Title: What is Robust Design or Taguchi
1What is Robust Design or Taguchis method?
- An experimental method to achieve product and
process quality through designing in an
insensitivity to noise based on statistical
principles.
2History of the method
- Dr. Taguchi in Japan 1949-NTT
- develops Quality Engineering
- 4 time winner of Demming Award
- Ford Supplier Institute, early 1980s
- American Supplier Institute, ASI
- Engineering Hall of Fame
- Statistics Community
- DOE
- S/N Ratio
3 Who uses Taguchis Methods
- Lucent
- Ford
- Kodak
- Xerox
- Whirlpool
- JPL
- ITT
- Toyota
- TRW
- Chrysler
- GTE
- John Deere
- Honeywell
- Black Decker
4Documented Results from Use
- 96 improvement of NiCAD battery on satellites
(JPL/ NASA) - 10 size reduction, 80 development time
reduction and 20 cost reduction in design of a
choke for a microwave oven (L.G. Electronics) - 50,000 annual cost savings in design of heat
staking process (Ann Arbor Assembly Corp) - 60 reduction in mean response time for computer
system (Lucent)
- 900,000 annual savings in the production of
sheet-molded compound parts (Chrysler) - 1.2M annual savings due to reduction in vacuum
line connector failures (Flex Technologies) - 66 reduction in variability in arrival time and
paper orientation (Xerox) - 90 reduction in encapsulation variation (LSI
Corp)
5Insensitivity to Noise
- Noise Factors which the engineer can not or
chooses not to control - Unit-to-unit
- Manufacturing variations
- Aging
- Corrosion
- UV degradation
- wear
- Environmental
- human interface
- temperature
- humidity
6How Noise Affects a System
7Step 1 Define the Project Scope 1/2
- A gyrocopter design is to be published in a
Sunday Comics section as a do-it-yourself project
for 6-12 year old kids - The customers (kids) want a product they can
easily build and have a long flight time.
WW
--- WL --- BL ----
--- --- 1/4
8Step 1 Define the Project Scope 2/2
- This is a difficult problem from an engineering
standpoint because - hard to get intuitive feel for effect of control
variables - cant control materials, manufacturing or assembly
- noise factors are numerous and have strong effect
on flight.
9Step 2 Identify Ideal Function
- Ideally want the most flight time (the quality
characteristic or useful energy) for any input
height (signal or input energy) - Minimize Noise Effect
- Maximize Slope
Time of Flight
Drop Height
10Step 3 Develop Noise Strategy 1/2
- Goal is to excite worst possible noise conditions
- Noise factors
- unit-to-unit
- aging
- environment
11Step 3 Develop Noise Strategy 2/2
- Noise factors
- unit-to-unit
- Construction accuracy
- Paper weight and type
- angle of wings
- aging
- damage from handling
- environment
- angle of release
- humidity content of air
- wind
many, many others
12Step 4 Establish Control Factors and Levels 1/4
- Want them independent to minimize interactions
- Dimensionless variable methods help
- Design of experiments help
- Confirm effect of interactions in Step 7
- Want to cover design space
- may have to guess initially and perform more than
one set of experiments. Method will help
determine where to go next.
13Step 4 Establish Control Factors and Levels 2/4
- Methods to explore the design space
- shot-gun
- one-factor-at-a-time
- full factorial
- orthogonal array (a type of fractional factorial)
14Step 4 Establish Control Factors and Levels 3/4
15Step 4 Establish Control Factors and Levels 4/4
16Step 5 Conduct Experiment and Collect Data
17Data for Runs 5 and 15
18Step 6 Conduct Data Analysis 1/7
- Calculate signal-to-noise-ratio (S/N) and Mean
- Complete and interpret response tables
- Perform two step optimization
- Reduce Variability (minimize the S/N ratio)
- Adjust the mean
- Make predictions about most robust configuration
19Step 6 Conduct Data Analysis 2/7
- Calculate signal to noise ratio, S/N, a metric in
decibels
variability S/N gain
reduction 3 27 6
50 12 75
Useful output Harmful output
S/N
Effect of Mean
Variability around mean
y2
10 log
Note This is one of many forms of S/N ratios.
s2
20Step 6 Conduct Data Analysis 3/7
21Step 6 Conduct Data Analysis 4/7Response Table
22Step 6 Conduct Data Analysis 5/7Response plot
23Step 6 Conduct Data Analysis 6/7Two Step
Optimization
- Reduce Variability (minimize the S/N ratio)
- look for control factor effects on S/N
- Dont worry about mean
- Adjust the mean
- To get desired response
- Use adjusting factors, those control factors
which have minimal effect on S/N
24Step 6 Conduct Data Analysis 7/7
- For gyrocopter
- wing width .75in
- wing length 2.00/0.75 2.67 in
- body length 2.00 x 2.67 5.33 in
- size 50
- no body folds
- no gussets
Predicted Performance S/N 9.44 dB Slope .31
sec/ft
25Step 7 Conduct Conformation Run
- To check validity of results
- To check for unforeseen interaction effects
between control factors - To check for unaccounted for noise factors
- To check for experimental error
Predicted Confirmed S/N 9.44 dB
9.86 Slope .31sec/ft .32 sec/ft
26How Taguchis Method Differs from an Ad-hoc
Design Process
- Organized Design Space Search
- Clear Critical Parameter Identification
- Focus on Parameter Variation (Noise)
- Clear Stopping Criteria
- Robustness centered not Failure Centered
- Reusable Method
- Concurrently Addresses Manufacturing Variation
- Concurrent Design-Test Not Design-Test-Fix
- Minimize Development Time (Stops Fire Fighting)
- Corporate Memory Through Documentation
- Encourages Technology Development Through System
Understanding
27How Taguchis Method Differs from Traditional
Design of Experiments
- Focused on reducing the impact of variability
rather than reducing variability - Focused on noise effects rather than control
factor effects - Clearly focused cost function - maximizing the
useful energy
- Tries to reduce interaction between control
factors rather than study them Requires little
skill in statistics - Usually lower cost
28How Taguchis Method Differs from Shainins Method
- Focused on both Product and Process Design rather
than Primarily on Process - Oriented to developing a robust system not
finding a problem (Red X). Taguchi tells what
parameter values to set to make system
insensitive to parameter Shainin identifies as
needing control.
- Widely Used Internationally
- Fire prevention rather than fire fighting
- Accessible
- Many Case Studies Available
29Plan for Application at Tektronix
- Select a parameter design problem
- Design the experiment
- Perform the experiment
- Reduce data
- Report results to Company
- Assuming success
- design more experiments
- train more engineers
- Plan for student-run experiments