Six Sigma and Its Implementation

SIX SIGMA AND ITS
IMPLEMENTATION ON
THE PROJECT

Six Sigma

Six Sigma
6 sigma is used by individual and
organizations to:
•Drive and sustain improvements
•Provide rigorous alignment of actions
with strategy
•Guide decision making with facts and
data
•Meet customer needs through improved
products and processes
•Deliver bottom-line results

Six Sigma
Sigma Level
Defect.10
-6
± 1σ
± 2σ
± 3σ
± 4σ
± 5σ
± 6σ
697,700
308,700
66,810
6,210
233
3.4

Six Sigma

Six Sigma Companies

The Six Sigma Evolutionary Timeline
1736: French
mathematician
Abraham de
Moivre publishes
an article
introducing the
normal curve.
1896: Italian sociologist Vilfredo
Alfredo Pareto introduces the 80/20
rule and the Pareto distribution in
Cours d’Economie Politique.
1924: Walter A. Shewhart introduces
the control chart and the distinction of
special vs. common cause variation as
contributors to process problems.
1941: Alex Osborn, head of
BBDO Advertising, fathers a
widely-adopted set of rules for
“brainstorming”.
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-and-effect diagram.
1818: Gauss uses the normal curve
to explore the mathematics of error
analysis for measurement, probability
analysis, and hypothesis testing.
1970s: Imai develop Dr.
Deming concept called 14 keys
of Deming or called kaizen in
Japanese.
1986: Bill Smith, a senior engineer
and scientist introduces the
concept of Six Sigma at Motorola
1994: Larry Bossidy launches
Six Sigma at Allied Signal.
1995: Jack Welch
launches Six Sigma at
GE.

Customer
Competitive Price
High Quality Products
On-time Delivery, etc
Company
Profitability
Repeat Business
Growth/Expansion
Cash !!
Cash !!$
Value !!
Value !!
Some Profit
Some Profit
Bigger ProfitBigger Profit
1
2
3
1
2
3
Price - Cost = Profit
Price - Cost = Profit
Price to
Sell
Price to
Sell
Cost to
Produce
Cost to
Produce
KEY BUSINESS CONCEPT OF SIX SIGMA

Six Sigma
Methods Production
Design
Service
Purchase
HRM
Administration
Quality
Depart.
Management
M & S
IT
Where can Six Sigma be applied?

COPQ against sales revenue
COPQ (Cost of Poor Quality)
Sigma Level DPMO COPQ as sales percentile
1-sigma 691.462 (very low competitive) N/A
2-sigma 308.538 (Average Indonesia’s Industry)N/A
3-sigma 66.807 25-40% of sales
4-sigma 6.210 (Average USA’s Industry) 15-25% of sales
5-sigma 233 (Average Japan’s Industry) 5-15% of sales
6-sigma 3,4 (World Class Industry) < 1% of sales

$600
$500$450
$380
$200
$2500
$1200
$700
$170
CostBenefit
1996
CostBenefit
1997
CostBenefit
1998
CostBenefit
1999
CostBenefit
2000
6 Sigma Cost
6 Sigma Productivity
Delighting Customers
SUCCESS STORY IN SIX SIGMASUCCESS STORY IN SIX SIGMA
General ElectricGeneral Electric
$2500
$3.0B
$0.5B
$2.5B$2.5B

Dupont Chronology
Periode Description Sigma
Before six sigmaDupont Total Cost of Poor Quality
= 20 -30 % of revenue
About 3 sigma
Implementing Cost of Implementing Six Sigma
= $ 20 million
-
1999 Q1 Pilot Project Six Sigma on Specialty Chemicals started
-Revenues $ 1.5 Billions
-Target $ 80 million savings (5% dari revenues)
1999 Q2 40 Black Belts done for training and start for project
1999 Q4 Total saving $ 35 million (initial target $ 25 million) 2.3% of Revenue
17.7% COPQ
4 Sigma
2000 Q4 Total saving $ 100 million (initial target $ 80 million) 6.7% of Revenue
13.3% COPQ
5 sigma

Difficult-to-Reach Fruit
Design for Six Sigma (DFSS)
Middle Fruit
Six Sigma tools
Lower Fruit
7 Basic Tools of QC
Ground Fruit
Logic and Intuition
66σσ Basic Concept Basic Concept

3 sigma level company 6 sigma level company
• <25~40% of sales is failure cost. • 5% of the sales is failure cost.
• Having 66,807 defects per million. • 3.4 defects per million.
• Depends on the detect to find
defect.
• Focusing on process not to produce
defects.
• Believes that high quality is expensive. • Realizes that high quality creates
low cost.
• Not available of systematic
approach.
• Uses know-how of measurement,
analysis, improvement & control.
• Benchmarking against competing
companies.
• Benchmarking to the best
in the world.
• Believes 99% is good enough.
 4 sigma Level? 1misspelled word per 30pages of newspaper.
 5 sigma Level? 1misspelled word in a set of encyclopedias.
 6 sigma Level? 1misseplled word in all of the books contained in a small library.
• Define CTQ’s internally.
• Believes 99% unacceptable.
66σσ Basic Concept Basic Concept

• It is important to understand the difference between accuracy and precision
• Sigma is a measure of variationvariation (the data spread)
• It is a statistical measure unit displaying a process capability and the
measured sigma value is expressed by DPU(Defect Per Unit), PPM
• It is said that the process with higher sigma value is the process having smaller
defects
• The more increase the sigma value, the more decrease the quality cost and
Cycle Time
The concept of sigma
1σ
μ USL
3σ
T
Inflection
Point
: The size or a standard deviation
shows the distances between
the inflection point and the mean.
We could say the process has 3
sigma capability if 3 deviations
are fit table between the target
and the specification limit.
Understanding Basic Concept of Statistics

What does variation mean?
•Variation means that a
process does not produce
the same result (the “Y”)
every time.
•Some variation will exist in
all processes.
•Variation directly affects customer experiences.
Customers do Customers do notnot feel averages! feel averages!
-10
-5
0
5
10
15
20

Measuring Process Performance
•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?
The pizza delivery example. . .

How often are we delivering
on time?
Answer: Look at
the variation!
•Managing by the average doesn’t tell the whole story.
The average and the variation together show what’s
happening.
s
x
30 min. or less
0 10 20 30 40 50
The pizza delivery example. . .

Reduce Variation to Improve Performance
•Sigma level measures how often we meet (or fail to
meet) the requirement(s) of our customer(s).
s
x
30 min. or less
0 10 20 30 40 50
How many standard
deviations can you
“fit” within
customer
expectations?

SIX SIGMA Basic Concept
• All work occurs in a system of interconnected processes
• Variation exists in all processes
• Understanding and reducing variation are the keys to
improving customer satisfaction and reducing costs

 Y = f(χ)
Question 1), Which one should we focus on the Y or X?
Question 2), Is needed to test and audit Y continually if the X is good?
• Y
• Dependent Variable
• Output
• Effect
• Symptom
• Monitor
• X1 … Xn
• Independent Variable
• Input
• Cause
• Problem
• Control object
 6 Sigma activity is concerned about the problem happened(in the sector of
manufacturing and non manufacturing). They could be improved by focusing
the factor which causes the problem.

Steps Activity
Measurement
4. Understanding process capability for ‘Y’
5. Clarifying measurement method of ‘Y’
6. Specific description of Target object for improving
against ‘Y’
Focus
Y
Y
Y
Analysis
7. Clarifying Target for improving ‘Y’
8. Clarifying factors which affect ‘Y’
Y
X1 .... Xn
Improvement
9. Extract the vital few factors through screening
10. Understanding correlation of vital few factors
11. Process optimization and confirmation experiment
X1 .... Xn
Vital Few X1
Vital Few X1
Control
12. Confirm measurement system for ‘X’
13. Selection method how to control vital few factors
14. Build up process control system & audit for vital few
Vital Few X1
Vital Few X1
Vital Few X1
 6 Sigma activity with 5 steps of D-M-A-I-C, will pass through the major 14 steps.
 6 Sigma activity have D-M-A-I-C process breaking down the problem through the condition analysis, finding
the potential causal factor , and improving the vital few factors
After the condition identification, we have the first action about the part being improved at first, and then we
proceed continually the improvement activity at the next step.
Define
1. Clarifying improvement target object.
2. Forecasting improvement effect.
3. CTQ selection for products and process. Y
The Approach of 6 Sigma Step

6 Sigma Roles6 Sigma Roles

SIX SIGMA CHALLENGES
•Six sigma less suitable for innovation.
•Six sigma emphasize process and cost,
while innovation constitutes something
new in which cost consuming.
•Six sigma only analyzing quantitative
data, qualitative data must be converted to
quantitative.

Six Sigma DMAIC Process
Develop Charter and
Business Case
Map Existing Process
Collect Voice of the
Customer
Specify CTQs / Requirements
Measure CTQs / Requirements
Determine Process Stability
Determine Process Capability
Calculate Baseline Sigma
Refine Problem Statement
Identify Root Causes
Quantify Root Causes
Verify Root Causes
Institutionalize Improvement
Control Deployment
Quantify Financial Results
Present Final Project Results
and Lessons Learned
Close Project
Select Solution (Including
Trade Studies,
Cost/Benefit Analysis)
Design Solution
Pilot Solution
Implement Solution
Define
Measure
Analyze
Improv
e
Control
DMAIC = Define, Measure, Analyze, Improve and Control

66σσ Methodology Methodology

DefineDefine
Defining the “Project Y”
Translate the external CTQ’s into internal product requirements or “Project Y”.
Example:
Project YCTQ
Voice of the
Customer
The range must heat
to the setting chosen
The refrigerator must
stay dry
Call-takers must be
available to answer calls
Answer rate
(% of incoming calls
answered within
20 seconds)
Call-takers must answer
95% of all incoming calls
within 20 seconds
(telephone promptness)
Calibration angle
of the thermostat
A thermostat setting of
350° must result in a
350° oven cavity
Foam densityNo sweat

Process MappingProcess Mapping
Start Finish
1 2
Process Targets and Specifications
Experimental
Input Parameters
Target Upper Spec. Lower Spec.
Y = f (X)
(SOP ) = Standard Operating
Parameters
( N ) = Noise Parameters
( X ) = Controllable Process
Parameters
Add the operating specification and process targets
For the controllable variable input

Courtesy of Daraius Patell
Continue to ask “Why?” until you Reach the Root Cause…...
Structure Tree
Example
RPM
Losses
Inductance
OD
Core length
STATOR
ASSEMBLY
ROTOR
Electromagnetic
Mechanical
Area A
Area B
Lamination
Endrings

Cause & Effect DiagramCause & Effect Diagram
•The final diagram will look like a fishbone with the backbone displaying every known
variable (Measurement, Method, Machine, People, Materials, Environment).
Measurement Method Machine
People Materials Environment

33
Why-
why
Diagram

Measurement Measurement
Determine Process Capability for Determine Process Capability for
Project YProject Y
Determining process capability for your "Project Y" allows you to do several important things.
–Establish a baseline for comparing the improvement of your product or process.
–Quantify the ability of your process to produce output that meets the performance
standard.
–Determine if there is a technology or control problem.
–Understand process capabilities for the design of future processes for DFSS (Design
for Six Sigma) projects.
–Compare your process with others (internally and externally) to judge relative
performance.
·Define the problem in
mathematical terms
·Predict probability of
producing defects

Measurement Measurement
Determine Current Sigma LevelDetermine Current Sigma Level

Used to break down problem into manageable groups to identify root cause
or area of focus.
Process for creating a Structure Tree:
•List your problem statement on the left hand side of the page.
• Break the problem down into causes by asking ‘Why?’ and record on
tree branches. Typical categories of causes include:
Technical Transactional
Manpower People
Machine Price
Material Product
Method Promotion
Measurement Physical Distribution
•Assign a High, Medium or Low impact to each branch and select the
highest impact branch.
•Continue breaking down by asking ‘Why?’ until you reach the root cause.
Structure Tree

What is a “Measurement System”?
-Everything associated with taking measurements:
the people, measurement tool, material, method
and environment is known as --
Think of the “Measurement System” as a sub-process that
can add additional variation to measurement data. The goal
is to use a measurement process that has the smallest
amount of measurement error as possible.
Observations
Measurements
Data
Inputs Outputs Inputs Outputs
-- The “Measurement System”.
Parts
More Frequently Asked Questions More Frequently Asked Questions
About Measurement DataAbout Measurement Data

Observed
Process
Variation
Actual Process
Variation
Measurement
Variation
Long Term
Process
Variation
s
lt
Short Term
Process
Variation
s
st
Within
Sample
Variation
Variation due
to
Measurement
Equipment
Variation
due to
Operators
Accuracy Linearity ReproducibilityStabilityRepeatability
Sources of Measurement System VariationSources of Measurement System Variation
The Gage R&R methods we will study in this class will provide estimates of the total
measurement variation, the variation attributed to measurement equipment repeatability
and the variation attributed to the appraisers.

Acceptable if less than 20%
Conditional if between 20% to 30%
Unacceptable if greater than 30%
Evaluation Criteria for Evaluation Criteria for
%GR&R and %Study Variation%GR&R and %Study Variation
Beware of the risk associated with using
data acquired from an unacceptable
measurement process.
Beware of the risk associated with using
data acquired from an unacceptable
measurement process.
s
2
gage
= s
2
repeatability
+ s
2
reproducibility
Repeatability: The variation in measurements taken by a single person or
instrument on the same or replicate item and under the same conditions.
Reproducibility: the variation induced when different operators, instruments, or
laboratories measure the same or replicate specimen.

AnalyzeAnalyze
Hypothesis Test (for variables)
Hypothesis Test (for attributes)

Correct
Decision
Correct
Decision
Type 1
Error
α
Type 2
Error
β
Ho Ha
Ho
Ha
True
Accept
The ratio which is
being “Ha” even if it’s false.
Where “β” is usually
set up at 10%.
The ratio which is
being rejected Ho even
though certain thing is true
where “ α” is α error.
(usually 5%)
*Ho(Null Hypothesis) is assumed to be true.
This is like the defendant being assumed
to be innocent.
Ha(Alternative Hypothesis is alternatives
the Null Hypothesis.
Ha is the one that must be proved.

Data Types
Variable Discrete

◎
t-Test
(Compares means less than 2 population)

◎
ANOVA
(Compares variances more than 2 population)

◎
F-Test
(Compares variances of two population)

◎
Chi Square
(Compares counts and
frequencies.)
Before t-Test/ANOVA, confirm the
homogeneity of variance conducting
F-Test
the gap delta(δ)
The larger Means and expected gap is getting,
the more different two variances of average in population.
T

•The tool depends on the data type. We use ANOVA when
we have categorical input(s) and a continuous response.
Continuous Categorical
C
a
t
e
g
o
r
i
c
a
l

C
o
n
t
i
n
u
o
u
s
Dependent Variable (Y)
I
n
d
e
p
e
n
d
e
n
t

V
a
r
i
a
b
l
e

(
X
)
Regression
ANOVA
Logistic
Regression
Chi-Square (c
2
)
Test

Variance Homogeneity Testing Means Testing
1.One population variance
testing
2.Two population variances
testing
3.Testing of population
variances for more than two
(Normal distribution)
4. Testing of population
variances for more than two
(Non-Normal distribution)
Chi Square
F
Homogeneity of
Variance
▶Bartlett’s Test
Homogeneity of
Variance
▶Levene’s Test
1.One population Mean
testing

1) When we know σ of the
population

2) When we know σ of the
population
2.Two population Mean
testing
1) When they know
σ1 and σ2
2) When they don’t know
σ1 and σ2
① σ1 = σ2
② σ1 ≠ σ2
Normal distribution
( 1 - Sample Z )
T distribution
( 1 - Sample t )
Normal distribution
T distribution
( 2 - Sample t )
The type of Hypothesis

What is DOE ?
•DOE is more than just a statistical DOE is more than just a statistical
technique. technique.
•It is the combination of effective planning, It is the combination of effective planning,
discipline, subject matter knowledge and discipline, subject matter knowledge and
statistical methods that make the statistical methods that make the
experiment a success.experiment a success.
ImprovementImprovement

DOE Steps
1.1.Define the objective of the experiment.Define the objective of the experiment.
2.2.Select the response and input factors.Select the response and input factors.
3.3.Determine the resources required.Determine the resources required.
4.4.Select suitable experiment design Select suitable experiment design
matrix and analysis strategy. matrix and analysis strategy.
5.5.Perform the experiment and record Perform the experiment and record
data.data.
6.6.Analyse the data, draw conclusions, Analyse the data, draw conclusions,
and perform confirmation runs.and perform confirmation runs.
Good Good
planning is planning is
critical to critical to
success!success!

Basic Concepts in DOEBasic Concepts in DOE
PressurePressure
SpeedSpeed
ProcessProcess
Quality Characteristic (Y)Quality Characteristic (Y)
+
+
-
-
Pr
118Ave
+
58Ave -
12++90104
6--70103
4-+9052
10+-7051
Y
Pr
x SpdSpdSpdPrRun
7
9
Y = 8
2 0 6
y = 8 + Pr + 3 x ( Pr Spd)y = 8 + Pr + 3 x ( Pr Spd)
^^
ProcessProcess
Process Process
ModelModel
D
O
E
D
O
E

FMEAFMEA
Failure Modes and Effects Analysis is a systematic method for identifying,
analyzing, prioritizing, and documenting potential failure modes and their
effects on a system, product, or process.

SAMPLES
A B C D E
· Controllable factors
- Assignable causes
- Adjustable
- Special
· Uncontrollable factors
- Common causes
- Noise
- Inherent causes
• SPC has been traditionally used to monitor and control the output of processes.
In this application, we are measuring the dimensions of finished parts or
characteristics of finished assemblies.
• Six Sigma Quality focuses on moving control upstream to the leverage input characteristic for Y. If we can
measure and control the vital few X’s, control of Y should be assured.
INPUT
Controller
α/2
α/2
X
Lower Control Limit
Upper Control Limit
Process
Capability
Desired
Output
OUTPUTPROCESS
The Logic of SPC(Statistical Process Control)?
L M N O P
10050Subgroup 0
0.5
0.0
-0.5
S
a
m
p
l
e

M
e
a
n
Mean=0.001188
UCL=0.4384
LCL=-0.4360
1.0
0.5
0.0
S
a
m
p
l
e

R
a
n
g
e
1
11
R=0.2325
UCL=0.7596
LCL=0
Xbar/R Chart for Sealing Angle Line #2
Control Control

Types of Control Charts Types of Control Charts
Variables Charts for
monitoring continuous X’s
• Average & Range
X bar & R
n ＜ 10
typically 3~5
• Averages & Std Deviation
X bar & σ
n ≥ 10
• Median & Range
X & R
n ＜ 10
typically 3~5
• Individual &
Moving Range
XmR
n = 1
Attributes Charts for
monitoring discrete X’s
• Fraction Defective
P Chart
typically n ≥ 50
tracks DPU/DPU
• Number Defective
np Chart
n ≥ 50 (constant)
tracks # def
• Number of Defects
c Chart
c ＞ 5
• Number of Def/Unit
U Chart
n variable
• In order to select the appropriate control chart for monitoring your process, first
determine if your key process variables (X’s) are continuous or discrete. There are
control charts for both continuous data and discrete data.

Control Chart
12
10
8
·
·
·
·
·
·
·
·
··· ·
·
·
·
·
·
·
·
·
Week
Upper Control Limit = Ave + 3 x Std Dev
14
13
7
6
Lower Control Limit = Ave - 3 x Std Dev
Central
Line =
Average
Note: Control limits should be established using subgroup standard deviation

Six Sigma DMAIC
Implementation Project
Example

Date …………………
Department / Team
Prepared : …………..
Project title

G Manager Pres. Dir.Director
Mgr.
6
s

C
h
a
m
p
i
o
n

R
e
v
i
e
w
F
i
n
a
l

R
e
p
o
r
t
Contents
1. Define Step
2. Measure Step
3. Analysis Step
4. Improvement Step
5. Control Step
-Attachment

Define Step

PJT Name
P
e
r
i
o
d
Team
NameDiv./Dept:
Breakthrough

KPI Current World Best Target
Main Improvement Object

Team Formation (Related Department Involved)
Name Dept. Level Role
Quantitative
Qualitative

Expected
Results
How to do ?
Why ?
(* Selection Background)
New Idea for Target Achievement

Engineer

A
p
p
r
o
v
a
l
Ka PartKa Group
Project Registration
Neck Point

CIAM
D
4500
5000
Current Current Target Target Unit:
(Nm
3
/day)
11%
How to do:

Target Saving cost:
Others
E
lectric
O2N2
LNG
0.00700.06050.12900.13090.1650
1.412.326.226.633.5
100.0 98.6 86.3 60.1 33.5
0.5
0.4
0.3
0.2
0.1
0.0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
Energy Usage Price
Expected result
Background
General Background

Start Finish
1 2
Process Targets and Specifications
Experimental
Input Parameters
Target Upper Spec. Lower Spec.
Y = f (X)
(SOP ) = Standard Operating
Parameters
( N ) = Noise Parameters
( X ) = Controllable Process
Parameters
Add the operating specification and process targets
For the controllable variable input
Process Mapping
CIAM
D

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.
P
r
o
c
e
s
s

U
n
d
e
r
s
t
a
n
d
i
n
g
CIAM
D

Measure Step

F(x) Machine
Man
Big Y X
1
X
2
X
3
Brainstorming Potential X’s List
Material
Material
CIAD
M

Sampling
Training
Gage R&R
Result
Conduct a training……
How to see what kind ……..
Defect that happened in process.
Purpose : find out the operator ability.
Conduct Gage R&R for 4 men to know the
judgment capability (in different times & do
not know the inspection result of each other)
--> Repeat 2 times for each persons
% Gage R&R : 0 %
acceptable
Date:………
Collected 12 ea Sampling
Observed Process
Variation
Measurement Process Variation
Sample M a n
M a n
Machine
Method
Material
To determine if the measurement error is To determine if the measurement error is
small and acceptable relative to the small and acceptable relative to the
process variation, we can process variation, we can
use Gage R&R study.use Gage R&R study.
Gage R & R
CIAD
M

Units of Measurem
m
Center of the bar
Smooth curve
interconnecting the
center of each bar
Process Capability
Current Condition
1 2 3 4 5 6
1.0
0.5
1.5
2.0
2.5
Z
S
h
ift
P
r
o
c
e
s
s
C
o
n
tr
o
l
Good
Poor
Technology
Good
Poor
Block A
Block C
Block B
Block D
Four Block Diagram
Z Shift
A : Poor control, inadequate technology
B : Must control the process better, technology is fine
C : Process control is good, inadequate technology
D : World class
Current Target
CIAD
M

Factor Detail Analysis Plan
Schedule
Mar 2
nd
Week Mar 3
rd
WeekMar 4
th
Week
X1.3
X1.1
Inspect correlation between “Y” and inspector
for each group
Analysis khole / dimension
Height, diameter, angle etc,
X1.2
X1.4
Analysis Plan
CIAD
M

Analyze Step

CIDM
A
Regression Analysis: Angle Value versus Rotate Gear
The regression equation is
Angle Value = - 0,380 + 3,74 Rotate Gear
Predictor Coef SE Coef T P
Constant -0,3796 0,1693 -2,24 0,042
Rotate G 3,744 1,153 3,25 0,006
S = 0,1510 R-Sq = 43,0% R-Sq(adj) = 38,9%
Analysis of Variance
Source DF SS MS F P
Regression 1 0,24032 0,24032 10,55 0,006
Residual Error 14 0,31905 0,02279
Total 15 0,55937
Unusual Observations
Obs Rotate G Angle Va Fit SE Fit Residual St Resid
2 0,230 0,4000 0,4815 0,1070 -0,0815 -0,77 X
7 0,130 0,5000 0,1071 0,0407 0,3929 2,70R
Regression Analysis: Angle Value versus Rotate Gear
The regression equation is
Angle Value = - 0,380 + 3,74 Rotate Gear
Predictor Coef SE Coef T P
Constant -0,3796 0,1693 -2,24 0,042
Rotate G 3,744 1,153 3,25 0,006
S = 0,1510 R-Sq = 43,0% R-Sq(adj) = 38,9%
Analysis of Variance
Source DF SS MS F P
Regression 1 0,24032 0,24032 10,55 0,006
Residual Error 14 0,31905 0,02279
Total 15 0,55937
Unusual Observations
Obs Rotate G Angle Va Fit SE Fit Residual St Resid
2 0,230 0,4000 0,4815 0,1070 -0,0815 -0,77 X
7 0,130 0,5000 0,1071 0,0407 0,3929 2,70R
-0,2 -0,1 0,0 0,1 0,2 0,3 0,4
-1
0
1
2
N
o
r
m
a
l

S
c
o
r
e
Residual
Normal Probability Plot of the Residuals
(response is Angle Va)
Motor
Gear RotationGear Rotation
Use regression is to express and analyze a mathematical equation of describing a relationship.
That is, it is to fit a mathematical equation of describing a relationship between the “YY” and “X’sX’s”.

One-way ANOVA: Gr.A - 3, Gr.A - 4, Gr.B - 3, Gr.B - 4, Gr.C - 3, Gr.C - 4
Analysis of Variance
Source DF SS MS F P
Factor 5 0.8761 0.1752 2.59 0.030
Error 102 6.8897 0.0675
Total 107 7.7659
Individual 95% CIs For Mean
Based on Pooled StDev
Level N Mean StDev -------+---------+---------+---------
Gr.A - 3 18 0.0572 0.3029 (-------*-------)
Gr.A - 4 18 -0.1217 0.2427 (-------*-------)
Gr.B - 3 18 0.0839 0.2231 (--------*-------)
Gr.B - 4 18 -0.1183 0.2403 (-------*-------)
Gr.C - 3 18 0.0156 0.2796 (-------*-------)
Gr.C - 4 18 0.0967 0.2626 (-------*--------)
-------+---------+---------+---------
Pooled StDev = 0.2599 -0.15 0.00 0.15
P-Value: 0.242
A-Squared: 0.452
Anderson-Darling Normality Test
N: 18
StDev: 0.302865
Average: 0.0572222
0.50.0-0.5
.999
.99
.95
.80
.50
.20
.05
.01
.001
P
r
o
b
a
b
ilit
y
Gr.A - 3
Normal Probability Plot
See from sealing angle specifications there’s no problem, cause all operator adjustment
Still in range (-0.5
o
~ 0.5
o
). But there’s a significant effect both of them seeing by characte
ristic variation result, each operator have a different mean adjustment.
Since p-value < 0.05;
Ho (reject), Ha (accept).
That is we can claim there’s
a difference between the level
Of adhesive
G
r
.
C

-

4
G
r
.
C

-

3
G
r
.
B

-

4
G
r
.
B

-

3
G
r
.
A

-

4
G
r
.
A

-

3
0.5
0.0
-0.5
Boxplots of Gr.A - 3 - Gr.C - 4
(means are indicated by solid circles)
UCL
LCL
Screen
Manual Adjustment
Operator Adjustment
Target Line
We represent characterized variation “Y”“Y” by the total sum of square, then this method is to find
what the factor’s level which influence enormously is, comparing both of them.
CIDM
A

CIDM
A
Factor Detail Analysis Content ResultConclusionAnalysis Purpose
Selected as
Vital “X”
P = 0.000
Selected as
Vital “X”
X1.1
Find most effected
to “ Y “
X1.2
Find most effected
to “ Y “
P = 0.000
Selected as
Vital “X”
Bottle Neck
Regression, to compare Dew Point and
Purge Flow rate
X1.3
X1.4 Regression, to compare Dew Point
and Out Air Temperature
Find most effected
to “ Y “
Find most effected
to “ Y “
ANOVA, to compare Dew Point and
Heating Time
ANOVA, to compare Dew Point and
Drying Time
P = 0.000
P = 0.003
Analysis Result

Improve Step

CDMA
I
Design of Experiment
Full Factorial Design
Factors: 3 Base Design: 3, 8
Runs: 8 Replicates: 1
Blocks: 1 Center pts (total): 0
Drying TIme
Flow rate
10 hour
6 hourHeating Time
12 hour
4 hour
520 m
3
/hr
Level 2Level 1Factors
810 m
3
/hr
The Improve phase identifies a solution and confirms that the proposed
Solution will meet or exceed the improvement goals of the project.
StdOrderRunOrderCenterPtBlocks Flow rateHeating TimeDrying TimeResult
8 1 1 1 810 6 12 -90.2
5 2 1 1 600 4 12 -52.6
1 3 1 1 600 4 10 -56.4
7 4 1 1 600 6 12 -57.6
2 5 1 1 810 4 10 -89.2
3 6 1 1 600 6 10 -68.1
6 7 1 1 810 4 12 -85.3
4 8 1 1 810 6 10 91.3

 Heating Time : 4 hour
 Drying Time : 10 hour
 Flow rate : 520 m
3
/hour
Optimize Condition:
CDMA
I

-40-50-60-70-80
LSL USL
Process Data
Sample N 24
StDev(Within) 4.29772
StDev(Overall)8.60376
LSL -80.00000
Target *
USL -40.00000
Sample Mean -64.18750
Potential (Within) Capability
CCpk 1.55
Overall Capability
Z.Bench1.81
Z.LSL 1.84
Z.USL 2.81
Ppk
Z.Bench
0.61
Cpm *
3.68
Z.LSL 3.68
Z.USL 5.63
Cpk 1.23
Observed Performance
% < LSL0.00
% > USL0.00
% Total0.00
Exp. Within Performance
% < LSL0.01
% > USL0.00
% Total0.01
Exp. Overall Performance
% < LSL3.30
% > USL0.25
% Total3.55
Within
Overall
Process Capability of Dew Point
-40-50-60-70-80
LSL USL
Process Data
Sample N 24
StDev(Within) 4.29772
StDev(Overall)8.60376
LSL -80.00000
Target *
USL -40.00000
Sample Mean -60.00000
Potential (Within) Capability
CCpk 1.55
Overall Capability
Z.Bench2.05
Z.LSL 2.32
Z.USL 2.32
Ppk
Z.Bench
0.77
Cpm *
4.51
Z.LSL 4.65
Z.USL 4.65
Cpk 1.55
Observed Performance
% < LSL0.00
% > USL0.00
% Total0.00
Exp. Within Performance
% < LSL0.00
% > USL0.00
% Total0.00
Exp. Overall Performance
% < LSL1.00
% > USL1.00
% Total2.01
Within
Overall
Process Capability of Dew Point
1 2 3 4 5 6
1.0
0.5
1.5
2.0
2.5
Z
S
h
ift
P
r
o
c
e
s
s
C
o
n
tr
o
l
Good
Poor
Technology
Good
Poor
Block A
Block C
Block B
Block D
Four Block Diagram
Z Shift
Current Target
Improvement Result
CDMA
I

4500
5000
Current Current Target Target
Result
8%
4600
Result Result
92%92%
Cost Saving: US$
Improvement Result
CDMA
I

Control Step

I
ProcessInput
Controller
Controllable factors:
- Miss adjust causes
- Adjustable check
- Pad control
- Education
Group Member
Process CapabilityDesired
Output
X
Upper Control Limit
Lower Control Limit
●
●
Six Sigma Quality focuses on moving control upstream to the leverage input characteristic
for Y. If we can measure and control the vital few X’s, control of Y should be assured.
10050Subgroup 0
0.5
0.0
-0.5
S
a
m
p
l
e

M
e
a
n
Mean=0.001188
UCL=0.4384
LCL=-0.4360
1.0
0.5
0.0
S
a
m
p
l
e

R
a
n
g
e
1
11
R=0.2325
UCL=0.7596
LCL=0
Xbar/R Chart for Sealing Angle Line #2
Output
Process Standard Change
C
DMA

I
C
DMA
P Chart enables us to control our process using statistical method's to signal when
process adjustments are needed.
50403020100
0.004
0.003
0.002
0.001
0.000
Sample Number
P
r
o
p
o
r
t
io
n
P Chart for Stem Crack
P=0.001198
UCL=0.002466
LCL=0

15105Subgroup0
10.2
10.1
10.0
9.9
9.8
9.7
S
a
m
p
l
e

M
e
a
n
1
Mean=9.943
UCL=10.13
LCL=9.755
1.0
0.5
0.0
S
a
m
p
l
e

R
a
n
g
e
R=0.5594
UCL=1.016
LCL=0.1030
Xbar/R Chart for Cullet Speed
X-bar/R Chart use to control daily average for CTQ.
CTQ’s daily control data

Six Sigma Project
Example

Date 20 MAY 2009
Team : Galvanize
Prepared by : Imam Mudawam
Optimalkan pemakaian Zinc
Pada Proses Galvanize

Prod Mgr CEO
.
Work Mgr
6
s

C
h
a
m
p
i
o
n

R
e
v
i
e
w
F
i
n
a
l

R
e
p
o
r
t
Contents
1. Define Step
2. Measure Step
3. Analysis Step
4. Improvement Step
5. Control Step
ISKDCNG

PJTName
P
e
r
io
d
Tea
m
Nam
e
Div./Dept:
CONST./GALV.
Breakthrough
KPI Current World Best Target
Main Improvement Object

Team Formation (Related Department Involved)
Name Dept. Level Role
Quantitative
Qualitative

Expected
Results
How to do ?Why ?
(* Selection Background)
New Idea for Target Achievement

CEO
A
p
p
r
o
v
a
l
Work MgrProd Mgr
Neck Point
Optimalkan Pemakaian Zinc
Z
Shift:
-
1.86.

Imam M
Lukman
Banbang
Spv
Supv
Form
•Ketebalan Galvanize
sesuai standard.
Amin Form Member
Leader
Leader
Member
Rp / kg
Rp / kg
Slamet Member
Fab
Galv
Galv
Galv
Qc
CIAM
D
Project Registration
D
M
A
I
C
Schedule
-Making Theme Reg.
-Analyzing Aging Root-Caused
-Determine Potential X List
-Find Current situation
-Find Vital X by analyzing
Potential X
-Find Improvement idea
-Confirmation run
-Process Control by
Monitoring Aging
Amount
May – W5
Junl – W5
Jul –W5
Optimalkan Pemakaian Zinc
Pada Proses Galvanize
Saving Zinc
Rp. 295 Juta /Tahun
Galvanize
Reduce
cost
Hans Ga Supv
Form
Member
Sept –W3
Sept –W5
Galvanize
183.6 micr.
Galvanize
130 micron
280240200160120
LSL TargetUSL
LSL 100
Target 130
USL 150
Sample Mean183.649
Sample N 35
StDev(Within)18.0955
StDev(Overall)31.8811
Process Data
Z.LSL 2.62
Z.USL -1.06
Ppk -0.35
Lower CL-0.49
Upper CL-0.21
Cpm 0.11
Lower CL0.09
Z.Bench-1.86
Lower CL-2.59
Z.LSL 4.62
Z.USL -1.86
Cpk -0.62
Lower CL-0.80
Upper CL-0.44
Z.Bench-1.07
Lower CL-2.11
Overall Capability
Potential (Within) Capability
PPM < LSL 0.00
PPM > USL885714.29
PPM Total885714.29
Observed Performance
PPM < LSL 1.89
PPM > USL968521.55
PPM Total968523.44
Exp. Within Performance
PPM < LSL4348.13
PPM > USL854388.04
PPM Total858736.18
Exp. Overall Performance
Within
Overall
Process Capability of t 6.5mm +
(using 95.0% confidence)
Worksheet: Worksheet 3; 11/13/2009
ISK DC NG

General Background
CIAM
D
•Hasil proses Galvanize ketebalannya melebihi standar yang
ditentukan dalam ASTM- A123 / A123M.
•Tujuan Proyek ini untuk dapat mengoptimalkan ketebalan lapisan zinc
pada hasil proses galvanize.
Produk tebal 6.0s/d….mm

sample mean micron 183.6 84.2 83.6
Percent 52.3 24.0 23.8
Cum % 52.3 76.2 100.0
Ketebalan material
M
a
t e
r i a
l t . 1
. 5
s
/ d
3
. 0
m
m
M
a
t e
r i a
l t . 3
. 5
s
/ d
6
. 0
m
m
M
a t e
r i a
l t . 6
. 5
s / d
. . . m
m
400
300
200
100
0
100
80
60
40
20
0
s
a
m
p
l
e

m
e
a
n

m
ic
r
o
n
P
e
r
c
e
n
t
Pareto Chart of Ketebalan material
Worksheet: Worksheet 3; 10/14/2009

Big Y X
1
X
2
X
3
Brainstorming Potential X’s List
CIA
M
D
Degrising
Pickling 1 & 2
Material
Konsentrasi Basa
Caustic Soda
Base Metal
Konsentrasi Keasaman
Optimalkan pemakaian Zinc
Pada Proses Galvanize
Temperatur
Waktu Pencelupan
Komposisi kimia
Ketebalan
1
Hcl

Optimalkan Pemakaian Zinc
Pada proses Galvanize
Fluxing
Konsentrasi Keasaman
Big Y X
1
X
2
X
3
1
Temperatur
Brainstorming Potential X’s List
CIA
M
D
Zinc Amunium Chloride
Dipping
Aluminium Alloy
Zinc Ingot
Temperatur
Komposisi Campuran
Waktu pencelupan

In doing gage R&R
we take 2 times
repeat in check
Gage R&R take
from 2 Inspector
who check this
sample of
Galvanize coating
thickness in 10
chek point
Decide
operator
who take Gage
R&R
Do Gage
R&R
Change
Method,measurement, etc
Analyze
Result
Gage
R&R
Next Step
NG ;
Total Gage
R&R > 20%
OK ;
Total Gage
R&R < 20%
Measure Step - GAGE R&R
CIA
M
D

Gage R&R
Gage R&R < 20%
Acceptable
CIA
M
D
Sample pengukuran ketebalan galvanize sebanyak 2 sample dengan 30 titik pengukuran , tiap sampel diambil 15 titik
pengukuran
Diukur secara berurutan dan secara acak oleh dua orang operator .
The result of Gage R&R total is 14.67 % the acceptance percentage is
bellow 20% (< 20 % ) ,
meanwhile the result of measurement between < 20 %, it accepted.
Study Var %Study Var
Source StdDev (SD) (6 * SD)
(%SV)
Total Gage R&R 2.0064 12.0384
14.67
Repeatability 2.0064 12.0384
14.67
Reproducibility 0.0000 0.0000
0.00
oprtr 0.0000 0.0000
0.00
Part-To-Part 13.5243 81.1460
98.92
Total Variation 13.6724 82.0341
100.00
Part-to-PartReprodRepeatGage R&R
100
50
0
P
e
r
c
e
n
t
% Contribution
% Study Var
8
4
0
S
a
m
p
l
e

R
a
n
g
e
_
R=2.633
UCL=8.604
LCL=0
A B
110
100
90
S
a
m
p
l
e

M
e
a
n
_
_
X=96.67
UCL=101.62
LCL=91.71
A B
321
110
100
90
part no
BA
110
100
90
oprtr
321
110
100
90
part no
A
v
e
r
a
g
e
A
B
oprtr
Gage name:
Date of study:
Reported by:
Tolerance:
Misc:
Components of Variation
R Chart by oprtr
Xbar Chart by oprtr
measure by part no
measure by oprtr
oprtr * part no Interaction
OPERATOR PENGUKURAN KETEBALAN GALVANIZE
Worksheet: Worksheet 3; 9/9/2009

CIA
M
DCurrent Condition
Rata rata ketebalan Galvanize pada produk dengan ketebalan > 6.0 mm saat ini mencapai 183.6 micron , berdasarkan data
dari tgl.20 Juni 09 Sampai 27 Juni 09 ,jumlah sample 35 , alat ukur menggunakan COATING THICKNESS DIGITAL merek
TIME - TYPE TT 220 .
Untuk mendapatkan ketebalan sesuai target yang diinginkan grafik harus bergeser kekiri , dan harus mengurangi keteba-
Pelapisan Galvanize yang sesuai standard ASTM A 123/A 123M antara ( 100 s/d 150 ) micron .
300250200150100
99
95
90
80
70
60
50
40
30
20
10
5
1
t 6.5mm +
P
e
r
c
e
n
t
Mean 183.6
StDev 31.88
N 35
AD 1.706
P-Value<0.005
Probability Plot Ketebalan Galvanize material 6.5mm s/d .......dst
Normal - 95% CI
Worksheet: Worksheet 3; 7/1/2009
KETEBALAN GALVANIZE
280240200160120
LSL TargetUSL
LSL 100
Target 130
USL 150
Sample Mean183.649
Sample N 35
StDev(Within)18.0955
StDev(Overall)31.8811
Process Data
Z.LSL 2.62
Z.USL -1.06
Ppk -0.35
Lower CL-0.49
Upper CL-0.21
Cpm 0.11
Lower CL0.09
Z.Bench-1.86
Lower CL-2.59
Z.LSL 4.62
Z.USL -1.86
Cpk -0.62
Lower CL-0.80
Upper CL-0.44
Z.Bench-1.07
Lower CL-2.11
Overall Capability
Potential (Within) Capability
PPM < LSL 0.00
PPM > USL885714.29
PPM Total885714.29
Observed Performance
PPM < LSL 1.89
PPM > USL968521.55
PPM Total968523.44
Exp. Within Performance
PPM < LSL4348.13
PPM > USL854388.04
PPM Total858736.18
Exp. Overall Performance
Within
Overall
Process Capability of t 6.5mm +
(using 95.0% confidence)
Worksheet: Worksheet 3; 11/13/2009

Four Block Diagram
CIA
M
D
Nilai sigma level saat ini adalah – 1.86 s, target yang ingin
dicapai 4.5 s
A B
C
0.5
1.0
1.5
2.0
2.5
1 2 3 4 5 6
Z

s
h
if
t
P
r
o
c
e
s
s

C
o
n
t
r
o
l
Good
Poor
Poor Good
Z st
Technology
Position of Current Condition was
column C , it was mean :
PROCESS CONTROL IS GOOD,
BUT TECHNOLOGY (METHOD)
IS BAD
EXPLANATION
D
SIGMA TARGET
SIGMA TARGET
Sigma current – 1.86

CI
A
MD
WAKTU PENCELUPAN
Optimalkan pemakaian Zinc
Pada proses Galvanize
F(x)
x
Analysis
TEMPERATUR ZINC
KETEBALAN GALVANIZE

Optimalkan
pemakaian
zinc pada
proses
Galvanize
(Continuous)
Waktu Pencelupan
Y Factor (x) Type Tools
Analyze – Type of factor & Tools Using CI
A
MD
CONTINUE
REGRESION

Regression Analysis: Ketbln.Galva versus Wktu fluxing, Wktu.deeping, ...
* Wktu.deeping is (essentially) constant
* Wktu.deeping has been removed from the equation.
* NOTE * All values in column are identical.
* Temp.deeping is (essentially) constant
* Temp.deeping has been removed from the equation.
The regression equation is
Ketbln.Galvanize (micron ) = 110 - 1.17 Wktu fluxing
Predictor Coef SE Coef T P VIF
Constant 109.667 3.658 29.98 0.000
Wktu fluxing -1.1668 0.4555 -2.56 0.034 1.000
S = 4.13685 R-Sq = 45.1% R-Sq(adj) = 38.2%
Analysis of Variance
Source DF SS MS F P
Regression 1 112.32 112.32 6.56 0.034
Residual Error 8 136.91 17.11
Total 9 249.22
Unusual Observations
Wktu Ketbln.Galvanize
Obs fluxing (micron ) Fit SE Fit Residual St Resid
1 3.0 114.00 106.17 2.43 7.83 2.34R
R denotes an observation with a large standardized residual.
Durbin-Watson statistic = 1.69422
Analysis Waktu Fluxing
( Regresion )
CI
A
MD
Ketebalan Galvanize VS
Waktu Fluxing
Nilai p-value < 0.05, maka
Ho ditolak, yang
menandakan variabel
waktu Fluxing
berpengaruh terhadap
ketebalan galvanize.
Nilai p-value < 0.05, maka
Ho ditolak, yang
menandakan variabel
waktu Fluxing
berpengaruh terhadap
ketebalan galvanize.

Regression Analysis: Ktbl . Galva versus Tmprt deepin,
Wkt.deeping, ...
* Wkt.deeping is (essentially) constant
* Wkt.deeping has been removed from the equation.
* NOTE * All values in column are identical.
* Wkt.fluxing is (essentially) constant
* Wkt.fluxing has been removed from the equation.
The regression equation is
Ktbl . Galvanize (micron ) = - 98 + 0.443 Tmprt deeping
Predictor Coef SE Coef T P VIF
Constant -98.4 157.1 -0.63 0.548
Tmprt deeping 0.4428 0.3562 1.24 0.249 1.000
S = 6.47038 R-Sq = 16.2% R-Sq(adj) = 5.7%
Analysis of Variance
Source DF SS MS F P
Regression 1 64.70 64.70 1.55 0.249
Residual Error 8 334.93 41.87
Total 9 399.63
Durbin-Watson statistic = 1.36705
Analysis Temperatur
Dipping ( Regresion )
CI
A
MD
Ketebalan Galvanize VS Temperatur Zinc
Nilai p-value > 0.05, maka
Ho diterima, yang
menandakan variabel
temperatur tidak
berpengaruh terhadap
ketebalan galvanize.
Nilai p-value > 0.05, maka
Ho diterima, yang
menandakan variabel
temperatur tidak
berpengaruh terhadap
ketebalan galvanize.

Regression Analysis: Ktbalan Galvanize micron versus Waktu
deeping
The regression equation is
Ktbalan Galvanize micron = 43.7 + 19.9 Waktu deeping
Predictor Coef SE Coef T P VIF
Constant 43.674 5.279 8.27 0.000
Waktu deeping 19.9082 0.6573 30.29 0.000 1.000
S = 5.96979 R-Sq = 99.1% R-Sq(adj) = 99.0%
Analysis of Variance
Source DF SS MS F P
Regression 1 32698 32698 917.49 0.000
Residual Error 8 285 36
Total 9 32983
Durbin-Watson statistic = 2.41646
Analysis Waktu Dipping
( Regresion )
CI
A
MD
Waktu Dipping VS Ketebalan Galvanize

Nilai p-value < 0.05, maka
Ho ditolak, yang
menandakan variabel
waktu dipping
berpengaruh terhadap
ketebalan galvanize.
Nilai p-value < 0.05, maka
Ho ditolak, yang
menandakan variabel
waktu dipping
berpengaruh terhadap
ketebalan galvanize.

CI
A
MD
Selected not Vital View
Item Content Result Remarks
Selected as Vital View
Galvanizing
P = 0.034
P = 0.024
P = 0.000
Waktu Fluxing
Temperatur Dipping
Waktu Dipping
Analysis Result ( Regression )
Selected not Vital View
Berdasarkan data diatas variabel waktu Dipping paling berpengaruh terhadap hasil ketebalan pada proses galvanize (99%)

C
Improvement
I
AMD
•Berdasarkan hasil analisa, untuk
produk dengan ketebalan 6 mm
didapatkan hubungan linier antara
lamanya waktu pencelupan dengan
hasil ketebalan galvanize.
•Grafik di bawah ini dapat dijadikan
acuan untuk menentukan tebal
galvanize yang diinginkan.
HUBUNGAN WAKTU PENCELUPAN DAN KETEBALAN GALVANIZE
PADA PRODUK DENGAN KETEBALAN 6 mm
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
250
260
0 1 2 3 4 5 6 7 8 9 10 11
Waktu Pencelupan (menit)
K
e
t
e
b
a
l
a
n

G
a
l
v
a
n
i
s

(
m
i
c
r
o
n
)
Standar 6mm
Y = 43.7 + 19.9 x

94
C
Improvement
I
AMD
Catatan : Proses improvement masih sedang berjalan dan akan dilaporkan
kemudian hari.

Session Start : 23.01.02
□○ Theme Register
□○ Team Organization
□○ Process Map
□○ Cpk Analysis(Current)
□○ Problem Description
□○ CTQ Selection
□○ Measurable Y Value
Selection
□○ 4 Block Diagram
□○ Brainstorming
□○ Logic Tree Analysis
□○ Analysis by Minitab
□○ Process Benchmarking
□○ CTQ Selection
□○ Process Map
□○ ANOVA
□○ Regression Analysis
□○ Factor Level Decision
□○ DOE
□○ Statistical Interpretation
□○ Data Gathering & Analysis
□○ Main Factor Analysis
□○ Hypothesis Test
□○ ANOVA
□○ Control Chart
□○ Rational Tolerance
selection
□○ Document Control plan
□○ Training Process Controller
□○ CTQ Process Monitoring
System set up
Session Finished :
□○ Y Identification
□○ Gauge R&R
□○ 4 Block Focus(Zst & Zshift)
□○ Problem analysis
reaffirmation
□○ Statistical skill of Y
□○ Graphical skill of Y
□○ Gap Analysis
□○ The 1st improvement of
X(Factor)
□○ Conclusion(the fixed X
factor)
□○ Test plan
□○ Control Plan
Implementation
□○ CTQ Process Monitoring
System Build-Up
□○ Double check of all the
problems
Measurement Analysis Improvement Control
Session Start : Session Start : Session Start :
Main Schedule
Session Finished : Session Finished : Session Finished :
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

A period of time taken off by an employee which is neither
planned or authorised
Definition Of Defect
DEFINITION
Sickness
Unauthorised absence
Lateness
DEFECT TYPES
Payment for overtime to cover absence = £800,000:00 pa
Loss in revenue due to lost production = £700,000:00 pa (est)
Adverse morale issues
Deterioration in productivity and quality
VALUE
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

Absence Logic Tree
Absence
Management
Style
Time With Company
Sex
=CTQ’s
Age
Guidelines
Team Leader
Section
Accidents
Morale issues
Cleanliness
Repetitive Work
Employee
Target Setting
Environment
Method
Consistent Approach
Use Of Policy
Shift Pattern
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

Graph shows reduction in absence, since the beginning of the project (DEC) spotlight effect has
reduced absence. One problem we do have is the calculation of absence data there is a difference
between the HR data collection method and the tube plant data collection method.
Sigma Level Calculation
Using The Percentage Defective (4.605%) We Calculated The DPMO Since Merger
DPMO = 46050
Using statistical tables the SIGMA LEVEL = 3.21
Measure Measure Define Define Improve Improve Control Control Analyse Analyse
-2
0
2
4
6
8
Tube 4.274.665.234.75.015.343.863.132.75
HR 4.315.225.864.565.095.023.973.452.72
Variance-0.04-0.56-0.630.14-0.080.32-0.11-0.320.03
JulAugSepOctNovDecJanFebMarAprMayJun

VOC - Harp Questionnaire
219 People Questioned Across The Tube Plant
Answered by all employees.
Name Marital Status
Clock Number Age
Department Children
Shift Home
Time with Company Job position
Time in current job
Have you taken any unauthorised time off since 5
th
July 2001? YES / NO
Do you currently have any warnings that relate to sickness or absence?YES / NO
Do you understand LGPDW's absence and sickness policy? YES / NO
Do you understand the affect of absence on our business? (e.g. cost, pressure on colleagues)
Only to be answered by employees with no absence history
How do you think high absence levels affects your ability to do your job?
What do you think of LG. PHILIPS Displays as an employer?
Are you happy working for LGPDW?
Ask employee all questions below, and mark the score they give ie strongly agree
= 10, agree = 5 and strongly disagree = 0.
Do you think absence is due to Management style?
Do you think absence is due to accidents in work?
Do you think absence is due to the current shift pattern?
Do you think absence is due to your working environment?
If I was late for work, I would take the rest of the shift off because it still counts as an absence
Are there any other reasons that you think contribute to absence that are not listed above?
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

Overall
catchment area
Main catchment
area
Catchment Area Of People Questioned
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

Others
Job rotation
P
ay rise
M
usic
Mgt attitude
Att Bonus
12 5 14 33 51137
4.8 2.0 5.613.120.254.4
100.0 95.2 93.3 87.7 74.6 54.4
250
200
150
100
50
0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
How would y ou tackle high absence?
Operators
Pareto’s Of Absence Data
Teambuilding
E
xtra m
anning
gather info on persistant absentees
1 - 1C
omm
unication
A
ttendance bonus
11224
10.010.020.020.040.0
100.0 90.0 80.0 60.0 40.0
10
5
0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
How would y ou tackle absence?
Section Leaders
Conclusions
Operators would like to be paid
more to come to work
Operators see a problem in the
way they are treated by Mgt
Music might help??
Conclusions
Section Leaders see the benefit
in some form of attendance
bonus, but more emphasis on
communication and data
collection
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

Pareto’s Of Absence Data
Others
deaths
car problem
s
accidents inside w
ork
low morale
injury outside work
flexible floatdays
sickness
flu
domestic
1111223444
4.3 4.3 4.3 4.3 8.7 8.713.017.417.417.4
100.0 95.7 91.3 87.0 82.6 73.9 65.2 52.2 34.8 17.4
20
10
0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
Main Reasons For Absence On Your Shif t
Others
experience
absence
poor m
aintenance
supply
low
m
anning
1 1 2 3 511
4.3 4.3 8.713.021.747.8
100.0 95.7 91.3 82.6 69.6 47.8
20
10
0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
Main issues af f ecting shif ts ability to meet output targets
Conclusions
No significant patterns have
emerged
Conclusions
Absence is not a major issue
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

Analysis Using M System
ImproveImproveDefineDefine MeasureMeasure AnalyseAnalyse ControlControl
Time With Company
Sex
Age
Employee
NEXT STEP IS TO CARRY OUT ANALYSIS BY
EMPLOYEE TO ESTABLISH IF THERE ARE
ANY STATISTICAL DIFFERENCES
DATABASE WAS CREATED LOOKING AT THE
NINE MONTH PERIOD (5th July-4th April) PRIOR
TO AND DURING PROJECT
empn
o
Emp Name
dept
cd
Dept Name Grade shift no
Distance
Details
date_hir
ed
Year/MonthgenderAge
Age
Group
Count
Absence
Abs Occ
12345 A N OTHER1 81000Tube ManufacturingSenior EngineerDT CF1 1AB17/03/00 02/01 Male34 D 0 0
12346 A N OTHER2 81000Tube ManufacturingSenior EngineerDT CF1 1AB06/10/99 02/06 Male34 D 0 0
12347 A N OTHER3 81000Tube Manufacturing Team Leader DT CF1 1AB11/08/97 04/08 Male33 D 3 1
12348 A N OTHER4 81100 Screen Production Section Leader T4 CF1 1AB18/05/98 03/11 Male51 H 0 0
12349 A N OTHER5 81100 Screen Production Section Leader T2 CF1 1AB26/04/99 02/12 Male42 F 1 1
12350 A N OTHER6 81100 Screen Production Section Leader T1 CF1 1AB10/05/99 02/11 Male41 F 0 0
12351 A N OTHER7 81100 Screen Production Section Leader DT CF1 1AB20/04/98 03/12 Male41 F 0 0
12352 A N OTHER8 81100 Screen Production Section Leader T2 CF1 1AB06/10/97 04/06 Male39 E 1 1
12353 A N OTHER9 81100 Screen Production Team Leader DT CF1 1AB12/01/98 04/03 Male39 E 0 0
12354 A N OTHER10 81100 Screen Production Technician DT CF1 1AB06/10/97 04/06 Female35 D 0 0
Time with Company Personal DetailsEmployment Details Unauthorised Absence

Analysis By Age
ImproveImproveDefineDefine MeasureMeasure AnalyseAnalyse ControlControl
Created Age Groupings
Age Group Age Range
A 16-20
B 21-25
C 26-30
D 31-35
E 36-40
F 41+
Analysed Percent Absent By Age Group
Conclusions
There is a significance showing that
D = 31-35 and F = 41+ are more likely to
not to take unauthorised absence and that
B = 21-25 are more likely to take
unauthorised absence
Chi-Square Test: A, B, C, D, E, F
Expected counts are printed below observed counts
A B C D E F Total
1 188 175 183 132 91 62 831
183.80 187.26 182.08 125.12 94.06 58.68
2 25 42 28 13 18 6 132
29.20 29.74 28.92 19.88 14.94 9.32
Total 213 217 211 145 109 68 963
Chi-Sq = 0.096 + 0.802 + 0.005 + 0.378 + 0.099 + 0.188 +
0.603 + 5.050 + 0.029 + 2.378 + 0.626 + 1.183 = 11.438
DF = 5,

P-Value = 0.043
19.35%
13.27%
8.97%
16.51%
8.82%
13.71%
11.74%
A B C D E F Grand
Total

Analysis By Gender
ImproveImproveDefineDefine MeasureMeasure AnalyseAnalyse ControlControl
Split By Gender
Analysed Percent Absent By Gender
Conclusions
There is significance showing that a
male person would more likely take
unauthorised absence
Female Male Grand Total
22.28% 30.88% 29.08%
157 526 683
45 235 280
Chi-Square Test: F, M
Expected counts are printed below
observed counts
F M Total
1 188 643 831
174.31 656.69
2 14 118 132
27.69 104.31
Total 202 761 963
Chi-Sq = 1.075 + 0.285 +
6.767 + 1.796 = 9.924
DF = 1,
P-Value = 0.002
15.51%
13.71%
6.93%
Female Male Grand Total

Group Range
A <1Year
B 1-2 Years
C 2-3 Years
D 3-4 Years
E >4 Years
Analysis By Time Served
ImproveImproveDefineDefine MeasureMeasure AnalyseAnalyse ControlControl
Analysed Percent Absent By Time Served
Conclusions
There is a significance in that
E = 4 years +
is more likely to take unauthorized
absence
Created Time Served Groupings
Chi-Square Test: A, B, C, D, E
Expected counts are printed below observed counts
A B C D E Total
1 117 129 151 324 110 831
113.04 126.85 154.46 314.97 121.67
2 14 18 28 41 31 132
17.96 20.15 24.54 50.03 19.33
Total 131 147 179 365 141 963
Chi-Sq = 0.138 + 0.036 + 0.078 + 0.259 + 1.120 +
0.872 + 0.229 + 0.489 + 1.630 + 7.050 =
11.902
DF = 4,
P-Value = 0.018
12.24%
15.64%
11.23%
21.99%
13.71%
10.69%
A B C D E Grand
Total

Analysis By Department
ImproveImproveDefineDefine MeasureMeasure AnalyseAnalyse ControlControl
Conclusions
This shows that there is a significance in departments
A = Tube Material Warehouse, B = Design/Process Engineering,
C = Sputter section.
In which all are more likely to take unauthorised absence
Percent Absent By Sub GroupCreated Department Sub-Groups
Groupdept_cd Dept Name TotalAbsNever
A 83210 Tube Material Warehouse 46.15%6 7
B 81400 Design/Process Engineering35.71%5 9
C 81320 Sputter section 32.50%1327
D 81310 Spin section 21.21%7 26
E 84120 Tube OQC section 16.13%5 26
81140 Maintenance 1 section
81510 Maintenance Section(2/3)
81230 Gunseal and Exhaust section
81330 I.T.C. section
81340 Outgoing section
81450 CS-Reinspection section
81210 Assembly Section
81220 Inner Dag section
I 84110 Tube IQC section 7.69% 1 12
J 81240 1st Inspection section 5.63% 4 67
81110 Screen Coating section
81120 Chemical section
L 81130 Shadow Mask Section 2.50% 2 78
Average 12.72%109748
F
G
H
14.55%
13.68%
10.29%
47
284
61
8
45
7
K 5.45% 6104
35.71%
32.50%
21.21%
16.13%
14.55%
13.68%
10.29%
7.69%
5.63%5.45%
2.50%
46.15%
A B C D E F G H I J K L

Analysis By Shift
ImproveImproveDefineDefine MeasureMeasure AnalyseAnalyse ControlControl

Improvement Suggestions
Improvement Actions/Suggestions by CTQ.
CTQ Proved By Improvement Recommendation
Age
Proved By Chi Square
Test
Use best fit when recruiting.
Gender
Proved By Chi Square
Test
Use best fit when recruiting.
Length Of Service
Proved By Chi Square
Test
Use best fit when recruiting.
Shift Pattern
Proved By Chi Square
Test
Build In More flexibility for day shift workers.
Department
Proved By Chi Square
Test
Compare Management Styles
Morale
Proved By HARP
Survey
New Incentive Scheme (Ongoing)
Accidents
Proved By HARP
Survey
New Health & Safety Structure In Place
Envirionment
Proved By HARP
Survey
Music & Improved Rest Room Facilities
Management Style
Proved By HARP
Survey
Training Courses For Manager On Interpersonal Skills.
Management Attitude Improvement Plan Next Slide
Aggressive Target Setting
Proved By HARP
Survey
Unable to improve due to the nature of our business.
Measure Measure Define Define Improve Improve Control Control Analyse Analyse

Improvement Suggestions - Management Attitude
Measure Measure Define Define Improve Improve Control Control Analyse Analyse
AbsenceMorale
More
Information
Team Building
Improved
Interpersonal
Skills
Treat
Operators As
Equal
More 1 To 1
Communication
Follow Correct
Procedures
Improved
Grading
System
HIGH MORALE = LOW ABSENCE

Improvement Suggestions - Attendance Bonus
Measure Measure Define Define Improve Improve Control Control Analyse Analyse
£100 Per
Year
Decided By
Incentive Scheme
Working Party
Deductions?
All Authorised
Non- Sickness
Absence
Paid Quarterly
Cash/Vouchers/Sa
vings
All Absence
Resulting In
Warnings
Total
Savings
£1,086,000
Verbal -25%
Written - 50%
F Written - 100%
All
Deductions
For 1 Year
From Date Of
Warning

Output
MeasureMeasureDefineDefine ImproveImproveControlControlAnalyseAnalyse
Initial Current
s level 3.21 3.43
PPM (Month) 46021 27300
Loss (£) £117000 £69405
£47595
Est. Monthly Saving of:
(Based on hours lost)

Contents:
1. Define Step
2. Measure Step
3. Analysis Step
4. Improvement
5. Control

Process Sterilization
Capability Up
6
s

C
h
a
m
p
i
o
n

R
e
v
i
e
w
F
i
n
a
l

R
e
p
o
r
t

Background
CIAM
D
Develop working efficiency and found 6 Sigma6 Sigma control for free
Salmonella in sterilization process.
0 100 200 300 400 500
-0.5
0.0
0.5
1.0
Individual and MR Chart
Obser.
In
d
i
v
i
d
u
a
l
V
a
lu
e
Mean=-0.01707
UCL=0.7674
LCL=-0.8016
0.0
0.3
0.6
0.9
M
o
v
.
R
a
n
g
e
R=0.2950
UCL=0.9638
LCL=0
480 490 500
Last 25 Observations
-0.6
-0.3
0.0
0.3
Observation Number
V
a
lu
e
s
-0.5 0.5
Capability Plot
Process Tolerance
I I I
I I I
I I
Specifications
Within
Overall
-0.5 0.0 0.5
Normal Prob Plot
-0.5 0.0 0.5
Capability Histogram
Within
StDev:
Cp:
Cpk:
0.261507
0.64
0.62
Overall
StDev:
Pp:
Ppk:
0.288617
0.58
0.56
Process Capability Sixpack for Sealing Angle Line #2
Process Capability Current Condition
2
0.65
1.5
0.51
Target Current
Cp
Cpk

D CIA
M
Executing analysis with Logic Tree for sterile product.
Potential X’ List
Big Y X
1
X
2
X
3
Sterile Product Machine Sterile Time Fixed
Rotate
Temp Warm
Hot > 97
Steam Supply Spec (6.5 ~ 9) bar
Pineapples Bracket Stand
Methods Automatic
Material
Juice pH
Manual

Gage R&R
D CIA
M
0.01
0.05
novi m
Mar 3rd,2006
Sealing angle Line 2
Misc:
Tolerance:
Reported by:
Date of study:
Gage name:
0
0.5
0.0
-0.5
CBA
Xbar Chart by Operator
S
a
m
p
le

M
e
a
n
Mean=0.03017UCL=0.03706LCL=0.02327
0
0.010
0.005
0.000
CBA
R Chart by Operator
S
a
m
p
le

R
a
n
g
e
R=0.003667
UCL=0.01198
LCL=0
10 9 8 7 6 5 4 3 2 1
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
Part
Operator
Operator*Part Interaction
A
v
e
r
a
g
e
A
B
C
CBA
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
Operator
By Operator
10 9 8 7 6 5 4 3 2 1
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
Part
By Part
%Contribution
%Study Var
Part-to-PartReprodRepeatGage R&R
100
50
0
Components of Variation
P
e
r
c
e
n
t
Gage R&R (ANOVA) for Measurement
Gage R&R
%Contribution
Source VarComp (of VarComp)

Total Gage R&R 0.000020 0.03
Repeatability 0.000020 0.03
Reproducibility 0.000000 0.00
Operator 0.000000 0.00
Part-To-Part 0.077747 99.97
Total Variation 0.077767 100.00
StdDev Study Var %Study
Var
Source (SD) (5.15*SD) (%SV)

Total Gage R&R 0.004437 0.02285 1.59

Repeatability 0.004425 0.02279 1.59

Reproducibility 0.000323 0.00166 0.12

Operator 0.000323 0.00166 0.12

Part-To-Part 0.278832 1.43598 99.99

Total Variation 0.278867 1.43616 100.00

Two-Way ANOVA Table With Interaction
Source DF SS MS F P

Part 9 4.19852 0.466502 21530.8 0.000000.00000
Operator 2 0.00004 0.000022 1.0 0.38742
Operator*Part 18 0.00039 0.000022 1.2 0.33365
Repeatability 30 0.00055 0.000018
Total 59 4.19950
If significant, P-value < 0.05 indicates
that a part is having a variation for
Some measuring
Ok for “product acceptance”
considering a products
tolerance.

D CIA
M
Measurement
Through analysis of process capability , getting sigma level 1.85 1.85 ss
1 2 3 4 5 6
1.0
0.5
1.5
2.0
2.5
Z
S
h
ift
P
r
o
c
e
s
s
C
o
n
tr
o
l
Good
Poor
Technology
Good
Poor
Block A
Block C
Block B
Block D
Four Block Diagram
1.85 s
4.5 s
Z Shift
1.00.50.0-0.5-1.0
Target USLLSL
Angle Line #2
Process Capability Analysis for Sealing
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
Ppk
Z.LSL
Z.USL
Z.Bench
Cpm
Cpk
Z.LSL
Z.USL
Z.Bench
StDev (Overall)
StDev (Within)
Sample N
Mean
LSL
Target
USL
83742.06
36602.01
47140.05
56399.74
24004.66
32395.08
4008.02
0.00
4008.02
0.56
1.67
1.79
1.38
0.58
0.62
1.85
1.98
1.59
0.288617
0.261507
499
-0.017074
-0.500000
0.000000
0.500000
Exp. "Overall" PerformanceExp. "Within" PerformanceObserved PerformanceOverall Capability
Potential (Within) Capability
Process Data
Within
Overall
Process capability for Sterile Product
A : Poor control, inadequate technology
B : Must control the process better, technology is fine
C : Process control is good, inadequate technology
D : World class
Target

Analysis - Regression D CIM
A
Regression Analysis: Sterile product versus time and temp
The regression equation is
Sterile =0.000303 + 0.00113 time + 0.000060 temp
Predictor Coef SE Coef T P
Constant 0.0003033 0.0002999 1.01 0.345
Time 0.00112859 0.00003939 28.65 0.000Time 0.00112859 0.00003939 28.65 0.000
Temp 0.00006005 0.00005694 1.05 0.327
S = 0.0003891 R-Sq = 100.0% R-Sq(adj) = 100.0%
Analysis of Variance
Source DF SS MS F P
Regression 2 0.82500 0.41250 2.725E+06 0.000
Residual Error 7 0.00000 0.00000
Total 9 0.82500
Regression Analysis: Sterile product versus time and temp
The regression equation is
Sterile =0.000303 + 0.00113 time + 0.000060 temp
Predictor Coef SE Coef T P
Constant 0.0003033 0.0002999 1.01 0.345
Time 0.00112859 0.00003939 28.65 0.000Time 0.00112859 0.00003939 28.65 0.000
Temp 0.00006005 0.00005694 1.05 0.327
S = 0.0003891 R-Sq = 100.0% R-Sq(adj) = 100.0%
Analysis of Variance
Source DF SS MS F P
Regression 2 0.82500 0.41250 2.725E+06 0.000
Residual Error 7 0.00000 0.00000
Total 9 0.82500
0.00050.0000-0.0005
1
0
-1
N
o
r
m
a
l

S
c
o
r
e
Residual
Normal Probability Plot of the Residuals
(response is Angle)
Use regression is to express and analyze a mathematical equation of describing a relationship. That is, it is
to fit a mathematical equation of describing a relationship between the “YY” and “X’sX’s”.
p-value < 0.05 :
Significant factor
R
2
and R
2
-adj are over 90% :
which indicates a potentially good fit

Regression Analysis: Sterile product versus Steam supply
The regression equation is
Sterile product = - 0,380 + 3,74 Steam supply
Predictor Coef SE Coef T P
Constant -0,3796 0,1693 -2,24 0,042
Steam Supply 3,744 1,153 3,25 0,006Steam Supply 3,744 1,153 3,25 0,006
S = 0,1510 R-Sq = 43,0% R-Sq(adj) = 38,9%
Analysis of Variance
Source DF SS MS F P
Regression 1 0,24032 0,24032 10,55 0,006
Residual Error 14 0,31905 0,02279
Total 15 0,55937
Regression Analysis: Sterile product versus Steam supply
The regression equation is
Sterile product = - 0,380 + 3,74 Steam supply
Predictor Coef SE Coef T P
Constant -0,3796 0,1693 -2,24 0,042
Steam Supply 3,744 1,153 3,25 0,006Steam Supply 3,744 1,153 3,25 0,006
S = 0,1510 R-Sq = 43,0% R-Sq(adj) = 38,9%
Analysis of Variance
Source DF SS MS F P
Regression 1 0,24032 0,24032 10,55 0,006
Residual Error 14 0,31905 0,02279
Total 15 0,55937
The P-value < 0.05
Reject Ho ; Accept ha
The P-value < 0.05
Reject Ho ; Accept ha
Comparing of Sterile product and steam supply to find what the factor’ level’s which influence
enormously by represent characterized variation “Y”“Y” by the total sum of square.
Analysis – Regression
D CIM
A
0.30.20.10.0-0.1-0.2-0.3
4
3
2
1
0
Residual
F
r
e
q
u
e
n
c
y
Histogram of the Residuals
(response is Cullet S)

Analysis – Chi-square D CIM
A
Since P-Value >> 0.05; there’s no significant
Effect between product sterile and factor.
Since P-Value >> 0.05; there’s no significant
Effect between product sterile and factor.
Conclusion:

◆
At least no one region is different, because a
dependence exists. (P > 0.05)

◆
It no appears that the dependence may exist with
Region 1 due to the large difference between the
observed and the expected values.(must
subtract the expected and observed values)
Conclusion:

◆
At least no one region is different, because a
dependence exists. (P > 0.05)

◆
It no appears that the dependence may exist with
Region 1 due to the large difference between the
observed and the expected values.(must
subtract the expected and observed values)
This Chi-Square is used to Test hypotheses about the frequency of occurrence of some event
happening with equal probability.
Chi-Square Test: matang, Stngh matang,
juice
Expected counts are printed below observed counts
Chi-Square contributions are printed below
expected counts
Stngh
matang matang juice Total
OK 1000 995 1013 3008
1000.01 993.03 1014.96
0.000 0.004 0.004
NG 3 1 5 9
2.99 2.97 3.04
0.000 1.308 1.269
Total 1003 996 1018 3017
Chi-Sq = 2.585, DF = 2, P-Value = 0.275
3 cells with expected counts less than 5.

Analysis –
two sample T-test
D CIM
A
Two-Sample T-Test and CI: Automatic, Chart
Two-sample T for Automatic vs Chart
N Mean StDev SE Mean
Automatic 12 14.70 1.47 0.42
Manual 12 14.13 2.19 0.63
Difference = mu (Automatic) - mu (Chart)
Estimate for difference: 0.562500
95% CI for difference: (-1.030754,
2.155754)
T-Test of difference = 0 (vs not =): T-
Value = 0.74 P-Value = 0.469 DF = 19
Hypothesis tests help to determine if a difference is real, or if it could be due to
chance
D
a
t
a
ChartAutomatic
19
18
17
16
15
14
13
12
11
Boxplot of Automatic, Chart
There is no statistically significant difference
if the confidence interval for m
1
- m
2
does
include 0.0.

9
6.5
97
90
107.5
Steam
Temp
Time
3
114
10
7
1215
0
Cube Plot (data means) for Salmonella
D CMAImprovement –
Response Surface Experiment
I
From the Main Effects Plot for the average of residue we
conclude:
• Temp has the greatest effect on average residue
• Time has a lesser effect on average residue
• Steam supply shows little or no effect (within the test range)
on the average residue
Main Effects Plots for Main Effects Plots for Time, Temp & Steam supplyTime, Temp & Steam supply Average and Standard Deviation of Residue. Average and Standard Deviation of Residue.
• Salmonella : 0
• Temp max : 97
o
C
• Time : 7.5 min
• Steam : 6.5 bar
Best
Condition:
Best
Condition:
M
e
a
n

o
f

S
a
lm
o
n
e
lla
10.07.5
10.0
7.5
5.0
9790
9.06.5
10.0
7.5
5.0
Time Temp
Steam
Main Effects Plot (data means) for Salmonella

Improvement –
Response Surface Experiment
D CMA
I
Contour Plot
Interpretation: to make sterile product (no salmonella) move towards the center corner of the
Contour Plot (samonella = 00). Read off potential “Time” and “Temp” values that will provide Salmonella
< 2.
Lines of target
response for
“0” Salmonella
1. Get to know the condition giving lower salmonella.
2. To get the regular response, we realize what variables
is important to control (temp & time)
3. Determine the level of independent variance needed for
getting salmonella 00 (When temp is approximately 97
o
C)
Time
T
e
m
p
10.09.59.08.58.07.5
97
96
95
94
93
92
91
90
Hold Values
Steam6.5
Salmonella
4.5- 7.0
7.0- 9.5
9.5- 12.0
12.0- 14.5
> 14.5
< 2.0
2.0- 4.5
Contour Plot of Salmonella vs Temp, Time

D CMA
I
Generally, main effect is more important than interaction. If interaction is regarded as a important thing,
then interaction can be used as a factor of interaction and another interaction might be confounded.
96
Salmonella
94
0
5
10
Temp
15
92
8
909
10Time
Hold Values
Steam6.5
Surface Plot of Salmonella vs Temp, Time
Improvement –
Response Surface Experiment

Improvement –
Result
D CMA
I
1 2 3 4 5 6
1.0
0.5
1.5
2.0
2.5
Z
S
h
ift
P
r
o
c
e
s
s
C
o
n
tr
o
l
Good
Poor
Technology
Good
Poor
Block A
Block C
Block B
Block D
Four Block Diagram
4.25 s
Z Shift
1.85 s
Improvement Result:
Saving Cost estimated: 2.7K U$/Year
0.500.250.00-0.25-0.50
Target USLLSL
Angle Line #2
Process Capability Analysis for Sealing
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
Ppk
Z.LSL
Z.USL
Z.Bench
Cpm
Cpk
Z.LSL
Z.USL
Z.Bench
StDev (Overall)
StDev (Within)
Sample N
Mean
LSL
Target
USL
63.70
46.70
17.00
41.55
30.86
10.69
0.00
0.00
0.00
1.30
4.14
3.91
3.83
1.34
1.34
4.25
4.01
3.94
0.124193
0.121123
84
0.014762
-0.500000
0.000000
0.500000
Exp. "Overall" PerformanceExp. "Within" PerformanceObserved PerformanceOverall Capability
Potential (Within) Capability
Process Data
Within
Overall

Control
DMAI
C
Process
Input
Controller
Controllable factors:
- Miss adjust causes
- Adjustable check
- Pad control
- Education
Group Member
Process CapabilityDesired
Output
X
Upper Control Limit
Lower Control Limit
●
●
Six Sigma Quality focuses on moving control upstream to the leverage input characteristic for Y.
If we can measure and control the vital few X’s, control of Y should be assured.
10050Subgroup 0
0.5
0.0
-0.5
S
a
m
p
l
e

M
e
a
n
Mean=0.001188
UCL=0.4384
LCL=-0.4360
1.0
0.5
0.0
S
a
m
p
l
e

R
a
n
g
e
1
11
R=0.2325
UCL=0.7596
LCL=0
Xbar/R Chart for Sealing Angle Line #2
Output
Process Standard Change

Contents
6
s

C
h
a
m
p
i
o
n

R
e
v
i
e
w
D
M
A
I
C
-
S
t
e
p

R
e
p
o
r
t
1.DEFINE
2.MEASURE
3.ANALYSIS
4.IMPROVEMENT
5.CONTROL
Optimizing Material DIO

‘06 6s Project Registration
PJT Name
P
e
r
io
d
Team
Name
Breakthrough
Main Improvement Object
(KPI) Current World Best Target
New Idea for Target Achievement
Team Formation, Related Department Involved
Name Dept.PositionMain Role
NECK POINT
How to do ?
Why ?
(* Selection Background)
Expected
Results
Quantitative
Qualitative
Optimizing Material D I O
1. JIT delivery system for press part
2. Door to door delivery for glass
3. Hub delivery system from Korea
4. Raw material issue control to process
5. Weekly stock taking
6. Minimize NG and rework stock in process
at end of the month
7. PO issued based on the latest production
plan
1. D I O is one of key performance indicators in
inventory management.
2. Good level of inventory will support production
line in effective and efficient way.
3. Fluctuated material D I O
Current Condition
USL : 2.9
LSL : -
Means : 2.58091
Sample N : 14
Z Bench : 1.03
- Warehouse Inventory Amount
D I O
( Days Inventory
Outstanding )
2.6 days 2.3 days
- Working In Process Inventory Amount
1,117
K
$
U
S
Current
1,050
Target - Continuous material supplies to production line
- All material used efficiently in production line
- Reduced warehouse and WIP inventory amount
1NG and rework stock in the end of month
2. Fluctuate production schedule
Just In Time Purchasing
Vendor Managed Inventory

K 67
3.53.02.52.01.5
USLUSL
Process Capability Analysis for C2
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
Ppk
Z.LSL
Z.USL
Z.Bench
Cpm
Cpk
Z.LSL
Z.USL
Z.Bench
StDev (Overall)
StDev (Within)
Sample N
Mean
LSL
Target
USL
200815.40
200815.40
*
151989.40
151989.40
*
214285.71
214285.71
*
0.28
*
0.84
0.84
*
0.34
*
1.03
1.03
0.380447
0.310413
14
2.58091
*
*
2.90000
Exp. "Overall" PerformanceExp. "Within" PerformanceObserved PerformanceOverall Capability
Potential (Within) Capability
Process Data
Within
Overall

Background
D
MAIC
One of the material inventory management control is DIO (Days Inventory Outstanding) that
has the formula:
Inventory Amount
Material DIO = ------------------------- X total days of current month
Sales Amount
The elements of material DIO are:
1.Warehouse Inventory (Raw Material)
2.Working In Process inventory (Semi Finished Goods )
3.Material In Transit Inventory
This large amount and high DIO have some effects :
 Risk in obsolescence, expired, lost, and
defect
 High inventory carrying cost
Current Material Inventory Condition :
Average Amount : $ K 1,117
D I O : 2.6 days
Optimizing inventory amount and D I O will bring material inventory management in a more
efficient cost

D
MAIC
X4X3X2X1Y X5
M. RATIO
Net BOM
Material Price
Src.
Performance
BOM Quantity
M. YIELD
CPT Price
Experience
Negotiation
Education
M. DIO
Material Loss
Net Req. Material
Material Cost Amt
Chemical
Others
Market Situation
Mat. Inv. Amt
Production Qty
Key Part
Sales Amt
Beginning Stock
Purchasing Qty
Ending Stock
Receive Amt
Total Days
In Transit Inv.
Warehouse Inv.
W I P Inv.
Sales Qty.
Current Month
Supply Condition
Glass
Mask
Press Part
Sub Mount
Process Part
Assy Part
Sales Target
Calculation Skill
Forecast Skill
Logic Tree

DM
AIC
Logic Tree
X4X3X2X1Y X5
Mat Inv Amt
In Transit Inv Delivery Sched.
Material
DIO
Sales Amt
Sales Qty
Price
Warehouse Inv Experience
A
Education
Total Days
Stock
Bulb
B
Part Stock
Assy Stock
Current Month
Sales Target
Production Capa
Graphite G 72 B
D Y
CMA
Glass
M/Assy
Simulation Skill
Marketing Nego
Market Situation
W I P Inventory
YS
TCL Prod.
Comp. Supply
Sub Month

D
M
AIC
This project with potential X-List will be focused to control Warehouse Inventory and working In
Process Inventory
X Level
2
X Level
2
D Y
D Y
Big Y
Big Y
Material DIO
Material DIO
Assy Stock
Assy Stock
Part Stock
Part Stock
P A D
P A D
Glass
Glass
Sub Mount
Sub Mount
Flat Mask
Flat Mask
Phosphor
Phosphor
G 72 B
G 72 B
X Level
1
X Level
1
Material Inv. Amt
Material Inv. Amt
W/house Inv.
W/house Inv.
WIP Inv.
WIP Inv.
X Level
3
X Level
3
X Level
4
X Level
4
Assy
Assy
CMA
CMA
good
good
Brainstorming Potential X

D
M
AIC
Gage R&R

StdDev Study Var %Study
Var
Source (SD) (5.15*SD) (%SV)
Total Gage R&R 0.109148 0.56211 16.58
Repeatability 0.109046 0.56159 16.56
Reproducibility 0.004723 0.02432 0.72
Operator 0.004723 0.02432 0.72
Part-To-Part 0.649363 3.34422 98.62
Total Variation 0.658472 3.39113 100.00
Number of Distinct Categories = 8
Misc:
Tolerance:
Reported by:
Date of study:
Gage name:
0
4
3
2
321
Xbar Chart by Operator
S
a
m
p
l
e

M
e
a
n
Mean=2.679
UCL=2.724
LCL=2.634
0
1.0
0.5
0.0
321
R Chart by Operator
S
a
m
p
l
e

R
a
n
g
e
R=0.02405
UCL=0.07857
LCL=0
1413121110 9 8 7 6 5 4 3 2 1
4
3
2
Part
Operator
Operator*Part Interaction
A
v
e
r
a
g
e
1
2
3
321
4
3
2
Operator
By Operator
1413121110 9 8 7 6 5 4 3 2 1
4
3
2
Part
By Part
%Contribution
%Study Var
%Tolerance
Part-to-PartReprodRepeatGage R&R
350
300
250
200
150
100
50
0
Components of Variation
P
e
r
c
e
n
t
Gage R&R (ANOVA) for Measure
To test validation of measurement , 3 (three) persons, twice inspection and 14 months calculation for
material D I O has carried out, and the result was acceptable.
The result of Gage R&R total is 16.58, the acceptance percentage is below 20 (<20), meanwhile the
result of measurement between <20, it accepted.
Gage R&R
<20%
Acceptable
Gage R&R
<20%
Acceptable
Gage R & R

Current Condition D
M
AIC
3.53.02.52.01.5
USLUSL
Process Capability Analysis for C2
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
Ppk
Z.LSL
Z.USL
Z.Bench
Cpm
Cpk
Z.LSL
Z.USL
Z.Bench
StDev (Overall)
StDev (Within)
Sample N
Mean
LSL
Target
USL
200815.40
200815.40
*
151989.40
151989.40
*
214285.71
214285.71
*
0.28
*
0.84
0.84
*
0.34
*
1.03
1.03
0.380447
0.310413
14
2.58091
*
*
2.90000
Exp. "Overall" PerformanceExp. "Within" PerformanceObserved PerformanceOverall Capability
Potential (Within) Capability
Process Data
Within
Overall
Z-Bench :
1.03

Block Diagram
DM
AIC
Z shift has been identified , the Z shift will be 0.19, regarding this issue the target
of the project is 4.5 sigma
A B
C
0.5
1.0
1.5
2.0
2.5
1 2 3 4 5 6
Z

s
h
if
t
P
r
o
c
e
s
s

C
o
n
t
r
o
l
Good
Poor
Poor Good
Z st
Technology
Position of DIO was column C ,
it was mean :
PROCESS CONTROL IS GOOD,
BUT TECHNOLOGY (METHOD)
IS BAD
EXPLANATION
Z st : 1.03
Z shift : Z st – Z lt
: 1.03 – 0.84
: 0.19
Z st : 1.03
Z shift : Z st – Z lt
: 1.03 – 0.84
: 0.19
D

DM
A
IC
The Pareto analysis has been done, the result shows that tube stock, Furnace, CMA, and Assy
have the highest contribution to WIP Assy Amount
WIP Assy Stock Analysis
O
th e rs
C
P
T D
o
n g b
a n g
M
o
u n t A
s s y
C
M
A
B
u lb
B
a
r e
T u b
e
14962 7725 33504 37174 48983165826
4.9 2.510.912.115.953.8
100.0 95.1 92.6 81.8 69.7 53.8
300000
200000
100000
0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
Avg WIP Stock 1Q '04
Analysis cetiEi
cetiEi
Bare TubeBare Tubecn
cn
BulbBulb

CMACMA
45 % conveyor stock ($ 78 K)
F’ce Stock
C/V StockGE aar
GE aar
Mount AssyMount Assy
Safety Stock to secure
supply
55 % ( $ 97 K) consists
Of :
- NG & rework,
- Pending Lot
- Remained Prod. Stock
Stock to keep production
2 shift Mount Assy Process
Pareto Chart for WIP Assy Inv.

DM
A
ICWIP Part Stock Analysis
O
t h e r s
P
h o
s p o
r
S
u b M
o u n t
G
la
s s
M
a
s k
7375 10535 32501 50418179511
2.6 3.811.618.064.0
100.0 97.4 93.6 82.0 64.0
200000
100000
0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
Pareto Chart for Desc.
The Pareto analysis result for WIP part stock shows that Mask Stock, Glass, Sub Mount and
Phosphor have the highest contribution to WIP part Amount
Analysis neEea
neEea
Flat MaskFlat Mask+neaa
+neaa
GlassGlass!EG
!EG
Sub MountSub Mount
86 % stock at PT YSI($ 156 K)
Stock
loading/7a%7t
/7a%7t
PhosphorPhosphor
Safety Stock for aging time
14 % stock at SM ( $ 23 K)
- Annealing
- Forming & Blackening
Mount Assy Process stock
High price
- Flat Mask (RM)
Pareto Chart for WIP Part Inv.

DM
A
ICWarehouse Stock Analysis
The Pareto analysis result for warehouse stock shows that Stock, Graphite, Phosphor and PAD
have the highest contribution to Warehouse Stock amount
Analysis >E9
>E9
D YD Y+te%7li
+te%7li
GraphiteGraphite/7a%7t
/7a%7t
PhosphorPhosphor
Inner Supply
Hardly PO revision
PO issued three month before/E E>
/E E>
P A DP A D
- Quality Problem
- Comp. Request to achieve
sales target
Decreasing consumptions
- Revised Production Plan
G
la
s s
P
A
D
P
h o
s p o
r
G
r a
p h i te
D
Y
1600717572396544144347520
9.910.824.425.629.3
100.0 90.1 79.3 54.8 29.3
150000
100000
50000
0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
Pareto Chart for C1
Blue Stock
So frequently
Pareto Chart for W/House Inv.
Minimum order 2304 kg

DM
A
IC
Not Selected as Vital View
Selected as Vital View
Item High Amount Stock Remarks
Selected as Vital View
Selected as Vital View
Selected as Vital View W I P
Part
NG and rework Stock
Mask Stock outsourcing process
Warehouse
Assy
Analysis Result
Glass Stock
Sub Mount
Remained production stock & Pending lot
conveyor stock
Selected as Vital View
Not Selected as Vital View
Y D Y
G 72 B & G 355
P A D
Selected as Vital View
Selected as Vital View
Phosphor
Selected as Vital View

DM
A
IC
Bottle Neck
1.High stock of NG, rework, pending lot in the end of month due to quality
problem, sourcing team can not fully controlled this situation. Actually, it’s depend
on process and quality performance.
Analysis Result
2. Outsourcing Mask Annealing process at Shin require more raw material
stock to keep production and secure supply.
3.Frequently change production plan for CIT and Y DY quality problem cause influence
high stock Y DY.

DMA
I
C
Improvement
Item Improvement Remarks
Mask Stock
- Identify and checking 3 days before closing
- Partial raw material delivery
Glass Stock
Stock
- Daily vendor managed inventory
- Communicate Stock to related dept and
push for action
- Working closely with PCT Team to input remained stock
In the end of month
- Pending Lot
- Remained Prod.
from March ‘04
Annealing
Outsourced

- Optimized in out stock control
- Daily input glass to production line
- Communicate and push process to minimize stock
- Check stock condition & make any necessary action
Y D Y
- Confirm I production plan
- Best effort to match PCT Production Plan & actual
- Just In Time purchasing
Phosphor
Graphite
- Improve Import delivery simulation skill
Weekly control
from March ‘04
Weekly control
from March ‘04
- Tightly control on ETD & ETA
- Maintain actual delivery performance on SRS
from March ‘04
from March ‘04

DMA
I
CImprovement Result
Result analysis after improvement actions : Z bench 2.60
2.92.72.52.32.11.91.71.5
USLUSL
Process Capability Analysis for DIO
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
Ppk
Z.LSL
Z.USL
Z.Bench
Cpm
Cpk
Z.LSL
Z.USL
Z.Bench
StDev (Overall)
StDev (Within)
Sample N
Mean
LSL
Target
USL
1307.83
1307.83
*
4651.78
4651.78
*
0.00
0.00
*
1.00
*
3.01
3.01
*
0.87
*
2.60
2.60
0.215663
0.249574
6
2.25093
*
*
2.90000
Exp. "Overall" PerformanceExp. "Within" PerformanceObserved PerformanceOverall Capability
Potential (Within) Capability
Process Data
Within
Overall
Z-Bench :
2.60

DMA
I
C Sigma Value
After improvement action, we can compare it with previous condition.
Improved Condition is better than Previous Condition.
PREVIOUS CONDITION
A B
C
0.5
1.0
1.5
2.0
2.5
1 2 3 4 5 6
Z

s
h
i
f
t
P
r
o
c
e
s
s

C
o
n
t
r
o
l
Good
Poor
Poor Good
Z st
Technology
IMPROVED CONDITION
Sigma=1.03
D
Z st : 2.60
Z shift : Z st – Z lt
: 2.60 – 3.01
: -0.41
Z st : 2.60
Z shift : Z st – Z lt
: 2.60 – 3.01
: -0.41
A B
C
0.5
1.0
1.5
2.0
2.5
1 2 3 4 5 6
Z

s
h
if
t
P
r
o
c
e
s
s

C
o
n
t
r
o
l
Good
Poor
Poor Good
Z st
Technology
D
Z st : 1.03
Z shift : Z st – Z lt
: 1.03 – 0.84
: 0.19
Z st : 1.03
Z shift : Z st – Z lt
: 1.03 – 0.84
: 0.19
Sigma=2.60

DMA
I
C Saving Cost
1,117
K
$
U
S
Current
1,050
Target
K 67
(6%)
TARGET
1,117
K
$
U
S
Before
1,075
After
K 42
(4%)
RESULT
Previous Average Material Inventory Amount : $ K 1,117
Current Average Material Inventory Amount : $ K 1,075
Saving Cost : $ K 42

DMAI
C
Control
Below check sheets are applied to ensure and maintain the material inventory DIO
stays optimized and some improvement activities stay controlled :
1. Mask daily inventory stock control at Shin
This is one of the application of vendor managed inventory (VMI)
2. Salvage glass daily input to process
3. Weekly Stock taking for warehouse and WIP (include Assy and Stock)
Desc. 31 1 2 3 4 5 6
F/MAS K 50001200001200001100001000008700080000
ANNEA 35447336122174019973233732515520373
FO RM 2550 1961 3170 5149 2090 2095 8300
BLAC K 7189 4841 7146 6972 6860 7207 2961
TO TAL 50186160414152056142094132323121457111634
F/MAS K 5000750007500075000750007500075000
ANNEA 2933 2933 2933 2933 2933 2933 2933
FO RM 0 0 0 0 0 0 0
BLAC K 2790 2790 2790 2790 2790 2790 2790
TO TAL 10723807238072380723807238072380723
14"
20"
1. 2.
3.
1 2 3 4 5 6 7 8 91 01 11 21 31 41 51 61 71 81 92 02 12 22 32 42 52 62 72 82 93 03 1 Tt l
14 0 0 067250416833602242243081543221680 0 03925041681680 0308078216816803361686242
20 0 0 0722662483100 0180722720 0 0 0720 0722080 01924162006462054864 3318
21 0 051225619206401923203201921283200 05763205121920 0 0384320641280 0 01285120
0 0512# #96241671004167247006184504880 0648712# #4323760 0884736# #360230088436014680
1 2 3 4 5 6 7 8 91 01 11 21 31 41 51 61 71 81 92 02 12 22 32 42 52 62 72 82 93 03 1 Tt l
142000 04004004000 040020020004000 0 02004002002004000 0 02002002000 02004005200
20 0 0810261902700900 01800 0 0 0 0 0 0 0 0 02700180270270180018090 2412
21 0 0405243162810 0814053741621622430 0324324324243810 04051620 02430811624667
200048664382357127005716055743425622430 05247245244434810270405542470470423046165212279
FUNNEL
PA NEL
JUMLA H
JUMLA H
Ta n g g a l
Ta n g g a l
1153-113V DY 14" LG STD 0 0 488 2,432 -2,4321.68205 0.00 820.84
2153-276F DY HARTONO/SA NKEN/V ESTEL 0 0 0 00.00000 0.00 0.00
33024GAFA01CMASK FLAT 21" MULTI 20,000 20,000 2,191 80,000 -60,0001.7221434,442.80 145.70
43040GA0001ABASE 20" 202,103 202,103 0 202,103 00.02246 4,539.23 0.00
53040GA0006ABASE 14" 210,000 210,000 0 210,000 00.03620 7,602.00 0.00
63210GBAA01AFRA ME SUPPORT 14" 0 0 4,284 0 00.12900 0.00 552.64
73210GBEA01AFRA ME SUPPORT 20" 0 0 1,920 0 00.58850 0.00 1,129.92
83210GBFA01AFRA ME SUPPORT 21" 0 0 3,120 0 00.63900 0.00 1,993.68
93300GB0001APLATE COMPENSATION 0 0 0 0 00.63900 0.00 0.00
103300GB0001BPLATE COMPENSATION 10,000 10,000 0 10,000 00.00428 42.80 0.00
113300GB0002APLATE COMPENSATION 20,000 20,000 0 20,000 00.00299 59.80 0.00
123300GC0001AB-S PLATE 20" 0 0 20,000 10,000 -10,0000.01220 0.00 244.00
133740GA0001ALEAD PROTECT 20" 12,500 12,500 11,500 12,500 00.01182 147.75 135.93
NoPart No Description Act.GdInv. Book PMS Ga pU/PRICEProce ss REMARKS
Amount
Process
Amount
INV

Attachment
Inventory and DIO Monthly Control 2004
I
n
v
e
n
t
o
r
y
(
K

$
)
INTR.
W/H
WIP
T/T
DESC. 1 2 3 4 5 6 7 8 9 10 11 12
15 62 84 135 82
303 333 432 469 394
706 786 522 597 509
1,024 1,181 1039 1201 984
Days 2.3 2.5 2.0 2.3 2.0
2. DIO
1024
1181
1039
1201
984
800
850
900
950
1000
1050
1100
1150
1200
1250
123456789101112
2.3
2.5
2.0 22.3
0
1
2
3
4
123456789101112
Targe
t
Actual
Targe
t
Actual

Attachment
Inventory and DIO Budget Control 2004
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Sales AmountTarget 12,060 12,176 12,569 12,256 12,290 12,322 12,432 12,879 12,874 12,197 10,740 11,551 146,346
Actual 13,784 13,668 15,892 15,388 15,423
In transitTarget 45 45 45 41 41 41 40 40 40 40 40 40 42
Actual 15 62 84 135 82
W/house Target 372 372 372 365 365 365 365 365 365 325 325 325 357
Actual 303 333 432 469 394
WIP Target 695 695 695 672 672 672 672 672 672 665 665 665 676
Actual 706 786 522 597 509
Total Target 1,112 1,112 1,112 1,078 1,078 1,078 1,077 1,077 1,077 1,030 1,030 1,030 1,074
Actual 1,024 1,181 1,039 1,201 984
DIO Target 2.9 2.6 2.7 2.6 2.7 2.6 2.7 2.6 2.5 2.6 2.9 2.8 2.7
Actual 2.30 2.51 2.03 2.34 1.98
Avg/TotalDesc.
2004

Attachment
1 2 3 4 5 6 7 8 9 10 11 12 1 2
In Transit 25 39 33 24 17 24 113 107 80 44 124 109 55 62 61
W I P 564 724 713 799 890 970 693 653 604 664 668 574 706 786 715
W/House 442 331 278 333 298 357 314 361 409 320 383 311 303 333 341
Total 1,0311,0941,0241,1561,2051,3511,1201,1211,0931,0291,175 9941,0641,1811,117
Sales Amt12,95313,09113,70112,08012,69812,32814,09914,39614,14714,38510,55113,11213,784 13,668 13,214
D I O 2.46752.33992.31692.87092.94123.28792.46322.41382.31862.21663.34032.35022.39362.50552.6205
Months
avgDESC.
0
0.5
1
1.5
2
2.5
3
3.5
4
1 2 3 4 5 6 7 8 9 1011 121314Avg
Material Inventory and DIO Control 2003 and 1 Q of 2004

Date Nov 24
th
, 2004
Process Technique Group
Prepared : Novi Muharam
Reduce LNG Usage
6
s

C
h
a
m
p
i
o
n

R
e
v
i
e
w
D
M
A
I
C

R
e
p
o
r
t
Contents
1. Define Step
2. Measure Step
3. Analysis Step
4. Improvement
5. Control

Background
CIAM
D
LNG (Liquid Natural Gas) is source of energy for combustion process in F’ce
Up to 3
rd
quarter LNG usage for exhaust furnace still higher, it’s about 5000 Nm
3
/day.
This project have a purpose to decreasing LNG usage in furnace
4500
5000
Current Current Target Target
LNG Usage
Unit:
(Nm
3
/day)
11%
How to do:
LNG & Air pressure system
Adjustment to find best ratio
Both of them.
Target Saving cost:
= (5000 –4500 )Nm
3
/day x 0.165 U$/Nm
3
x 30 x 12
= 500 Nm
3
/day x 0.165 x 30 x 12
= 29,700 U$/Years29,700 U$/Years
O
thers
E
lectric
O
2
N
2
LNG
0.00700.06050.12900.13090.1650
1.412.326.226.633.5
100.0 98.6 86.3 60.1 33.5
0.5
0.4
0.3
0.2
0.1
0.0
100
80
60
40
20
0
Defect
Count
Percent
Cum %
P
e
r
c
e
n
t
C
o
u
n
t
Energy Usage Price
Energy Unit Price (U$)

G/S
Load Exh
P-pipe
Tip off B.B.D
Exhaust Furnace
Zone #1 ~ #44
Unload
Robot
Robot
Control Panel
-F/F C/V
-Cart
In
Out
Keeping
Zone
Up Slope
Zone
Down Slope
Process Mapping
CIAM
D
CTQ Area : - LNG & Air Usage

LNG Ratio
Usage
Material LNG Pressure Ratio
Air Pressure Ratio
Machine TIC Temperature
RC Fan RPM Motor
Dumper Valve
Exh Blower Pressure
Big Y X
1
X
2
X
3
D CIA
M
Brainstormed Potential X List

GaR StdDev Study Var
%Study Var
Source (SD) (5,15*SD) (%SV)

Total Gage R&R 0,009704 0,04998 1,40

Repeatability 0,002582 0,01330 0,37

Reproducibility 0,009354 0,04817 1,35

Operator 0,002566 0,01321 0,37

Operator*Part 0,008995 0,04633 1,30

Part-To-Part 0,692590 3,56684 99,99

Total Variation 0,692658 3,56719 100,00

GaR StdDev Study Var
%Study Var
Source (SD) (5,15*SD) (%SV)

Total Gage R&R 0,009704 0,04998 1,40

Repeatability 0,002582 0,01330 0,37

Reproducibility 0,009354 0,04817 1,35

Operator 0,002566 0,01321 0,37

Operator*Part 0,008995 0,04633 1,30

Part-To-Part 0,692590 3,56684 99,99

Total Variation 0,692658 3,56719 100,00

Gage name:
Date of study:
Reported by:
Tolerance:
Misc:
Exhaust F'Ce Measurement
Oct 19th, 2005
Novi M
0
-1,5
-1,0
-0,5
0,0
0,5
Eng'r Gijo Maker 1 maker 2 PQC
Xbar Chart by Operator
S
a
m
p
l
e

M
e
a
n
Mean=-0,5057UCL=-0,5032LCL=-0,5082
0
0,000
0,005
0,010
Eng'r Gijo Maker 1 maker 2 PQC
R Chart by Operator
S
a
m
p
l
e

R
a
n
g
e
R=0,001333
UCL=0,004356
LCL=0
a b c d e f
-1,0
-0,5
0,0
0,5
Part
Operator
Operator*Part Interaction
A
v
e
r
a
g
e
Eng'r
Gijo_1
Gijo_2
Gijo_3
PQC
Eng'r GijoMaker 1maker 2PQC
-1,0
-0,5
0,0
0,5
Oper
Response By Operator
a b c d e f
-1,0
-0,5
0,0
0,5
Part
Response By Part
%Contribution
%Study Var
Gage R&RRepeat ReprodPart-to-Part
0
50
100
Components of Variation
P
e
r
c
e
n
t
Gage R&R (ANOVA) for Auto 14"
Gage R&R for Exhaust F’ce Line #1 Measurement
Gage R&R
D CIA
M
Gage R&R <20%
Acceptable
Gage R&R <20%
Acceptable

1086420
USLLSL
Exhaust F'ce 14" Capa'
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
Ppk
Z.LSL
Z.USL
Z.Bench
Cpm
Cpk
Z.LSL
Z.USL
Z.Bench
StDev (Overall)
StDev (Within)
Sample N
Mean
LSL
Target
USL
137981.99
90265.06
47716.94
14111.87
11722.77
2389.10
81081.08
81081.08
0.00
0.45
1.67
1.34
1.09
*
0.76
2.82
2.27
2.19
1.66305
0.98276
37
5.27297
2.50000
*
7.50000
Exp. "Overall" PerformanceExp. "Within" PerformanceObserved PerformanceOverall Capability
Potential (Within) Capability
Process Data
Within
Overall
Capability Process
1 2 3 4 5 6
1.0
0.5
1.5
2.0
2.5
Z
S
h
if
t
P
r
o
c
e
s
s
C
o
n
t
r
o
l
Good
Poor
Z st
Technology
GoodPoor
Block A
Block C
Block B
Block
D
A : Poor control, inadequate technology
B : Must control the process better, technology is fine
C : Process control is good, inadequate technology
D : World class
Four Block Diagram
D CIA
M

Regression Analysis: Press LNG versus Temp
The regression equation is
Press LNG = 3.7 + 21.9 Temp
Predictor Coef SE Coef T P
Constant 3.75 36.18 0.10 0.918
Temp 21.934 6.556 3.35 0.002
S = 64.96 R-Sq = 24.2% R-Sq(adj) = 22.1%
Analysis of Variance
Source DF SS MS F PP
Regression 1 47242 47242 11.19 0.0020.002
Residual Error 35 147702 4220
Total 36 194945
Regression Analysis: Press LNG versus Temp
The regression equation is
Press LNG = 3.7 + 21.9 Temp
Predictor Coef SE Coef T P
Constant 3.75 36.18 0.10 0.918
Temp 21.934 6.556 3.35 0.002
S = 64.96 R-Sq = 24.2% R-Sq(adj) = 22.1%
Analysis of Variance
Source DF SS MS F PP
Regression 1 47242 47242 11.19 0.0020.002
Residual Error 35 147702 4220
Total 36 194945
Analysis
Accept H
a
Fail to reject H
o
The p-value < 0.05, Fail to accept
Ho, which demonstrate statistical
significance (an equation with a
“good” fit)
The p-value < 0.05, Fail to accept
Ho, which demonstrate statistical
significance (an equation with a
“good” fit)
LNG
Pressure
Manometer
Gauge
D CIM
A
P-Value: 0.321
A-Squared: 0.414
Anderson-Darling Normality Test
N: 37
StDev: 1.65154
Average: 5.27297
9.28.27.26.25.24.23.22.2
.999
.99
.95
.80
.50
.20
.05
.01
.001
P
r
o
b
a
b
ilit
y
Qty LNG
Normal Probability Plot
Analysis of Temp furnace with result has significant effect to Pressure LNG
Hypothesis Analysis :
Ho : Temperature furnace has no significant effect to LNG Pressure
Ha : Temperature furnace has significant effect to LNG Pressure
Hypothesis Analysis :
Ho : Temperature furnace has no significant effect to LNG Pressure
Ha : Temperature furnace has significant effect to LNG Pressure

0100200300400500600700800900
95% Confidence Intervals for Sigmas
Bartlett's Test
Test Statistic: 3.282
P-Value : 0.858
Levene's Test
Test Statistic: 2.096
P-Value : 0.131
Factor Levels
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
75
0.1
0.2
0.5
0.6
1.7
2.0
3.0
4.0
5.0
6.0
6.8
7.0
7.3
8.0
9.0
9.1
9.3
9.8
11.0
11.6
11.8
12.0
12.9
13.5
14.2
14.6
Test for Equal Variances for LNG Ratio
One-way ANOVA: Air Press, Press LNG,
Temp
Analysis of Variance
Source DF SS MS F PP
Factor 2 1769458 884729 118.10 0.0000.000
Error 108 809084 7492
Total 110 2578542
Individual 95% CIs For Mean
Based on Pooled St Dev
Level N Mean StDev --+---------+---------+---------+----
Air Pres 37 424.73 73.53 (-*--)
Press LN 37 117.00 73.04 (--*--)
Temp 2 37 297.54 108.32 (--*--)
--+---------+---------+---------+----
Pooled StDev = 86.55 100 200 300 400
One-way ANOVA: Air Press, Press LNG,
Temp
Analysis of Variance
Source DF SS MS F PP
Factor 2 1769458 884729 118.10 0.0000.000
Error 108 809084 7492
Total 110 2578542
Individual 95% CIs For Mean
Based on Pooled St Dev
Level N Mean StDev --+---------+---------+---------+----
Air Pres 37 424.73 73.53 (-*--)
Press LN 37 117.00 73.04 (--*--)
Temp 2 37 297.54 108.32 (--*--)
--+---------+---------+---------+----
Pooled StDev = 86.55 100 200 300 400
T
e
m
p

L
N
G
P
r
e
s
s
A
ir

P
r
e
s
600
500
400
300
200
100
0
Boxplots of Air Press - Temp
(means are indicated by solid circles)
Analysis
D CIM
A
Analysis of Temp furnace with result has significant effect to LNG & Air press
Hypothesis Analysis :
Ho : Temperature furnace has no significant effect to LNG & Air Pressure
Ha : Temperature furnace has significant effect to LNG & Air Pressure
Hypothesis Analysis :
Ho : Temperature furnace has no significant effect to LNG & Air Pressure
Ha : Temperature furnace has significant effect to LNG & Air Pressure

Regression Analysis: LNG Press versus RPM
Motor
The regression equation is
RPM = 1448 + 0.0344 LNG Press
Predictor Coef SE Coef T P
Constant 1448.00 5.02 288.20 0.000
LNG press 0.03440 0.01568 2.19 0.035
S = 8.711 R-Sq = 12.1% R-Sq(adj) = 9.6%
Analysis of Variance
Source DF SS MS F PP
Regression 1 365.11 365.11 4.81 0.0350.035
Residual Error 35 2655.97 75.88
Total 36 3021.08
Regression Analysis: LNG Press versus RPM
Motor
The regression equation is
RPM = 1448 + 0.0344 LNG Press
Predictor Coef SE Coef T P
Constant 1448.00 5.02 288.20 0.000
LNG press 0.03440 0.01568 2.19 0.035
S = 8.711 R-Sq = 12.1% R-Sq(adj) = 9.6%
Analysis of Variance
Source DF SS MS F PP
Regression 1 365.11 365.11 4.81 0.0350.035
Residual Error 35 2655.97 75.88
Total 36 3021.08
RC Fan
Rotation
Analysis
Accept H
a
Fail to reject H
o
D CIM
A
P-Value: 0.097
A-Squared: 0.624
Anderson-Darling Normality Test
N: 37
StDev: 19.1748
Average: 1462.22
14901480147014601450144014301420
.999
.99
.95
.80
.50
.20
.05
.01
.001
P
r
o
b
a
b
i
l
i
t
y
RPM
Normal Probability Plot
Analysis of RPM motor with result has significant effect to Pressure LNG
Hypothesis Analysis :
Ho : RPM motor has no significant effect to LNG Pressure
Ha : RPM motor has significant effect to LNG Pressure
Hypothesis Analysis :
Ho : RPM motor has no significant effect to LNG Pressure
Ha : RPM motor has significant effect to LNG Pressure
The p-value < 0.05, Fail to accept Ho, which
demonstrate statistical significance (an
equation with a “good” fit)
The p-value < 0.05, Fail to accept Ho, which
demonstrate statistical significance (an
equation with a “good” fit)

Analysis
Regression Analysis: LNG Press Versus Air Blower
The regression equation is
Air Blower = 475 - 0.169 LNG Press
Predictor Coef SE Coef T P
Constant 474.97 35.14 13.52 0.000
Air Blower -0.1689 0.1111 -1.52 0.138
S = 72.23 R-Sq = 6.2% R-Sq(adj) = 3.5%
Analysis of Variance
Source DF SS MS F PP
Regression 1 12044 12044 2.31 0.1380.138
Residual Error 35 182604 5217
Total 36 194647
Regression Analysis: LNG Press Versus Air Blower
The regression equation is
Air Blower = 475 - 0.169 LNG Press
Predictor Coef SE Coef T P
Constant 474.97 35.14 13.52 0.000
Air Blower -0.1689 0.1111 -1.52 0.138
S = 72.23 R-Sq = 6.2% R-Sq(adj) = 3.5%
Analysis of Variance
Source DF SS MS F PP
Regression 1 12044 12044 2.31 0.1380.138
Residual Error 35 182604 5217
Total 36 194647
Exhaust F’ce
Blower
Fail to accept H
a
Reject H
o
D CIM
A
Average: 424.730
StDev: 73.5314
N: 37
Anderson-Darling Normality Test
A-Squared: 0.200
P-Value: 0.875
300 400 500 600
.001
.01
.05
.20
.50
.80
.95
.99
.999
P
r
o
b
a
b
i
l
i
t
y
Blower
Normal Probability Plot
Analysis of Air Blower with result has no significant effect to Pressure LNG
Hypothesis Analysis :
Ho : RPM motor has significant effect to LNG Pressure
Ha : RPM motor has no significant effect to LNG Pressure
Hypothesis Analysis :
Ho : RPM motor has significant effect to LNG Pressure
Ha : RPM motor has no significant effect to LNG Pressure

Regression Analysis: LNG Press versus Damper
The regression equation is
Damper = 428 - 0.43 LNG Press
Predictor Coef SE Coef T P
Constant 427.96 24.17 17.70 0.000
LNG press -0.425 2.745 -0.15 0.878
S = 74.55 R-Sq = 0.1% R-Sq(adj) = 0.0%
Analysis of Variance
Source DF SS MS F PP
Regression 1 133 133 0.02 0.8780.878
Residual Error 35 194514 5558
Total 36 194647
Regression Analysis: LNG Press versus Damper
The regression equation is
Damper = 428 - 0.43 LNG Press
Predictor Coef SE Coef T P
Constant 427.96 24.17 17.70 0.000
LNG press -0.425 2.745 -0.15 0.878
S = 74.55 R-Sq = 0.1% R-Sq(adj) = 0.0%
Analysis of Variance
Source DF SS MS F PP
Regression 1 133 133 0.02 0.8780.878
Residual Error 35 194514 5558
Total 36 194647
Analysis
Damper
2001000-100
2
1
0
-1
-2
N
o
r
m
a
l

S
c
o
r
e
Residual
Normal Probability Plot of the Residuals
(response is Damper)
Fail to accept H
a
Reject H
o
D CIM
A
Analysis of Damper with result has no significant effect to Pressure LNG
Hypothesis Analysis :
Ho : RPM motor has significant effect to LNG Pressure
Ha : RPM motor has no significant effect to LNG Pressure
Hypothesis Analysis :
Ho : RPM motor has significant effect to LNG Pressure
Ha : RPM motor has no significant effect to LNG Pressure

Analysis Resume
D CIM
Select most effected factor in Regression and ANOVA with P value < 0.05Select most effected factor in Regression and ANOVA with P value < 0.05
Factor Detail Analysis Content Result ConclusionAnalysis Tool
Selected as
vital few
P < 0.05
P < 0.05
Selected as
vital few
LNG Ratio
Pressure and quantity
adjustment
For f’ce combustions
P < 0.05
Selected as
vital few
Sample test for kind of air blower
effect to gas ratio.
RPM Motor
With Gas
Heating result measurement
inside
Furnace.
P > 0.05
Not
selected as
vital few
Air & LNG
Pressure
Checking both of pressure
Compare with temperature result
Air blower
With Gas
Sample test for damper setting
for
Each position
P > 0.05
Not
selected as
vital few
Damper
With Gas
A
Regression
Regression
Regression
Regression
ANOVA

Bottle Neck
D CIM
A
From furnace build until now, this furnace have never been done by
Cleaning inside of f’ce
burner
Dilution
air
Gas
Exhaust
duct
To many Carbon (C)
To make efficiency,
Need cleaningNeed cleaning inside of f’ce
From Carbon (C) result of combustion.

Response Surface Regression: Result versus LNG Press, LNG
Qty, ...
The analysis was done using coded units.
Estimated Regression Coefficients for Result
Term Coef SE Coef T P
Constant 115.3 37.23 3.098 0.008
Block -25.0 17.44 -1.436 0.173
LNG Pres 108.1 18.39 5.878 0.000
LNG Qty -137.5 18.39 -7.478 0.000
AIR Pres -79.2 18.39 -4.310 0.001
RPM 3.9 18.39 0.213 0.834
LNG Pres*LNG Pres 2.4 17.20 0.142 0.889
LNG Qty*LNG Qty 78.6 17.20 4.568 0.000
AIR Pres*AIR Pres 26.9 17.20 1.566 0.140
RPM*RPM -10.3 17.20 -0.600 0.558
LNG Pres*LNG Qty -76.0 22.52 -3.375 0.005
LNG Pres*AIR Pres -51.5 22.52 -2.287 0.038
LNG Pres*RPM 3.5 22.52 0.155 0.879
LNG Qty*AIR Pres 39.2 22.52 1.743 0.103
LNG Qty*RPM -2.5 22.52 -0.111 0.913
AIR Pres*RPM -1.7 22.52 -0.078 0.93
S = 90.08 R-Sq = 91.7% R-Sq(adj) = 82.8%
Response Surface Regression: Result versus LNG Press, LNG
Qty, ...
The analysis was done using coded units.
Estimated Regression Coefficients for Result
Term Coef SE Coef T P
Constant 115.3 37.23 3.098 0.008
Block -25.0 17.44 -1.436 0.173
LNG Pres 108.1 18.39 5.878 0.000
LNG Qty -137.5 18.39 -7.478 0.000
AIR Pres -79.2 18.39 -4.310 0.001
RPM 3.9 18.39 0.213 0.834
LNG Pres*LNG Pres 2.4 17.20 0.142 0.889
LNG Qty*LNG Qty 78.6 17.20 4.568 0.000
AIR Pres*AIR Pres 26.9 17.20 1.566 0.140
RPM*RPM -10.3 17.20 -0.600 0.558
LNG Pres*LNG Qty -76.0 22.52 -3.375 0.005
LNG Pres*AIR Pres -51.5 22.52 -2.287 0.038
LNG Pres*RPM 3.5 22.52 0.155 0.879
LNG Qty*AIR Pres 39.2 22.52 1.743 0.103
LNG Qty*RPM -2.5 22.52 -0.111 0.913
AIR Pres*RPM -1.7 22.52 -0.078 0.93
S = 90.08 R-Sq = 91.7% R-Sq(adj) = 82.8%
Analysis of Variance for Result
Source DF Seq SS Adj SS Adj MS F P
Blocks 1 16733 16733 16733 2.06 0.173
Regression 14 1238887 1238887 88492 10.91 0.000
Linear 4 885220 885220 221305 27.27 0.000
Square 4 193821 193821 48455 5.97 0.005
Interaction 6 159846 159846 26641 3.28 0.031
Residual Error 14 113594 113594 8114
Lack-of-Fit 10 113594 113594 11359 * *
Pure Error 4 0 0 0
Analysis of Variance for Result
Source DF Seq SS Adj SS Adj MS F P
Blocks 1 16733 16733 16733 2.06 0.173
Regression 14 1238887 1238887 88492 10.91 0.000
Linear 4 885220 885220 221305 27.27 0.000
Square 4 193821 193821 48455 5.97 0.005
Interaction 6 159846 159846 26641 3.28 0.031
Residual Error 14 113594 113594 8114
Lack-of-Fit 10 113594 113594 11359 * *
Pure Error 4 0 0 0
Improvement
The Improve phase identifies a solution and confirms that the proposed solution will meet or
exceed the improvement goals of the project.
D CAM
I

10
-100
5
0
ratio
0
500
100
1000
200
300
0
400
Result
500LNG Press
Surface Plot of Result
Hold values: AIR Pres: 452.5 RPM: 1459.0
Improvement
Optimum condition when:
-LNG pressure 60.5 mmH
2
O
-Ratio of gas 4.5
-Air Pressure 452.5 mmH
2
O
-RPM Motor 1459 rpm

Optimum condition when:
-LNG pressure 60.5 mmH
2
O
-Ratio of gas 4.5
-Air Pressure 452.5 mmH
2
O
-RPM Motor 1459 rpm

D CAM
I

D CAM
Improvement
7.56.55.54.53.52.5
USLLSL
Exhaust F'ce 14" Capability
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
PPM Total
PPM > USL
PPM < LSL
Ppk
Z.LSL
Z.USL
Z.Bench
Cpm
Cpk
Z.LSL
Z.USL
Z.Bench
StDev (Overall)
StDev (Within)
Sample N
Mean
LSL
Target
USL
1.71
0.00
1.71
0.07
0.00
0.07
0.00
0.00
0.00
1.55
4.64
6.12
4.64
*
1.76
5.27
6.94
5.27
0.464622
0.409704
37
4.65757
2.50000
*
7.50000
Exp. "Overall" PerformanceExp. "Within" PerformanceObserved PerformanceOverall Capability
Potential (Within) Capability
Process Data
Within
Overall
1 2 3 4 5 6
1.0
0.5
1.5
2.0
2.5
Z
S
h
if
t
P
r
o
c
e
s
s
C
o
n
t
r
o
l
Good
Poor
Z st
Technology
GoodPoor
Block A
Block C
Block B
Block D
Four Block Diagram
I

4500
5000
Current Current Target Target
LNG Usage
Unit: (Nm
3
/day)
8%
D CAM
Improvement
I
4600
Result Result
92%92% Cost Saving:
= (5000 – 4600 )Nm
3
/day x 0.165 U$/Nm
3
x 30 x 12
= 400 Nm
3
/day x 0.165 x 30 x 12
= 23,760 U$/years23,760 U$/years
Improvement Result

Control Plan
Control
Item Period Gauge Method Chart Type PIC
LNG Press
Each Zone
Weekly Digital
Manometer
Use check sheet Xbar - R
PQC
Leader exh
Air Press
Each Zone
Weekly Digital
Manometer
Use check sheet Xbar - R
PQC
Leader exh
LNG meter
control
Daily Visual Use check sheet Xbar - R
Leader exh
RPM
Motor
Weekly Tachometer Use check sheet Xbar - R
PQC
D IAM
C

Process standard changes:
D IAM
Control
F’ce
Control
LNG Meter
Daily check
Record
Data
F’ce
Control
LNG Meter
Daily check
Record
Data
LNG & AIR Press
Weekly check
Standard
Record
Data
Setting Gas
Ratio
OK
NG
Record
Data
Process Control
Weekly check sheet for Gas measurement
Standard:
-LNG pressure 60.5 ± 10 mmH
2
O
-Ratio of gas 4.5
-Air Pressure 452.5 ± 50 mmH
2
O
Standard:
-LNG pressure 60.5 ± 10 mmH
2
O
-Ratio of gas 4.5
-Air Pressure 452.5 ± 50 mmH
2
O
PIC:
Leader Exh
PIC:
PQC Exh
Before Before After After
C
PIC:
Leader Exh

Six Sigma and Its Implementation

About This Presentation

Slide Content

Tags

Categories

Download

Quick Actions

Statistics

Related Slideshows

Six Sigma and Its Implementation

About This Presentation

Slide Content

Slide 1

Slide 2

Slide 3

Slide 4

Slide 5

Slide 6

Slide 7

Slide 8

Slide 9

Slide 10

Slide 11

Slide 12

Slide 13

Slide 14

Slide 15

Slide 16

Slide 17

Slide 18

Slide 19

Slide 20

Slide 21

Slide 22

Slide 23

Slide 24

Slide 25

Slide 26

Slide 27

Slide 28

Slide 29

Slide 30

Slide 31

Slide 32

Slide 33

Slide 34

Slide 35

Slide 36

Slide 37

Slide 38

Slide 39

Slide 40

Slide 41

Slide 42

Slide 43

Slide 44

Slide 45

Slide 46

Slide 47

Slide 48

Slide 49

Slide 50

Slide 51

Slide 52

Slide 53

Slide 54

Slide 55

Slide 56

Slide 57

Slide 58

Slide 59

Slide 60

Slide 61

Slide 62

Slide 63

Slide 64

Slide 65

Slide 66

Slide 67

Slide 68

Slide 69

Slide 70

Slide 71

Slide 72

Slide 73

Slide 74

Slide 75

Slide 76

Slide 77