home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!sun-barr!olivea!mintaka.lcs.mit.edu!bloom-picayune.mit.edu!athena.mit.edu!cid
- From: cid@athena.mit.edu (Derek H Cedillo)
- Newsgroups: sci.engr.mech
- Subject: Taguchi Program Flopped (for me at least)
- Message-ID: <1992Aug31.123940.20238@athena.mit.edu>
- Date: 31 Aug 92 12:39:40 GMT
- Sender: news@athena.mit.edu (News system)
- Organization: Massachusetts Institute of Technology
- Lines: 23
- Nntp-Posting-Host: e40-008-9.mit.edu
-
-
- Well,
- for all of you interested in knowing about the great taguchi base program,
- I have no idea. I was short on time, so I only got a few hours to try
- and figure out the program. I couldn't. There is a class offered for
- it though, that would be real interesting. In my case, I didnt
- have so many variables. I had a lot of interspersed dependant
- variables, and seems like the program is better for multiple
- independant variables. It also seemed like a worthless peice of junk
- in my case. It asked for the indep. variable ranges, and then asked
- how I expected them to relate, quadraticly or simple, then it had the nerve
- to ask me for my results. I thought this was an aide to help you design
- the experiment, But I no longer think it does. It plotted my results
- then, and I got nothing better than if I hand plotted, or used Delta
- or Cricket Graph or whatever. But this very well could be becuase I only
- had 2 hours with the program. Im sure with the class and more time,
- there is something worthwhile in there. I will post the name of the program,
- and any contacts in regards to tthe class (May have only been GE offered)
- and keep you posted.
-
-
-
- Derek
-