Optimization techniques
Six Sigma – iSixSigma › Forums › Old Forums › General › Optimization techniques
 This topic has 16 replies, 12 voices, and was last updated 18 years, 1 month ago by MBB Help.

AuthorPosts

July 25, 2002 at 8:34 am #29956
Nitin AgrawalParticipant@NitinAgrawal Include @NitinAgrawal in your post and this person will
be notified via email.I want to know of different optimization techniques I can use to find out the best specs / value for my process parameters.
0July 25, 2002 at 8:46 am #77574
vidya KulkarniMember@vidyaKulkarni Include @vidyaKulkarni in your post and this person will
be notified via email.hi Nitin,
I think DOE can be the best tool for optimisation of a process or spec value.
vidya0July 25, 2002 at 12:54 pm #77588
Ø6 Sigma BB CoordinatorParticipant@Ø6SigmaBBCoordinator Include @Ø6SigmaBBCoordinator in your post and this person will
be notified via email.When we know Y = f(Xs), we can optimize our process easily.
These are just some of the name of the techniques to find Y = f (Xs) and how to use it after discovery the relationship.
0. Variable Search
1. Factorial DOE
2. RSM
3. Scatter plot
4. Robust design
5. Parameter design
6. Tolerance design
7. Statistical tolerance analysis
8. Simulation
Hope this helps.
Six Sigma Black Belt Coordinator0July 25, 2002 at 1:47 pm #77600
Mike CarnellParticipant@MikeCarnell Include @MikeCarnell in your post and this person will
be notified via email.Nitin,
DOE’s in and of themselves will point you in the right direction but your best treatment combination is not neccessarily an optimized solution. There are two techniques that can be used as a follow on to a DOE that will provide a systematic methodology to get to optimization. They are Evolutionary Operation (EVOP) and Steepest Ascent. They are similar in concept.
Good luck.0July 26, 2002 at 3:12 am #77623
Nitin AgrawalParticipant@NitinAgrawal Include @NitinAgrawal in your post and this person will
be notified via email.Dear Ms./Mr. BB Coordinator,
Thank you for your reply. Can you tell me more on the following techniques
1) RSM (Response Surface Methodology)
2) Statistical Tolerance Design
How can I use these techniques to find out the optimum condition for my process?0July 26, 2002 at 3:15 am #77624
Nitin AgrawalParticipant@NitinAgrawal Include @NitinAgrawal in your post and this person will
be notified via email.Dear Mike,
Thanks for introducing me to the two techniques of EVOP and Steepest Ascent. Can you send me some more details on the two so that I can develop some understanding of them on what is the basic funda of the two and how can they be used.0July 26, 2002 at 1:28 pm #77630Don’t forget to consider the effects of process variation of your factors in the optimiztion effort. The sensitivity of one factor vs another may drive you to different “ideal” zones. EVOP and RSM only give extremely short term glimpses into variation. You definitely will want to do some Monte Carlo type modleing . Remember group 6S is not just about the mean – vartiation is the key.
0July 26, 2002 at 2:52 pm #77634Nitin
Robust Design or Taguchi Method is the best technique for optimization of process parameters.
Manee0July 26, 2002 at 3:52 pm #77637Nitin,
Evolutionary Operation (EVOP) was a strategy of experimentation first espoused by Box and Draper. The basic idea was to run incremental experimentation on fullscale manufacturing processes without disrupting the overall quality of the product and/or to negate bringing the process offline to run text book applications of factorial experimentation. The results of EVOP will be a continous improvement of the process, but will not result in breakthroughtype performance improvements. The overall strategy is heavily dependent on process knowledge and stability of the process that is being targeted. The basic elements include:Series of sequential experiments
Simple factorial designs at 2**2 or 2**3
Small changes to factors
Large # of runs per factor combination
Simple graphical analysis
Steepest Ascent is an element within Response Surface Methodology (RSM). DOE like other tools builds on a foundation of knowledge. The maturity of DOE to a process goes from screening, to characterizing, and culminates with optimization. RSM is the technique used to optimize a process. At its basic level, RSM is looking to model curved relationships to find the optimum response through the quadratic model. This is due to the fact that most functions can be approximated quite effectively by the 2nd order Taylor series expansion and further complexity is not typically needed. Additionally, all of the x’s, for RSM, need to be continuos. So what about this method of steepest ascent? Well, it builds on the factorialtype DOE. Let’s say that you’ve fractionated a design. Picked a smaller set of variables and then followed up with a full factorial. Once you have the Y=f(x) and S=g(x) you can look to see whick path takes you to more desirable response levels. This is accomplished by either taking the practical approach of asking Minitab to make a cube plot and to pick off any trends of desirable improvement or you can go analytical and take the the 1st order derivative of the Y=f(x) model that you’ve developed. This will provide you with a gradient vector (g) that will provide insight into where the next iteration of the experiment should be run. So the cliff note version of RSM would be:Run factorial designs
Add center points
Look for significance of a quadratic factor (if missing move further up the gradientif present add axial points.)
Hope this helps. To do this topic justice we would need a discussion forum focused on just DOE, but this gives you some insight into the general concepts. I’d highly recommend the following books: Design and Analysis of Experiments, Montgomery, 5th ed. and Reponse Surface Methodology, Myers and Montgomery.
Regards,
Erik
0July 26, 2002 at 4:25 pm #77638Manee,
I’d be cautious in advocating Taguchi methods as a preferred experimentation approach. The designs, and analysis via SN ratios, has been shown over and over to yield unnecessarily complex designs that are inefficient in ultimately yielding the Y=F(x) and S=g(x) models. There is a lot of literature out there that reinforces the sequential building approach of factorial experimentation and design.
Regards,
Erik0July 26, 2002 at 9:49 pm #77644Erik
I think you should read the book by Phadke “Quality Engineering using Robust Design”. There is a comparison given in one chapter between DOE and Robust Design. There are other books on that subject too
Manee0July 27, 2002 at 12:12 am #77647Douglas Montgomery’s latest edition (5th?) of his Design & Analysis of Experiments has a chapter on Robust Design. His approach provides the modern (post Taguchi) approach based on Response Surface methods. I do wish that section provide a little more guidance and more examples though.
0July 27, 2002 at 7:26 pm #77654Manee,
I have read Phadke’s book on the subject. While I do recognize the loss function that Taguchi advocated and pushing for robust design, I do not believe the experiments that he ‘modified’, based on S/N ratios, to be as efficient and/or effective in deriving the optimal Y=f(x) and S=g(x) response models.
Thanks,
Erik0July 28, 2002 at 12:41 am #77655
John J. FlaigParticipant@JohnFlaig Include @JohnFlaig in your post and this person will
be notified via email.Erik,
Your analysis is absolutely correct. The Taguchi approach to optimization is fundamentally flawed. It may work on some problems, but not on others. When it works people say “see I told you it works”. But the two step optimization has serious heuristic problems. For example, Taguchi breaks the controllable factors down into control and signal. Control varibles affect the variance, signal variables affect the mean. OK suppose the only active variable you have affects both the mean and the variance, and as you reduce variance the mean is moved away from the target. Two step optimization will not work.
John Flaig, Ph.D.
Applied Technology
http://www.eATUSA.com
0July 28, 2002 at 8:46 pm #77659For more details / exploration of this area take a look at:
– Myers, R.H. and Montgomery, D.C., Response Surface Methodology: Process and Productr Optimization Using Designed Experiments
– Hu, C.F.J. and Hamada, M., Experiments: Planning, Analysis, and Parameter Design Optimization
Bob0August 2, 2002 at 5:38 pm #77801The best way I have found is by using a very awesome 6sigma software package I purchased a couple years ago. It is very easy to use and does a great job. It actually is the best software for the entire sixsigma process I have ever seen. It does it all from problem statement to DOE to SPC. If you are intersted let me know and I will track down some contact info for you. [email protected]
Best wishes to you,
Tom Dunn0November 5, 2004 at 5:38 am #110320
MBB HelpParticipant@MBBHelp Include @MBBHelp in your post and this person will
be notified via email.If you’re still out there. . . can you tell me more about the software you mentioned.
0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.