WhyWhen is Taguchi Method Appropriate Tip 1 Taguchis

Why/When is Taguchi Method Appropriate? Tip #1 Taguchi’s Signal-to-Noise Ratios are in Log form Friday, 6 April 2001

Earlier Tip Taguchi’s Signal-to-Noise Ratios are in Log form Friday, 6 April 2001

Friday, 6 -April-2001 Taguchi’s Signal-to-Noise Ratios are in Log form • very important reason – Taking “logarithm” improves ‘additivity’. Why? • • Most functions in ‘nature’ follow the power-laws Y = A exp (-Ea/k. T) (EXPONENTIAL) Y = B Ax. . ( a – RAISED – POWER X) Y = B Xa. . ( X – RAISED – POWER a) – On taking logarithm these become ‘additive’ • For example Log 10 ( B Xa. exp(-Ea/k. T) ) = Log 10(B) + a. Log 10 (X) + (-Ea/k. T). Log 10 (e) • These are ‘additive’ for variables (Log 10 X) and (1/T)

Friday, 6 -April-2001 Taguchi’s Signal-to-Noise Ratios are in Log form • Let us say we are trying reduce ‘defects’ in a process using L 9 array • The prediction will be, say, – 7 defects +- a few defects – The prediction error may be high due to errors in individual 9 experiments. Say it is +- 9 – 7 defects +-9 -2 to 17 defects ? ? ? ? (-2 is problem) • Taguchi’s SN-Ratio for smaller-the-better – = -10 Log 10 ( 1/n Yi 2 ) – Its range is – infinity to +infinity (- /10) 2 – Extreme values are = +infinity defects = 0 [average Y = 10 ] – = - infinity defects = infinity – Always predicts defects between allowed range of 0 to infinity

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