How to use DOE in practical world
Six Sigma – iSixSigma › Forums › Old Forums › General › How to use DOE in practical world
- This topic has 4 replies, 5 voices, and was last updated 15 years, 2 months ago by
Deep.
-
AuthorPosts
-
April 29, 2007 at 12:24 pm #46848
Hi,
I just finished my DOE course and working in a an automobile manufacturing environment where we manufacture shock absorbers for tier-1 vendor.
Could anyone guide me how could I use DOE in this environment. I am very enthusiastic!!
Nina0April 29, 2007 at 12:56 pm #155428
Robert ButlerParticipant@rbutlerInclude @rbutler in your post and this person will
be notified via email.You will need to give more details of your situation before anyone can offer much in the way suggestions.
For example, if you can halt the line, run a series of experiments, and then restart production your approach will be very different from the case where you have to schedule your experiments as part of the daily adjustments to the line (for an example of how to approach the latter see the post cited below).
Your approach will also depend on what you want to vary vs. what you are allowed to vary. I’m sure there will be issues surrounding raw material/supplier changes and how you are allowed to approach these will also impact your final approach to a design. (In all of the above we are, of course, assuming you did everything you should have done before considering a design – proper variable identification, examination of existing production data, etc.)
https://www.isixsigma.com/forum/showmessage.asp?messageID=1037710April 29, 2007 at 5:09 pm #155432Hi Nina,
Well you can use DOE to optimize the smooth functioning of shock absorbers. For ex.. if Helical spring is a critical component of an shock absorber. You can optimize the dimensions of the Helical spring based on response. For example if your out put is Displacement and the measurable parameters in Helical spring is 1) dia of spring 2) length of spring 3) number of winding etc.. you can optimize these 3 dimensions for an expected response displacement “Y”. Hence we try to represent in mathematical formulae it can be represented as Y(Displacement)=F(dia, length,num) hence for a predefined weight and each variable(dia,length,num) can have 2 extreme value ex. dia =( 2 mm to 5 mm) so the optimal will be either of u or any value b/w these two limits. We arrive to transfer function or prediction equation based on existing analysis, data & fact and obtain TF or PF. Then we will work backward to get the optimal values..Hope i have explained if you need further info.. do feel free to mail to [email protected].0April 29, 2007 at 5:53 pm #155433
AllthingsidiotoParticipant@AllthingsidiotoInclude @Allthingsidioto in your post and this person will
be notified via email.Excellent Elaboration.One important thing most people ignore in DOE,first to brainstorm a session with the SME,to decide the requirements:smaller is better,bigger is better or Nominal is best as quality charecterstics.good luck
0April 29, 2007 at 6:05 pm #155435Robert Butler:Can you check your email. I send you a question few days back. sorry to tell this over here, just want to make sure you got my mailthank youDeep
0 -
AuthorPosts
The forum ‘General’ is closed to new topics and replies.