Sample size calculation given machine accuracy
Six Sigma – iSixSigma › Forums › Old Forums › General › Sample size calculation given machine accuracy
 This topic has 7 replies, 6 voices, and was last updated 16 years, 3 months ago by R.M.Parkhi.

AuthorPosts

April 10, 2006 at 3:11 pm #43038
TinocoParticipant@anthony Include @anthony in your post and this person will
be notified via email.I have been struggling in determining the required sample size given a machine’s accuracy. The testing will consist simply of checking the parts dimensions, and the machine’s accuracy is given in these terms. The population size is also known. Any help with this subject would be great. Thanks.
0April 10, 2006 at 4:08 pm #136194
Benjamin BantumParticipant@BenjaminBantum Include @BenjaminBantum in your post and this person will
be notified via email.I always use 14.
0April 10, 2006 at 4:09 pm #136195
ImpecuniousParticipant@Impecunious Include @Impecunious in your post and this person will
be notified via email.Sample size is a function of a few key properties: the minimum difference you want to detect, your desired power, the standard deviation of the data, and your desired confidence level.
The detectable difference seems odd to many people at first glance, but it’s simply the smallest difference that you “have” to detect to be meaningful for your particular purpose. For example, if your engineers begin to care when a part’s width is more than 0.5″ from some specification, then your minimum difference would be 0.5. All else equal, the smaller the difference you want to detect, the larger your sample size will need to be.
Power is the probability that you’ll correctly reject the null hypothesis when you should. Power is generally set somewhere between 0.7 and 0.85, although your particular situation may dictate otherwise.0April 10, 2006 at 8:25 pm #136206
GriswoldParticipant@chipgiii Include @chipgiii in your post and this person will
be notified via email.Anthony,
What’s your email address. I will send you a little sample calculator from excel. I use it from timetotime. But I think it will suit your needs.
chip0April 10, 2006 at 10:33 pm #136209
TinocoParticipant@anthony Include @anthony in your post and this person will
be notified via email.Hi Chip,
Thanks for your response. My email is [email protected]
Thanks.0April 11, 2006 at 1:54 am #136215
GriswoldParticipant@chipgiii Include @chipgiii in your post and this person will
be notified via email.Anthony,
I sent you the file, let me know how you make out. The post previous to this is on target as well. If you have minitab you can do the same thing and see the different power levels. The file I sent you is establishing the sampling amount at a 95% CI. There are two sheets, one for discrete data and one for continous. The curve on the chart to the right starts going to the right, that’s when increasing you sample size has a diminishing return.
Good luck,
Chip0April 17, 2006 at 2:12 pm #136470
Eugene LanningParticipant@EugeneLanning Include @EugeneLanning in your post and this person will
be notified via email.I’m looking at how often samples should be taken from a machine. Currently 5 samples are taken each hour, independent of production rate, standard deviation, confidence level desired, or anything else. Could the spreadsheet mentioned help bring some rigor into this situation. I also have access to MINITAB. Is there an ANSI std for this problem?
The spreadsheet may be fwd to [email protected]
0April 22, 2006 at 7:52 am #136683
R.M.ParkhiParticipant@R.M.Parkhi Include @R.M.Parkhi in your post and this person will
be notified via email.1. Machine accuracy as per manufacture’s recomendation is a prime requirement.
2. Since you the job accuracy, you start producing the jobs till you achieve 80 % of the total spread of the specs. This is more than sufficient. You should carry out Multivari analysis for the same.This will help you to identify the rootcauses of variations viz. within job, job to job & temporal i.e. time related. Attack no. one & no. two causes to reduce the variations to obtain CPk of 1.67.
You may get details in the book ” World Class Quality ” by Mr. Keki Bhote, publishers American Management Association.
With regards,
R.M.Parkhi0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.