tion

Yes
New &
Other
People
call in

Take off
IPS
system

Set 14
month
flag (on
IPS?)

schedule
for
training

Show
up?

Yes

Reach

No

Call
1X

No

Gen Event Roster


rpt in IPS

Notify
HR

Yes

Training

Gen rpt for


Ops Kronos
Recruit

Train

No

Pass?

Yes

Update
IPS

Topics (Session 2)

Problem Identification
Cost of Poor Quality
Problem Refinement
Process Understanding
Potential X to Critical X
Improvement

Potential X to Critical X
Y is the dependent output of a variable process. In other
words, output is a function of input variables (Y=f(x1, x2,
x3).
Through hypothesis testing, Six Sigma allows one to
determine which attributes (basic descriptor (generally
limited or binary in nature) for data we gather ie. day of
the week, shift, supervisor, site location, machine type,
work type, affect the output. For example, statistically,
does one shift make more errors or have a longer cycle
time than another? Do we make more errors on Fridays
than on Mondays? Is one site faster than another? Once we
determine which attributes affect our output, we determine
the degree of impact using Design of Experiment (DOE).

Potential X to Critical X
A Design of Experiment (DOE) is a structured, organized
method for determining the relationship between factors
(Xs) affecting a process and the output of that process (Y).
Not only is the direct affect of an X1 gauged against Y but
also the affect of X1 on X2 against Y is also gauged. In
other words, DOE allows us to determine - does one input
(x1) affect another input (x2) as well as Output (Y).

Potential X to Critical X
DOE Example

Main Effects Plot (data means) for Elapsed

Main Effects Plot


Direct impact to Y

1.4

Elapsed

1.3

1.2

1.1

1.0
Jams

DCDEL

SK

P2Jam

Interaction Plot (data means) for Elapsed


1.50

Jams
1

1.25

1.00
1.50

DCDEL

Interaction Plot
Impacts of Xs on
each other

1.25

1.00
1.50

SK
3

1.25

1.00

P2Jam

1.50

1.25

1.00

Potential X to Critical X
DOE Optimizer
Allows us to
statistically predict the
Output (Y) based on
optimizing the inputs
(X) from the Design of
experiment data.

Topics (Session 2)

Problem Identification
Cost of Poor Quality
Problem Refinement
Process Understanding
Potential X to Critical X
Improvement

Improvement
Once we know the degree to which inputs (X) affect our
output (Y), we can explore improvement ideas, focusing
on the cost benefit of a given improvement as it relates
to the degree it will affect the output. In other words, we
generally will not attempt to fix every X, only those that
give us the greatest impact and are financially or
customer justified.

Control
Once improvements are made, the question becomes, are the
improvement consistent with predicted Design of Experiment
results (ie are they what we expected) and, are they statistically
different than pre-improvement results.

Process Capability Analysis for Sept


Process Data
USL
0.23000
Target
*
LSL
-1.00000
Mean
-0.02391
Sample N
23
StDev (Within) 0.166425
StDev (Overall) 0.221880

LSL

USL
Within
Overall

Potential (Within) Capability


Z.Bench
1.53
Z.USL
1.53
Z.LSL
5.87
Cpk
0.51
Cpm

Overall Capability
Z.Bench
1.14
Z.USL
1.14
Z.LSL
4.40
Ppk
0.38

-1.0

-0.5

Observed Performance
% < LSL
0.00
% > USL
13.04
% Total
13.04

0.0

0.5

Exp. "Within" Performance


% < LSL
0.00
% > USL
6.35
% Total
6.35

1.0
Exp. "Overall" Performance
% < LSL
0.00
% > USL
12.62
% Total
12.62

Control
Control Chart - A graphical tool for monitoring changes that occur
within a process, by distinguishing variation that is inherent in the
process(common cause) from variation that yields a change to the
process(special cause). This change may be a single point or a series
of points in time - each is a signal that something is different from
what was previously observed and measured.
I and MR Chart for Sept
Individual Value

1
UCL=0.5293

0.5

Mean=0.03

0.0
2

LCL=-0.4693

-0.5
Sept 13

Sept 20

Date

9/13

9/25

Moving Range

Subgroup
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0

UCL=0.6134

R=0.1877
LCL=0

Wrap Up