Need some help with a question that has me stumped
Six Sigma – iSixSigma › Forums › Old Forums › General › Need some help with a question that has me stumped
 This topic has 5 replies, 5 voices, and was last updated 12 years, 8 months ago by Remi.

AuthorPosts

January 23, 2009 at 7:58 pm #51726
I have this question that has me completely stumped. I am in no way looking for an answer just some guidance on how to approach the question. I know it has to be done using minitab.Here is what I have:
One variable or factor we will call X1 is varied in a one factor experiment and results in an X2. I am given an equation X1 = a/X2 + b noting that X1 and X2 follow the relationship defined by this equation. I am being asked ot compute a and b.
If I perform a simple linear regresssion analysis I get X2 = # – (#*X1). If I perform a linear fitting line plot I get the same format.
I am lost on how to try to find a & b in Minitab.
Any guidance would be appreciated
0January 23, 2009 at 8:48 pm #180152I assume X1 is the “y” and X2 is the “x”
In that case, let y = 1/X2,
then y = X1/a – b/a.
Use regression in Minitab to solve y = m*X1 + c where m = 1/a and c = b/a.
Forgive me if I made a mistake in algebra. I think you can figure out what I’m showing.
BC0January 23, 2009 at 9:45 pm #180155I am afraid the sustitution I made may have made this unclear…
Let me use different substitutions to try to make it clearer..I am varying p = pressure to get a volume = v. p and v follow this relationship of p = a/v +b. I am gven a set of runs and I am being asked to compute a and b using mintab.0January 23, 2009 at 9:58 pm #180156
Boyles LawParticipant@BoylesLaw Include @BoylesLaw in your post and this person will
be notified via email.P*V=C
Pressure * Volume=Constant
Boyles Law This may help you get there0January 24, 2009 at 11:30 am #180162
KRISHNANMember@saravanan Include @saravanan in your post and this person will
be notified via email.i need more informations about six sigma concept as your concern
0January 26, 2009 at 9:29 am #180192Hai William2;
General: With Regression you solve Y= c0 + c1*X where X is independent and Y is dependent (either Regression > Regression or Regression > Fitted Line Plot).
In your case p is independent and v is dependent and the relation you try to fnd will be looking like: v = a/(pb) or v = a*1/p – a/b.
Make a column of 1/p (carefull with the 0’s). And analyse Y=v as function of X=1/p.
If your relation was not p=a/v+b but a*v+b you would have a slightly different formula and X=p.
Hope this helps.
Remi0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.