Transcript Six-Sigma by Day Release

INDUSTRIAL STATISTICS
RESEARCH UNIT
We are based in the School of Mechanical and
Systems Engineering, University of Newcastle upon
Tyne, England
Our work we do can be broken
down into 3 main categories:
• Consultancy
• Training
• Major Research Projects
All with the common goal of promoting quality
improvement by implementing statistical
techniques
European Research Projects
• The Unit has provided the statistical input into many
major European projects - Examples include • Assessing steel rail reliability
• Testing fire-fighter’s boots for safety
• Calibsensory - Effect on food of the taints
and odours in packaging materials
• Pro-Enbis - Network of Six-Sigma and other
statistical practitioners around Europe
• Kensys - Kansei Engineering
Six-Sigma Basics
Basics
• Effective application of statistical tools within
a structured methodology
• Repeated application of Continuous
Improvement strategy to individual projects
• Projects deliberately selected to have a
substantial impact on the ‘bottom line’
Emperical Approach
A scientific and practical method to achieve
improvements in a company
Scientific:
• Structured approach.
• Assuming quantitative data.
”Show me
the money”
Practical:
• Emphasis on financial result.
• Start with the voice of the customer.
“Show me
the data”
Where can Statistical techniques be applied?
Service
Design
Management
Purchase
Administration
Statistical
Methods
Production
IT
Quality
Depart.
HRM
M&S
Using statistics can
integrate all of these issues
Knowledge
Management
Statisticians
Technical
Skills
CD
Course
delegates
Advanced
course
delegates
ACD
Quality Improvement
Facilitators
Soft Skills
Focus
• Accelerating fast breakthrough
performance
• Significant financial results in 4-8
months
• Results first, then change!
Improvement cycle
• PDCA cycle
Plan
Act
Do
Check
11
Alternative interpretation (Six Sigma
structure)
Prioritise (Define)
Hold gains
(Control)
Measure (M)
Improve (I)
Problem solve (Analyse - Improve)
Interpret
(Analyse)
Scientific method (after Box)
Data
Facts
INDUCTION
Theory
Hypothesis
Conjecture
Idea
Model
INDUCTION
DEDUCTION
Plan
Act
Do
Check
DEDUCTION
The “Success” of Change Programs?
“Performance improvement efforts …
have as much impact on
operational and financial results as a
ceremonial rain dance has on the weather”
Schaffer and Thomson,
Harvard Business Review (1992)
Change Management:
Two Alternative Approaches
Activity Centered
Programs
Change
Management
Result Oriented
Programs
Reference: Schaffer and Thomson, HBR, Jan-Feb. 1992
ISO 9000
Data
Deduction Induction
Hypothesis
No Checking with Empirical Evidence, No
Learning Process
Result Oriented Programs
• Project based
•
•
•
•
•
Experimental
Guided by empirical evidence
Measurable results
Easier to assess cause and effect
Cascading strategy
ISRU Training
Open and in-House courses
• Six-Sigma
• TPM
Distance Learning
Essentially a further learning resource for statistical tools and
methodology
Six-Sigma by Day Release
The Plan
Plan
Plan
Week 1
Mon
Tues
Wed
Thurs Fri
Day1
Plan
Plan
Mon
Tues
Wed
Thurs Fri
Day1
Week 1
Week 2
Day1
Day2
Plan
Plan
Mon
Tues
Wed
Thurs Fri
Day1
Week 1
Week 2
Week 3
Day1
Day1
Day2
Day2
Day3
Plan
Plan
Mon
Tues
Wed
Thurs Fri
Day1
Week 1
Day1
Day2
Day1
Day2
Day3
Day2
Day3
Day4
Week 2
Week 3
Week 4
Day1
Plan
Plan
Mon
Tues
Wed
Thurs Fri
Week 1
Day1
Week 2
Week 3
Day1
Day2
Day1
Day2
Day3
Week 4
Day1
Day2
Day3
Day4
Week 5 Day1
Day2
Day3
Day4
Day5
Plan
Plan
Mon
Tues
Wed
Week 2
Week 3
Thurs Fri
Day1
Day2
Day1
Day2
Day3
Week 4
Day1
Day2
Day3
Day4
Week 5 Day1
Day2
Day3
Day4
Day5
Week 6 Day2
Day3
Day4
Day5
Rest
Plan
Plan
Mon
Tues
Wed
Thurs Fri
Week 3
Day1 Day2 Day3
Week 4
Day1 Day2 Day3 Day4
Week 5 Day1 Day2 Day3 Day4 Day5
Week 6 Day2 Day3 Day4 Day5 Rest
Week 7 Day 3 Day4 Day5 Rest
Day 6
Plan
Plan
Week 4
Mon
Tues
Wed
Thurs Fri
Day1 Day2 Day3 Day4
Week 5 Day1 Day2 Day3 Day4 Day5
Week 6 Day2 Day3 Day4 Day5 Rest
Week 7 Day 3 Day4 Day5 Rest
Week 8 Day 4 Day5 Rest
Day6
Day6 Day7
Plan 2
Plan
Week 1
Mon
Tues
Wed
Thurs Fri
Day1
Plan 2
Plan
Mon
Tues
Wed
Week 1
Week 2
Thurs Fri
Day1
Day1
Day2
Plan 2
Plan
Mon
Tues
Wed
Week 1
Thurs Fri
Day1
Week 2
Day1
Day2
Week 3 Day 1
Day2
Day3
Plan 2
Plan
Mon
Tues
Wed
Week 1
Thurs Fri
Day1
Week 2
Day1
Day2
Week 3 Day 1
Day2
Day3
Week 4 Day2
Day3
Day4
Plan 2
Plan
Mon
Tues
Wed
Week 1
Thurs Fri
Day1
Week 2
Day1
Day2
Week 3 Day 1
Day2
Day3
Week 4 Day2
Day3
Day4
Week 5 Day3
Day4
Rest
Plan 2
Plan
Mon
Tues
Wed
Thurs Fri
Week 2
Day1
Day2
Week 3 Day 1
Day2
Day3
Week 4 Day2
Day3
Day4
Week 5 Day3
Day4
Rest
Week 6 Day4
Rest
Day5
Plan 2
Plan
Mon
Tues
Wed
Thurs Fri
Week 3 Day 1
Day2
Day3
Week 4 Day2
Day3
Day4
Week 5 Day3
Day4
Rest
Week 6 Day4
Rest
Day5
Week 7 Rest
Day5
Day6
Different Belts
• Yellow Belts - 5 days
– days - 1 - 3 - 8 - 12 - 17
• Green Belts - 10 days
– days - 1 - 2 - 3 - 4 - 8 - 9 - 12 - 13 - 17 - 18
• Black belts - 20 days
Additional Project Support
• Project support - on-going
• Yellow Belts - 1 day
• Green Belts - 2 days
• Black-Belts - 4 days
Project Funding
• SMEs - Training free! Just pay £150
per delegate per support day - green
belts = £300 for complete course.
• Non-SMEs - half usual cost
– e.g. green belt is £1,200 per man - by
day release in this way.
End
Six-Sigma Case study
Case study: project selection
Coffee
beans
Roast
Cool
Grind
Pack
Sealed
coffee
Savings:
-Savings on rework and scrap
-Water costs less than coffee
Potential savings:
500 000 Euros
Moisture
content
Case study: Measure
1. Select the Critical to Quality (CTQ)
characteristic
2. Define performance standards
3. Validate measurement system
Case study: Measure
1. CTQ
Moisture contents of
roasted coffee
2. Standards
- Unit: one batch
- Defect: Moisture% > 12.6%
Case study: Measure
3. Measurement reliability
Gauge R&R study
Measurement system
too unreliable!
So fix it!!
Case study: Analyse
Analyse
4. Establish product capability
5. Define performance
objectives
6. Identify influence factors
Improvement opportunities
USL
USL
CTQ
CTQ
CTQ
CTQ
Diagnosis of problem
Discovery of causes
Man
Machine
Material
6. Identify factors
-Brainstorming
-Exploratory data analysis
Roasting
machines
Batch
size
Moisture%
Amount of
added water
Reliability
of Quadra Beam
Weather
conditions
Method
Measurement
Mother
Nature
Discovery of causes
Regelkaart
voor for
Vocht%
Control chart
moisture%
5.2
1
Individual Value
1
1
3.0SL=4.410
4.2
X=3.900
-3.0SL=3.390
3.2
0
10
20
30
40
Observation Number
50
A case study
Potential influence factors
- Roasting machines (Nuisance variable)
- Weather conditions (Nuisance variable)
- Stagnations in the transport system
(Disturbance)
- Batch size (Nuisance variable)
- Amount of added water (Control
variable)
Case study: Improve
Improve
7. Screen potential causes
8. Discover variable
relationships
9. Establish operating
tolerances
Case study: Improve
7. Screen potential causes
- Relation between humidity and
moisture% not established
- Effect of stagnations confirmed
- Machine differences confirmed
8. Discover variable relationships
Design of Experiments (DoE)
Experimentation
How do we often conduct experiments?
Possible settings for X2
Experiments are run based on: Intuition
Knowledge
Experience
Power
Emotions
X
X
X: Settings with which
an experiment is run.
X
X
X
X
X
Possible settings for X1
Actually:
• we’re just trying
• unsystematical
• no design/plan
Experimentation
A systematical experiment: Organized / discipline
One factor at a time
Other factors kept constant
Possible settings for X2
Procedure:
X
X: First vary X1; X2 is kept constant
X
X
X X X X X X XO X
O: Optimal value for X1.
X
X
X
X
X
Possible settings for X1
X: Vary X2; X1 is kept constant.
: Optimal value (???)
Design of Experiments (DoE)
One factor (X)
X1
low
2
1
high
Two factors (X’s)
Three factors (X’s)
high
high
X2
2
2
2
X2
low
X1
high
X3
low
X1
high
3
A case study: Experiment
Surface Plot of Moisture
Experiment:
14
Y: moisture%
X1: Water (liters)
X2: Batch size (kg)
13
12
Moisture
11
110
10
105
600
100
610
Batch size
620
630
95
640
Water
A case study
9. Establish operating tolerances
Feedback adjustments for influence
of weather conditions
A case study: feedback adjustments
4.35
4.25
4.15
4.05
Moisture% Vocht%
without adjustments
989
937
885
833
781
729
677
625
573
521
469
417
365
313
261
209
157
105
53
1
3.95
A case study: feedback adjustments
4.35
4.25
4.15
4.05
Controlled Vocht%
Moisture%
with adjustments
989
937
885
833
781
729
677
625
573
521
469
417
365
313
261
209
157
105
53
1
3.95
Case study: Control
Control
10. Validate measurement
system (X’s)
11. Determine process
capability
12. Implement process
controls
Results
Before
long-term = 0.532
ProcessCapability
CapabilityAnalysis
Analysisfor
forMoisture
Moisture
Process
Objective
USL
USL
Process Data
Process Data
USL
12.6000
USL
13.0000
Target
*
Target
*
LSL
*
LSL
9.0000
Mean
11.0026
Mean
10.9921
Sample N
490
Sample N
200
StDev (Within) 0.335675
StDev (Within) 0.105808
StDev (Overall) 0.531635
StDev (Overall) 0.102497
long-term < 0.280
Within
Within
Overall
Overall
Result
Potential (Within) Capability
Potential
(Within) Capability
Cp
*
Cp
6.30
CPU
1.54
CPU
6.33
CPL
*
CPL
6.28
Cpk
1.54
long-term < 0.100
Cpk
6.28
Cpm
*
Cpm
*
Overall Capability
Pp Overall Capability *
PPU
0.96
Pp
6.50
9
9
10
10
Observed Performance
PPM
< LSL Performance *
Observed
PPM
0.00
PPM >< USL
LSL
0.00
11
11
12
12
Exp. "Within" Performance
PPM
LSL Performance*
Exp. <"Within"
PPM
1.79
PPM >< USL
LSL
0.00
13
13
Exp. "Overall" Performance
PPM
LSL Performance*
Exp. <"Overall"
PPM >
1987.68
< USL
LSL
0.00
Benefits
Benefits of this project
long-term < 0.100
Ppk = 1.5
This enables us to increase the mean to
12.1%
Per 0.1% coffee: 100 000 Euros saving
Benefits of this project:
1 100 000 Euros per year
Approved by controller
Case study: control
12. Implement process controls
- SPC control loop
- Mistake proofing
- Control plan
- Audit schedule
Project closure
- Documentation of the results and
data.
- Results are reported to involved
persons.
- The follow-up is determined
Approach to this project
- Step-by-step approach.
- Constant testing and double checking.
- No problem fixing, but: explanation  control.
- Interaction of technical knowledge and
experimentation methodology.
- Good research enables intelligent decision
making.
- Knowing the financial impact made it easy to find
priority for this project.
Re-cap
• Structured approach – roadmap
• Systematic project-based improvement
• Plan for “quick wins”
– Find good initial projects - fast wins
• Publicise success
– Often and continually - blow that trumpet
• Use modern tools and methods
• Empirical evidence based improvement