WhyWhen is Taguchi Method Appropriate Tip 5 Taguchi

Why/When is Taguchi Method Appropriate? Tip #5 Taguchi Method Appropriate Signal-to-Noise Ratio for Quality Characteristics approaching IDEAL value Friday, 18 th May 2001

Tip #5 Taguchi Method Appropriate Signal-to-Noise Ratio for Quality Characteristics approaching IDEAL value (next 7 slides) Friday, 18 th May 2001

Taguchi’s SN Ratio Quality Characteristics approaching IDEAL value • Taguchi’s SN-Ratio for smaller-the-better – quality characteristics is usually an undesired output, for example • Defects like pin holes, particulates in deposition processes • Unwanted by-product or side effect • Taguchi’s SN-Ratio for LARGER-THE-BETTER – quality characteristics is usually a desired output, for example • Bond strength • Critical Current • Taguchi’s SN-Ratio for NOMINAL-the-best – quality characteristics is usually a nominal output, for example • most parts in mechanical fittings have nominal dimensions • Ratios of chemicals or mixtures are nominally the best type. • Thickness should be uniform in deposition /growth /plating /etching. .

Taguchi’s SN Ratio Quality Characteristics approaching IDEAL value • But what about quality characteristics that approach an ideal value? – Examples are • Efficiency : all efficiencies approach the ideal value of 100% • Weld strength : approaches the ideal strength of the material • Critical temperature or Critical current density for High Temperature superconductors (YBCO) : These approach ideal values, say 92 K and 108 A/cm 2 • Which SN-Ratio is most suitable among the following ? – smaller-the-better – LARGER-THE-BETTER – NOMINAL-the-BEST

Taguchi’s SN Ratio Quality Characteristics approaching IDEAL value • Taguchi’s SN-Ratio for smaller-the-better ( quality characteristics is usually an undesired output, say Defects ) – S/N Ratio = – 10 Log 10 ( 1/n Y 2 i ) • Taguchi’s SN-Ratio for LARGER-THE-BETTER ( quality characteristics is usually a desired output, say Current ) – S/N Ratio = – 10 Log 10 ( 1/n 1/Y 2 i ) • Taguchi’s SN-Ratio for NOMINAL-the-best ( quality characteristics is usually a nominal output, say Diameter ) – S/N Ratio = 10 Log 10 ( 2 / 2 )

Taguchi’s SN Ratio Quality Characteristics approaching IDEAL value • Generally, we would state that this value should be “LARGER-THE-BETTER” • However, there exists an IDEAL value!!! • QUESTION still IS “Which SN-Ratio is most suitable among the following ? ” – LARGER-THE-BETTER – smaller-the-better – Nominal-the-best

Taguchi’s SN Ratio Quality Characteristics approaching IDEAL value • Reword the SN-Ratio for smaller-the-better to indicate ‘the approach to IDEAL value’ by saying “Smaller-the-difference-from-IDEAL-the-better” • Taguchi’s SN-Ratio for smaller-the-better is now modified ( for quality characteristics approaching IDEAL output ) – S/N Ratio = – 10 Log 10 [ 1/n (YIDEAL- Y ) 2 ] i

Taguchi’s SN Ratio Quality Characteristics approaching IDEAL value • If our nominal values are far away from IDEAL value – SN-Ratio for LARGER-THE-BETTER will give good additivity (prediction and results will match well) • If our nominal values are already close to IDEAL value – SN-Ratio for smaller-the-better (the modified form) will give good additivity (prediction and results will match well) = – 10 Log 10 [ 1/n (YIDEAL- Y ) 2 ] i

Taguchi’s SN Ratio Quality Characteristics approaching IDEAL value • SN-Ratio for LARGER-THE-BETTER tends to predict optimistic (LARGER) values – Sometimes the value exceeds maximum possible value !? • SN-Ratio for smaller-the-better (the modified form) always predicts values pessimistically (LESS than ideal value) ü TRY BOTH, Select one with less ANOVA error Friday, 18 th May 2001

More Tips 16. Links below Taguchi Method e 1 st Priority : Va ia ce R duction 2 nd Priority : Factor Effects Friday, 3 rd Aug 2001 15. “inner” L 9 array with “outer” L 4 and L 9 No. Is. E arrays Friday, 27 th July 2001 14. Taguchi Method r n “inner” L 18 array with “outer” L 4 and L 9 No. Is. E arrays 13. Friday, 20 th July 2001 Taguchi Method Why/When is Taguchi Method not Appropriate? Friday, 13 th July 2001 Tips 12, 11, 10

More Tips 12. Taguchi Method Links below Friday, 6 th July 2001 “inner” L 8 array with “outer” L 4 and L 9 No. Is. E arrays 11. Taguchi Method Friday, 29 th June 2001 Useful at ALL Life-stages of a Process or Product 10. Taguchi Method Friday, 22 nd June 2001 Performs Process “centering” or “fine tuning” Tips 9, 8, 7

More Tips 9. Taguchi Method Links below Friday, 15 th June 2001 Identifies the “right” No. Is. E factor(s) for Tolerance Design 8. Taguchi Method Friday, 8 th June 2001 Finds best settings to optimize TWO quality characteristics Simultaneously 7. Taguchi Method Friday, 1 st June 2001 When to select a ‘Larger’ OA to perform “Factorial Experiments” Tips 6, 5, 4

More Tips 6. Taguchi Method Links below Friday, 25 th May 2001 Using Orthogonal Arrays for Generating Balanced Combinations of No. Is. E Factors 5. Taguchi Method Friday, 18 th May 2001 Signal-to-Noise Ratio for Quality Characteristics approaching IDEAL value 4. Taguchi Method Friday, 11 th May 2001 improves " quality “ at all the life stages at the design stage itself Tips 3, 2, 1

More Tips Links below 3. Taguchi Method Appropriate for Concurrent Engineering Friday, 4 th May 2001 2. Taguchi Method can study Interaction Friday, 27 th April 2001 between Noise Factors and Control Factors 1. Taguchi’s Signal-to-Noise Ratios are in Log form Friday, 6 th April 2001

end