Title: Why/When is Taguchi Method Appropriate?
1Why/When is Taguchi Method Appropriate?
Friday, 8th June 2001
2Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
(next 7 slides)
Tip 8
Friday, 8th June 2001
3Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
- A process has several quality characteristics
- Desirable Qualities
- e.g. Larger-the-Better (strength, throughput
etc) - or Nominal-the-Best (specified dimensions,
uniformity) - Undesirable properties
- e.g. Smaller-the-better (defects, resources -
material/time, cost) - Objective of Taguchi Method is to determine the
best settings of Control Factors such that - Desirable qualities are enhanced
- .and.
- Undesirable properties are minimized/eliminated
- Process becomes ROBUST i.e. insensitive to NoIsE
4Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
- Taguchi Method is most effective when there is at
least one quality characteristic that is
sensitive to variations or NoIsE - Desirable Qualities
- e.g. Nominal-the-Best type are sensitive to
NoIsE - S/N Ratio ? 10 Log10 ( mean2 /
Variance ) - Undesirable properties
- e.g. Smaller-the-better type are also
sensitive to NoIsE - S/N Ratio ? 10 Log10 ( 1/n ? Yi2 )
- ? 10 Log10
(Variance) . . . if ideal value zero
5Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
- Let us use the quality Loss function for
quality characteristics - Desirable Qualities
- e.g. Nominal-the-Best
- Quality Loss, Qa (Variance / mean2 )
( ?2 / ?2 ) - e.g. Larger-the-Better
- Quality Loss, Qa 1/n ? (1/Yi2 )
- Undesirable properties
- e.g. Smaller-the-better
- Quality Loss, Qa 1/n ? Yi2
6Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
- We can minimize the Total quality Loss
function for TWO or more quality characteristics - Total Quality Loss, Qa,total Qa,1 Qa,2 . .
. - To obtain SN Ratio, take the Log-form of the
Total quality Loss function - SN Ratio -10 Log Qa,total
- -10 Log Qa,1 Qa,2 Qa,3
. . .
7Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
- We can add the quality Loss functions for TWO
quality characteristics, as given below - Desirable Qualities
- e.g. TWO quality characteristics of
Nominal-the-Best type - Combined Quality Loss, Qa ( ?12 / ?12 ) (
?22 / ?22 ) - Desirable Qualities
- e.g. TWO quality characteristics of
Larger-the-Better type - Combined Quality Loss, Qa 1/m ?(1/Y1j ) 2
1/n ?(1/Y2j ) 2 - Undesirable properties
- e.g. TWO quality characteristics of
smaller-the-better type - Combined Quality Loss, Qa 1/m ?(Y1j ) 2
1/n ?(Y2j ) 2
8Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
- Example Computer Benchmarking
- Time required for obtaining the time and date ?
T1 - Time required for opening and closing a file ?
T2 - Desired quality characteristics
- time T1 and T2 both ? smaller-the-better
- Total Quality Loss, Qa,total 1/m ?(T1j ) 2
1/n ?(T2j ) 2 - To obtain SN Ratio, take the Log-form of the
Total quality Loss - SN Ratio -10 Log 1/m ?(T1j ) 2 1/n ?(T2j
) 2
9Taguchi Method Finds best settings to optimize
TWO quality characteristics Simultaneously
- Example Telecom Connectivity
- Calls connected per hour on XMas day ? C1
- Calls connected per hour on Monday morning ?
C2 - Desired quality characteristics
- calls C1 and C2 both ? Larger-the-Better
- Total Quality Loss, Qa,total 1/m ?(1/C1j ) 2
1/n ?(1/C2j ) 2 - To obtain SN Ratio, take the Log-form of the
Total quality Loss - SN Ratio -10 Log 1/m ?(1/C1j ) 2 1/n
?(1/C2j ) 2
Friday, 8th June 2001
10Earlier Tips Links below
- 7. Taguchi Method
- When to select a Larger OA to perform
Factorial Experiments - Taguchi Method Using Orthogonal Arrays for
Generating Balanced Combinations of NoIsE Factors - 5. Taguchi Method Signal-to-Noise Ratio for
Quality Characteristics approaching IDEAL value
- Friday, 1st June 2001
- Friday, 25th May 2001
- Friday, 18th May 2001
Tips 4,3,2,1 ?
11Earlier Tips Links below
- Friday, 11th May 2001
- Friday, 4th May 2001
- Friday, 27th April 2001
- Friday, 6th April 2001
- 4. Taguchi Method improves " quality at all
the life stages - at the design stage itself
-
- 3. Taguchi Method Appropriate for Concurrent
Engineering - 2. Taguchi Method can study Interaction
- between Noise Factors and Control
Factors - 1. Taguchis Signal-to-Noise Ratios are in Log
form
12More Tips Links below
- Friday, 3rd Aug 2001
- Friday, 27th July 2001
- Friday, 20th July 2001
- Friday, 13th July 2001
- Taguchi Method
- 1st Priority Variance Reduction 2nd Priority
Factor Effects - 15. inner L9 array with outer L4 and L9
NoIsE arrays - Taguchi Method
- inner L18 array with outer L4 and L9 NoIsE
arrays - Taguchi Method Why/When is Taguchi Method not
Appropriate?
Tips 12, 11, 10 ?
13More Tips Links below
- Friday, 6th July 2001
- Friday, 29th June 2001
- Friday, 22nd June 2001
- Taguchi Method
- inner L8 array with outer L4 and L9 NoIsE
arrays - Taguchi Method
- Useful at ALL Life-stages of a Process or
Product - Taguchi Method
- Performs Process centering or fine tuning
Tips 9, 8, 7 ?
14More Tips Links below
- Taguchi Method Identifies the right NoIsE
factor(s) for Tolerance Design - Taguchi Method
- Finds best settings to optimize TWO quality
characteristics Simultaneously - 7. Taguchi Method
- When to select a Larger OA to perform
Factorial Experiments
- Friday, 15th June 2001
- Friday, 8th June 2001
- Friday, 1st June 2001
Tips 6, 5, 4 ?
15More Tips Links below
- Friday, 25th May 2001
- Friday, 18th May 2001
- Friday, 11th May 2001
- Taguchi Method Using Orthogonal Arrays for
Generating Balanced Combinations of NoIsE Factors - Taguchi Method Signal-to-Noise Ratio for Quality
Characteristics approaching IDEAL value - 4. Taguchi Method improves " quality at
all the life stages - at the design stage itself
Tips 3, 2, 1 ?
16More Tips Links below
- Friday, 4th May 2001
- Friday, 27th April 2001
- Friday, 6th April 2001
- 3. Taguchi Method Appropriate for Concurrent
Engineering - 2. Taguchi Method can study Interaction
- between Noise Factors and Control
Factors - 1. Taguchis Signal-to-Noise Ratios are in Log
form
17end