Comments on: Taguchi Loss Function http://lssacademy.com/2007/04/15/taguchi-loss-function/ Lean Manufacturing, Six Sigma, Lean Six Sigma, and Kaizen Sat, 20 Mar 2010 07:33:05 -0600 http://wordpress.org/?v=2.9.2 hourly 1 By: walt Eschmann http://lssacademy.com/2007/04/15/taguchi-loss-function/comment-page-1/#comment-2251 walt Eschmann Sun, 06 Jul 2008 19:09:01 +0000 http://lssacademy.com/?p=200#comment-2251 Need information on Tagushi loss function for one sided models. The formula I have is L(X) =kx^2 for less is better and l(x)=k*1/x^2. Is x the value from the target, or is x the value of x-t? For example, x=.48 and t+ .50. Should K be multiplied by x and then squared? This is for applications such as chemicals. thanks in advance for any help. Need information on Tagushi loss function for one sided models. The formula I have is L(X) =kx^2 for less is better and l(x)=k*1/x^2. Is x the value from the target, or is x the value of x-t? For example, x=.48 and t+ .50. Should K be multiplied by x and then squared? This is for applications such as chemicals.

thanks in advance for any help.

]]> By: Taguchi Index – Cpm | Lean Six Sigma Academy http://lssacademy.com/2007/04/15/taguchi-loss-function/comment-page-1/#comment-1692 Taguchi Index – Cpm | Lean Six Sigma Academy Mon, 24 Mar 2008 03:56:14 +0000 http://lssacademy.com/?p=200#comment-1692 [...] night we discussed the Taguchi Loss Function and how Taguchi methods are more concerned with hitting the target compared to more traditional [...] [...] night we discussed the Taguchi Loss Function and how Taguchi methods are more concerned with hitting the target compared to more traditional [...]

]]> By: Ron Pereira http://lssacademy.com/2007/04/15/taguchi-loss-function/comment-page-1/#comment-368 Ron Pereira Mon, 16 Apr 2007 18:52:00 +0000 http://lssacademy.com/?p=200#comment-368 Great question Maciej. There main beef is with his DOE's. A Taguchi DOE is like an enhanced fractional factorial design. Since a Taguchi DOE is not completely orthogonal (balanced) some say they can be misleading. I will write more about Taguchi DOE's soon as there is some truth to their concern if you choose the wrong Taguchi DOE. By my favorite, the L18, is bullet proof and probably my all time favorite DOE design out there. Great question Maciej. There main beef is with his DOE’s. A Taguchi DOE is like an enhanced fractional factorial design. Since a Taguchi DOE is not completely orthogonal (balanced) some say they can be misleading. I will write more about Taguchi DOE’s soon as there is some truth to their concern if you choose the wrong Taguchi DOE. By my favorite, the L18, is bullet proof and probably my all time favorite DOE design out there.

]]> By: Maciej http://lssacademy.com/2007/04/15/taguchi-loss-function/comment-page-1/#comment-369 Maciej Mon, 16 Apr 2007 18:40:00 +0000 http://lssacademy.com/?p=200#comment-369 Interesting.. I hadn't heard of this guy before. What is so statistician unfriendly about this approach though? Interesting.. I hadn’t heard of this guy before. What is so statistician unfriendly about this approach though?

]]> By: Ron Pereira http://lssacademy.com/2007/04/15/taguchi-loss-function/comment-page-1/#comment-370 Ron Pereira Mon, 16 Apr 2007 16:23:00 +0000 http://lssacademy.com/?p=200#comment-370 Thanks Jon. Right back at ya brutha! Thanks Jon. Right back at ya brutha!

]]> By: Jon Miller http://lssacademy.com/2007/04/15/taguchi-loss-function/comment-page-1/#comment-371 Jon Miller Mon, 16 Apr 2007 15:42:00 +0000 http://lssacademy.com/?p=200#comment-371 Rock on. Rock on.

]]> By: Anonymous http://lssacademy.com/2007/04/15/taguchi-loss-function/comment-page-1/#comment-372 Anonymous Mon, 16 Apr 2007 14:01:00 +0000 http://lssacademy.com/?p=200#comment-372 good stuff. look forward to future taguchi discussion! good stuff. look forward to future taguchi discussion!

]]>