NPTEL Taguchi Methods

Quality Engineering and
Taguchi Methods
Copyright Tapan Bagchi 1

“Robust” Chocolate Bars are
better!
Poor
performance
Robust
Plasticity
performance
Ambient
Temperature
Taguchi Methods
(or Quality Engineering
or Robust Design)
Focus is on reducing variability of response

to maximize robustness, generally achieved
through Orthogonal Array Experiments

The Genesis of DOE
Sir Ronald Alymer Fisher

(1890-1962) was the pioneer
of DOE. He was responsible
for statistics and data analysis
at the Rothamsted Agricultural
Experiment Station in London,
England. Fisher developed
and was the first to use
ANOVA in the statistical
analysis of experimental data.

Historical Perspective
George E. P. Box (born 1919)

was a student of R A Fisher. He
made several advances to
Fisher’s work in DOE theory
and statistics. The founding
chair of the University of
Wisconsin’s Department of
Statistics, Box was appointed
the R. A. Fisher Professor of
Statistics at UW in 1971.

QE Overview
Objective of this Lecture:
● To explore the basic ideas of two-level factorial design of
experiments (DOE) and the connection of QE to statistical
process control (SPC)
Key Points:
● DOE can help uncover significant variables and

interactions among variables
● SPC can help uncover process shifts
● Quality engineering tools help the investigator to
discover a path for process improvement

Typical QE Applications
 In manufacturing - improve performance of a

manufacturing process
 In process development - improve yields, reduce
variability and cost.
 In design - evaluation and comparison of basic
configurations, materials, and parameters
 The method is called Taguchi Methods.
 The key tool is DOE.

Taguchi Methods

C R Rao
You’d know Rao from his
Cramer-Rao Inequality. Rao is
recognized worldwide as a
pioneer of modern multivariate
theory and as one of the world's
top statisticians, with
distinctions as a mathematician,
researcher, scientist, and
teacher. Taught Taguchi.
Author of 14 books and over
300 papers.

Genichi Taguchi
An engineer who developed
an approach (now called
Taguchi Methods) involving
statistically planned
experiments to reduce
variation in quality. Learned
DOE from Professor Rao.
In 1960’s he applied his
learning in Japan.
In 1980’s he introduced his
ideas to US at AT&T.

What are Taguchi’s
Contributions?
 Quality Engineering Philosophy—Targets

and Loss functions
 Methodology—System, Parameter,
Tolerance design steps
 Experiment Design—use of Orthogonal
arrays
 Analysis—use Signal-to-Noise (S/N ratios)

Conventional DOE focuses only on -++
7
+++
8
Average Response -+- ++-
3 4
FACTOR LOW(-) HIGH (+)
D (Driver) regular oversized B
B (Beverage) beer water
5 --+ 6
+-+
O (Ball) 3-piece balata
1 2 O
standard --- +--
D B O Avg Response D
order
1 - - - 67 77 87
2 + - - 79
3 - + - 61 61 75
4 + + - 75
5 - - + 65
6 + - + 60
- + +
B
7 77
8 + + + 87 65 60
O
QE focuses on Variability of 67 79
Response Copyright Tapan Bagchi D 12
Taguchi’s Key Contributions
 Quality Engineering Philosophy
 Methodology
 Experiment Design
 Analysis

The Taguchi Loss Function
and the typically assumed Loss to the
Customer
Loss
Lo Spec Target Hi Spec

Taguchi’s Quality Philosophy
Taguchi’s view Conventional view
Loss = k(P - T)2

not 0 if within specs
and 1 if outside
LS T US
On Target Production
is more important than
producing within Specs
LS T US

Taguchi focused on
Off-Line Quality Control
Off-Line Quality Control = Improving quality and reducing

total cost in the product or process design stage
Total Cost means cost to society so it includes the cost of

problems in manufacturing and the cost of problems in the
field.

Taguchi’s key contribution is
Robust Design
Definition:
Robust Design—A Design that results in
products or services that can function over
a broad range of usage and environmental
conditions

Taguchi’s Contributions Contd.
 Methodology
 Analysis

Taguchi’s Product Design Approach
has 3 Steps
1. System Design
Choose the sub-systems, mechanisms, form of
the prototype—develop the basic design. This is
similar to conventional engineering design
2. Parameter Design
Optimize the system design so that it improves
quality (robustness) and reduces cost
3. Tolerance Design
Study the tradeoffs that must be made and
determine what tolerances and grades of
materials are acceptable

Parameter Design
(the Robust Design step)
● Optimize the settings of the design parameters to minimize

its sensitivity to noise–ROBUSTNESS.
● By highlighting ―robustness‖ as a key quality requirement,
Taguchi really opened a whole area that previously had been
talked about only by a few very applied people.
● His methodology is heavily dependent on design of
experiments like Fisher’s and Box’s methods, but the
difference he made was that for response he looked at not
only the mean but also the variance of performance

Robust Design—how it is done
Design Parameters (D)

Target
Performance () Product / Process Actual
Performance (P)
Noise Factors (N): Internal & External
 Identify Product/Process Design Parameters that

 Have significant / little influence on Performance
 Minimize performance variation due to Noise factors
 Minimize the processing cost
 Methodology: Design of Experiments (DOE)
 Examples - Chocolate mix, Ina Tile Co., Sony TV
Taguchi’s Experimental Factors
Parameter design step identifies and
optimizes the Design Factors
 Control Factors – Design factors that are to be set at optimal levels
to improve quality and reduce sensitivity to noise
• Size of parts, type of material, Value of resistors, etc
 Noise Factors – Factors that represent the noise that is expected in
production or in actual use of the product
• Dimensional variation
• Operating Temperature
 Adjustment Factor – Affects the mean but not the variance of a
response
• Deposition time in silicon wafer fabrication
 Signal Factors – Set by the user to communicate desires of the user
• Position of the gas pedal

Taguchi’s Contributions Contd.
 Methodology
 Experiment Design  use orthogonal arrays
 Analysis

Several different types of Experimental plans
(“designs”) are available to the design engineer—
Factorial, Fractional, Central Cuboid, etc.
Taguchi used “Orthogonal” Designs
C S
Center Screening
F O
Factorial Orthogonal
FF
Fractional
factorial
Focus: Handle many factors

Output: List of Important Factors, Best Settings, Good design
Full Factorial Array Example: The 23
(8-trial) array
A B
Response
C
Array
Columns  1 2 3 4 5 6 7
1 1 1 1 1 1 1
1 1 1 2 2 2 2
1 2 2 1 1 2 2
1 2 2 2 2 1 1
2 1 2 1 2 1 2
2 1 2 2 1 2 1
2 2 1 1 2 2 1
2 2 1 2 1 1 2
C B -BC A -AC -AB -ABC
Full Factorial Factor Assignments to Experimental Array Columns

Such experiments can find all Main & two- and three-factor Interactions
The L8 Orthogonal Array Example:
Taguchi used these
A B D E
Response
C F G
Array
Columns  1 2 3 4 5 6 7
1 1 1 1 1 1 1
1 1 1 2 2 2 2
1 2 2 1 1 2 2
1 2 2 2 2 1 1
2 1 2 1 2 1 2
2 1 2 2 1 2 1
2 2 1 1 2 2 1
2 2 1 2 1 1 2
C B D A E F G
Orthogonal Array Factor Assignments to Experimental Columns

Such experiments can find all 7 Main effects.
Taguchi’s Orthogonal Experimental Plan—
7 Factors (A, B, C, D, E, F and G) may potentially
influence the production of defective tiles

Calculation of Factor Effects

Main Effects of Process
Factors on %Defects in Tiles

Alternative Design Notations for
Orthogonal Arrays
S td. Fisher's Original Yates Group Theory Taguchi

Order XA1 XB2 XC3 XA1 XB2 XC3 A B C
1 – – – 1 0 0 0 1 1 1
2 + – – a 1 0 0 2 1 1
3 – + – b 0 1 0 1 2 1
4 + + – ab 1 1 0 2 2 1
5 – – + c 0 0 1 1 1 2
6 + – + ac 1 0 1 2 1 2
7 – + + bc 0 1 1 1 2 2
8 + + + abc 1 1 1 2 2 2

Taguchi’s OA-based
Experimental Design Matrix
Notation
Number of Factors
LN 2  
k
Total Number of Runs
Number of Levels per Factor

Linear Graphs for the L8 Array
Linear graphs guide assignment of factors to L8
columns
1 1
7
3 5 3 5
7
2 6 2
4 4
6
 Main effects are assigned to columns at nodes in the graph.

 Interactions are assigned to the columns on the lines.
Some Orthogonal Array
Designs
―Classical‖ ―Taguchi‖
(2-level Factorials)
23-1=L4 L12
23 26-3
24 L18
27-4=L8
25 215-11=L16 L27
27-1
… … …
See Montgomery (1997), Design and Analysis of Experiments, P. 631

Taguchi Orthogonal Array Tables
 2-level (fractional factorial) arrays
 L4(23). L8(27), L16(215). L32(231), L64(263)
 2-level array
 L12(211) (Plackett-Burman Design)
 3-level arrays
 L9(34). L27(313), L81(340)
 4-level arrays
 L16(45). L64(421)
 5-level array
 L25(56)
 Mixed-level arrays
 L18(21x37), L32(21x49), L50(21x511)
 L36(211x312), L36(23x313), L54(21x325)

Comments on Taguchi Arrays
 Taguchi designs are large screening designs
 Assumes most interactions are small and those that aren’t are
known ahead of time.
 Taguchi claims that it is possible to eliminate interactions
either by correctly specifying the response and design
factors or by using a sliding setting approach to those factor
levels.
 Doesn’t guarantee that we get ―highest resolution‖ design.
 Instead of designing the experiment to investigate potential
interactions, Taguchi prefers to use three-level factors to estimate
curvature

Taguchi’s Robust Design
Experiments
 Taguchi advocated using
inner and outer array
designs to take into
account noise factors
(outer) and design factors
(inner)
 Design factors: I1, I2, I3
 Noise factors: E1 & E2
 Objective: Maximize
response while minimizing
its variance

Example: Robust Design OAs of
Starter Motor Parameter Design
Outer array: battery voltage, ambient temperature

Replicates 
Starter torque
Inner array: armature turns, gage of wire, ferric content of alloy

Taguchi’s Contributions
 Methodology
 Analysis  Finding the robust design parameter values

Taguchi’s Analysis uses SN Ratios
 To maximize robustness, Taguchi uses signal-to-noise ratios as

response variables, for example,
y 2 
SNt  -10 log  2 
s 
 However, it is often more informative to analyze mean and
standard deviation separately, rather than combine into a signal-
to-noise ratio
 analyze stddev in the same manner that we have previously
analyzed the mean.
 Taguchi’s analysis techniques are often inefficient…

SN Ratios are Maximized
 To maximize robustness, when Target performance is the
best, Taguchi uses the signal-to-noise ratio
 y2 
SN t  10 log 2 
s 
 When response is to be maximized, Taguchi uses
 1 / y 2 
SN t  -10 log 
 n 
 When response is to be minimized, Taguchi uses
 y 2 
SN t  -10 log 
 n 

Taguchi Analysis of Motor Design Data
80
78
Torque
76
74
72
I1 = -1 I1 = +1 I2 = -1 I2 = +1 I3 = -1 I3 = +1
Inner Array Factors
Robustness is maximized with
SN ratio is maximized.
Standard Dev Torque
25
20
Design (inner array) factor
15
settings that maximize SN ratio
are:
10
I1 = -1 I1 = +1 I2 = -1 I2 = +1 I3 = -1 I3 = +1
I1 (turns) = -1
Inner Array Factor Settings
I2 (gage) = +1
I3 (ferric %) = -1
S/N Ratio of Response
0.0015
0.001
Note: This system is not additive!
 Results are approximately OK.
0.0005
0
I1 = -1 I1 = +1 I2 = -1 I2 = +1 I3 = -1 I3 = +1
Inner Array Factor Settings

Epilogue
 Designers should embrace Taguchi’s philosophy of quality

engineering. It makes very good sense.
 Note, however, that a key weakness of Taguchi method is its
assumption of a ―main factor only‖ (or ―additive‖ model)…
Taguchi ignores interactions
 Therefore, rather than use inner  outer arrays, we may use
more efficient and exact methods that are no more difficult to
learn and apply to carry Taguchi’s robust design philosophy
into practice…
 You may use any of the various experimental and optimization
techniques available in the literature such as multiple
regression/RSM to develop robust designs.
 An example of such extension is shown in the next slides.

Multiobjective Robust
Design
by Metaheuristic Methods
Tapan P Bagchi
and
Madhu Ranjan Kumar (1993)

The Empirical Framework for
progressing Knowledge
Psychology
Engineering
Weather
Medicine Electronics
Economy Chemistry

Robust Design search by GA


NPTEL Taguchi Methods

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

NPTEL Taguchi Methods

Hochgeladen von

Copyright:

Verfügbare Formate

Quality Engineering and

Copyright Tapan Bagchi 1

Focus is on reducing variability of response

Copyright Tapan Bagchi 3

Sir Ronald Alymer Fisher

Copyright Tapan Bagchi 4

George E. P. Box (born 1919)

Copyright Tapan Bagchi 5

● DOE can help uncover significant variables and

Copyright Tapan Bagchi 6

 In manufacturing - improve performance of a

Copyright Tapan Bagchi 7

Copyright Tapan Bagchi 8

Copyright Tapan Bagchi 9

Copyright Tapan Bagchi 10

 Quality Engineering Philosophy—Targets

Copyright Tapan Bagchi 11

 Quality Engineering Philosophy

Copyright Tapan Bagchi 13

Lo Spec Target Hi Spec

Copyright Tapan Bagchi 14

Taguchi’s view Conventional view

Loss = k(P - T)2

Copyright Tapan Bagchi 15

Off-Line Quality Control = Improving quality and reducing

Total Cost means cost to society so it includes the cost of

Copyright Tapan Bagchi 16

Copyright Tapan Bagchi 17

 Quality Engineering Philosophy

Copyright Tapan Bagchi 18

Copyright Tapan Bagchi 19

● Optimize the settings of the design parameters to minimize

Copyright Tapan Bagchi 20

Design Parameters (D)

Noise Factors (N): Internal & External

 Identify Product/Process Design Parameters that

Copyright Tapan Bagchi 22

 Quality Engineering Philosophy

 Experiment Design  use orthogonal arrays

Copyright Tapan Bagchi 23

Focus: Handle many factors

Full Factorial Factor Assignments to Experimental Array Columns

Orthogonal Array Factor Assignments to Experimental Columns

Copyright Tapan Bagchi 27

Copyright Tapan Bagchi 28

Copyright Tapan Bagchi 29

S td. Fisher's Original Yates Group Theory Taguchi

Copyright Tapan Bagchi 30

Total Number of Runs

Number of Levels per Factor

Copyright Tapan Bagchi 31

 Main effects are assigned to columns at nodes in the graph.

See Montgomery (1997), Design and Analysis of Experiments, P. 631

 L36(211x312), L36(23x313), L54(21x325)

Copyright Tapan Bagchi 34

Copyright Tapan Bagchi 35

Copyright Tapan Bagchi 36

Outer array: battery voltage, ambient temperature

Inner array: armature turns, gage of wire, ferric content of alloy

 Quality Engineering Philosophy

 Analysis  Finding the robust design parameter values

Copyright Tapan Bagchi 38

 To maximize robustness, Taguchi uses signal-to-noise ratios as