Attribute MSA Study
Six Sigma – iSixSigma › Forums › Old Forums › General › Attribute MSA Study
- This topic has 97 replies, 77 voices, and was last updated 13 years, 2 months ago by
Wissanu.
-
AuthorPosts
-
December 18, 2002 at 6:28 pm #31049
I am trying to understand how to perform attribute measurement system study. I started at page 125 of the 3rd edition of the Measurement Systems Analysis manual and got lost as to how to perform these studies. Does anyone have any insight on what to read to better understand this stuff?
0December 18, 2002 at 6:59 pm #81590It is not real well documented.
Go back to the folks who taught you Six Sigma — they should have good information.0December 18, 2002 at 9:19 pm #81595… and if I have not gone through BB / six sigma training, where might I go to look for infomation on attribute MSA studies?
0December 18, 2002 at 9:27 pm #81596
Joy CowingParticipant@Joy-CowingInclude @Joy-Cowing in your post and this person will
be notified via email.Matt,
Here are some basic guidelines for conducting an attribute study:
1) Select 20 parts or samples for inspection. These samples should include some clearly good, some clearly bad, and some borderline. Overall, I would try to end up with 10 good and 10 bad.
2) Have 2 – 3 appraisers inspect the samples in random order. They should each do this twice. This will allow you to assess how often they agree with each other (reproducibility) and how consistent they are in their own assessments (repeatability).
3) Compare these results to the standard. This part can be tricky. When you put the study together, you should have a team or expert do this evaluation. I say it is tricky because many times, the standards are just being developed or haven’t been developed at all.
4) Plug the data into Excel or Minitab for the analysis. You are looking for an overall assessment agreement of 90% or more to accept your measurement system. This means that 90% of the time your appraisers need to agree with themselves, each other and the standard. Deficiencies should be addressed with training and development of standards (basic rules, visuals, etc).
I hope this helps.
Joy0December 18, 2002 at 9:53 pm #81599Go to the AIAG website and order their MSA book.
0December 18, 2002 at 10:00 pm #81600I have AIAG’s the MSA’s book, but I am attempting to figure out the Cross-Tab stuff. It is not very clear how this is done. I will re-re-re-read the 10 pages. Perhaps the light will turn on.
0December 31, 2002 at 5:14 am #81778Dear Matt, will you please give me a copy about MSA book. my email address: [email protected] thank you!
Best Regards!0January 2, 2003 at 6:13 am #81798
HemanthParticipant@HemanthInclude @Hemanth in your post and this person will
be notified via email.Hi
Adding further to Joy’s reply, try analysing this way:
Overall compliance: This is the ratio of number of cases where all findings are matching / total number of cases. say out of 20 defects there were only 18 defects where all the readings of inspectors matched with the experts, then overall compliance = 18/20 = 0.9 or 90%. Overall compliance muste be at least 80%, >90% is desirable.
Individual operator repeatablity: This is the ratio of cases where both the readings for the inspector match / total number of cases. Say for example Inspector A repeated his results in 15 cases out of 20, then his repeatablity is 75%. guidelines are same for this criteria.
Individual operator reproduciblity: This is the ratio of cases where both the readings for the inspector match with the expert’s / total number of cases.
This method is very simple to analyse and explain to team.
Hope this helped
Hemanth0January 2, 2003 at 2:57 pm #81808The MSA book that I am using is sold through the AIAG (Automotive Industry Action group.) You can find out more from: http://www.aiag.org/publications/quality/dcxfordgm.html
0January 2, 2003 at 5:36 pm #81814
Chip HewetteParticipant@Chip-HewetteInclude @Chip-Hewette in your post and this person will
be notified via email.Think of attribute measurement studies as a way to describe the randomness or error in the attribute inspection. An old QC saying is “100% inspection is 85% effective.”
One must also consider how many defects or nonconformances exist generally, in the process of interest. If the number of defects is very small, relative to the production of goods or services, then the attribute measurement system study should be expanded to ensure that the truth can be measured. If, for instance, the defect rate is 5,000 dpmo, and the attribute measurement system shows error of 4,000 dpmo, there is a barrier to improvement. Conversely, if the defect rate is 30,000 dpmo, the same measurement system would likely show improvement.
Don’t forget if you have products or services with multiple opportunities for nonconformances that the measurement system study must encompass these opportunities. 10 objects with 10 possible mistakes gives the measurer 100 opportunities.0January 2, 2003 at 8:23 pm #81819My (new) question has to do with the confidence interval bounds for attribute MSA studies.
On page 131 of the AIAG, MSA 3rd edition, it shows non-symmetrical confidence interval bounds. Can someone help explain where or how are these derived? If you know the answer, does this also explain why they are not symmetrical?0February 10, 2003 at 2:54 pm #82876Hello,
I am going to set up a gage R&R and have mostly found information as to pass/fail criteria only. I am measuring length and width and don’t necessarily have a pass/fail criteria. Can I use the raw data instead of using a pass/fail criteria?
Thanks,0June 23, 2003 at 6:30 am #87231Hi,
I have been working on Attribute MSA study and i have found some formulas and answers about that cross tabulation methot. I want to discuss it with you,if you are interested please email me on [email protected].
Best regards
Ozler0October 30, 2003 at 5:39 am #91793Hi..I am new to MSA, and am doing some own studies.
I noticed that most tests are made on 10-3-3 version (10 parts, 3 appraisers, 3 trials). I was wondering whether there are other combinations (eg 5-2-2, 20-4-6..). Plus, can the 10-3-3 formulas/calculations applied to those other versions? How about the constant values of K1, K2 and K3 ?
I heard that the MSA book by AIAG has the complete values of the constants, but I do not have access to purchasing the book. Is there anyway I can get the list of the constants..?
Please help…0November 4, 2003 at 6:59 pm #92051
Brian KarneskyParticipant@Brian-KarneskyInclude @Brian-Karnesky in your post and this person will
be notified via email.All:
I’m struggleing with an Attribute Gage R&R set up for a vision system. I’m confused about appraisers – since parts are fixtured and the Vision system is the judge based on programed limits.
Help – please feel free to e-mail me directly at [email protected]
Thanks0November 4, 2003 at 7:49 pm #92056Brian,
As with any experiment, you must first decide what factors to examine. Choosing factors is most often done by informed intuition, not hard data, because if you had the data you would not need the experiment.
Having pontificated, I’ll say that your issue is to determine the effect of humans on your process. You could decide that there is no effect and skip the appraiser study; however, it has been my experience that everyone uses fixturing differently unless of course it is designed and built to be fail-safe.
The win-win scenario is you do the appraiser test and find no effect. I would consider trying to “force” the appraisers into fixturing the part differently to eliminate the chance that they randomly do it the same. Remember that your true objective is to identify potential variability in the measurement process, not to derive a number. Tests are done to confirm or refute our assumptions.0December 10, 2003 at 6:12 pm #93436This message is for Joy.
Joy, where did you obtain the 90% value for acceptance purposes ??
I am a technical trainer and this is the question I’m asked the most often ??
A detailed explanation of any sort would be most appreciated.0December 19, 2003 at 5:27 pm #93713Bob, in response to your question to Joy, I’ve found the 90% acceptable reference in one of Air Academy Associates’ training manual (Six Sigma Expert).
0December 19, 2003 at 5:48 pm #93714OK,
Let me see if I understand the experts correctly.
I have a process that produces 1,000 parts per day. The average cost of scrap or repair internally is $2,000 per part. The average cost of repair or replacement externally is $10,000. The process has a current capability of 4 sigma (66800 PPM defective) and therefore produces 67 defective parts per day on average. My gage R&R on the inspection was acceptable at .90. For simplicity, lets say that the alpha and beta risk of the inspection were equal components of this at 0.05 each.
Then
Number of false positives per day = (1,000-67)*.05 = 47
Number of false negatives = 67*.05 = 3.35
Then my cost of unnecessary repair or scrape = $2,000*47 = $94,000 per day
And cost due to field repair or replacement = ($10,000 $2,000)*3.35 = $26,800 per day
Total annual cost of the qualified inspections system = $120,800 * 365 = $44,092,000
Well I guess I have a different definition of an acceptable inspection system!
Statman0December 19, 2003 at 6:03 pm #93715I’m sorry 3 sigma is 66800 PPM not 4 sigma
Oops.0December 19, 2003 at 6:13 pm #93716
GabrielParticipant@GabrielInclude @Gabriel in your post and this person will
be notified via email.Statman, I think you are overreacting.
First, it is pretty possible that with such a huge cost of repair, parts that are found non conforming go to further investigation (such as 2 more checks) before reapair. What would be the chance that a good part is found to be defective in 2 out of 3 checks with this betta risk?
Second, specifications do have a margin of safety. A part that is just a little bit out of spec is not much different from a part that is just a little bit in spec (ask Taguchi). In othr words, if the specification is 10±0.2, a part at 10.19 is not that different from a part at 10.21, so if a part at 10.21 goes out you will probably not have a complaint because of it, at least not with a significant higher frequency than the return of parts at 10.19. And the chance to detect parts out of spec increases as they are farther out of spec. So the chances to miss a very bad part is typically pretty low when you have such a measurement system.
Do you really think that your PC, where you are writing this post from, did comply with ALL product and process specifications at all process stages if you went all the way back to the raw silicon, polimers and gold? And it is still working, isn’t it?0December 19, 2003 at 6:58 pm #93718Hi Gabriel,
My post wouldnt have had the impact if I had set the parameters so that the annual cost of the alpha and beta risk was $50 dollars a year J.
What you have done is emphasized my point. Which is that you should not accept some general rule of thumb about an inspection system without a cost/benefit analysis of the consequences.
Since it would be unacceptable to have such a high cost of unnecessary internal repair due to the inspections system alpha risk, we would have to compensate by adding additional gates. But OK, lets go with that. The parts are inspected two more times before repair, each at the same alpha and beta level of the first (.05).
Therefore,
Test 2 will see 47 false positives and 64.65 actual defectives per day and have these results:
47*.05 = 2.35 false positives going on to inspection number 3
64.65*.05 = 3.23 false negative shipped to the customer
Test 3 will see 2.35 false positives and 61.32 actual defectives per day with these results:
2.35*.05 = .1175 false positives per day on to repair and
61.32*.05 = 3.066 false negatives shipped to the customer
Then there will be (3.35 + 3.23 + 3.066) = 9.646 false positives per day shipped to the customer and
0.1175 false negatives repaired
Then my cost of unnecessary repair or scrape = $2,000*0.1175 = $235 per day
And cost due to field repair or replacement = ($10,000 $2,000)*9.646 = $77,168per day
Total annual cost of the qualified inspections system with 2 additional inspections = $77,403 * 365 = $28,252,095
So yes, we can reduce the cost of false positives with more internal inspection. However, we increase the number of defectives shipped to the customer as well as the cost of additional inspection, complexity, WIP, and process cycle time, and lower customer satisfaction.
To your second point, if there was a variables measurement with in which safety limits could be set up, you would go with that measurement as opposed to the pass/fail attribute gage.
Dont get me started on the number of defects associated with my PC K.
Cheers,
Statman0December 23, 2003 at 2:38 am #93784Bill,
Always remember that if possible use a continuous data rather that attribute data because continuous data is rich of information.
Width and Length are continuous measurement, just use either of the two in your GRR. You may test 2 operators, 10 or more samples, 2 trials each operator. Make sure your samples are numbered but must be measured randomly ( 2,4,3,6,10,…7 ). Randomize the samples to eliminate bias ( operator might memorize the length/width per sample ). If possible after the first operator measured all the samples, let the second operator do the measuring, then back to the first operator.
You can use minitab to get the result or if you don’t have minitab, you can post your collected data and your e-mail address. I will do it for you.
AbetF
0July 29, 2004 at 2:06 pm #10461866800 PPM is 1.5 sigma….not 3 sigma. Unless you have some basis for a 1.5 sigma shift between long and short term.
0September 9, 2004 at 1:09 am #107037Dear Ozler,
I am very new in MSA study and have lost when I reach the cross tabulation method for attribute measurement. I hope you can share the formulas that you have found.
Thanks a lot.
Wilfred0September 13, 2004 at 9:33 am #107263Hi Wilfred,
If you can write me your email address, i can send you a working sheet about MSA attribute study.
Regards,
Ozler0September 13, 2004 at 2:08 pm #107285Dear Ozler,
Please send the workig sheet to my yahoo e-mail address: [email protected].
Thank you very much.
Regards,
Wilfred0September 13, 2004 at 5:53 pm #107297These type of studies are more and more frequent as six sigma moves off the shop floor and into office processes. If you have minitab follow the examples given. If you do not look up Kaplan and follow that model.
0September 14, 2004 at 4:05 am #107320Try these two links. I think they will help you alot. The sight is under construction however, it has a lot of helpfull information right now. The first link is just about gage R&Rs. The second link is just tons of Six Sigma help. Good luck and I hope this helps
http://dbar-innovations.com/sixsigma/password/mainmenu.gif
http://dbar-innovations.com/sixsigma/general.html0September 14, 2004 at 4:55 am #107323
Tom OooopsMember@Tom-OooopsInclude @Tom-Oooops in your post and this person will
be notified via email.I am sorry I gave the wrong link to Gage R&R before. I could screw up a one car funneral. I hope this helps.
http://dbar-innovations.com/sixsigma/password/gagrr.html0October 14, 2004 at 6:35 am #109075
A.KarimParticipant@A.KarimInclude @A.Karim in your post and this person will
be notified via email.Hai Ozler
I am Karim, just starting to implement Attribute MSA Studies. Can you e-mail me your MSA working sheet
Thank
Regards
Karim
0December 28, 2004 at 7:28 am #112862Hi Ozler,
ShRLi this, got in the same situation here, can i also have a copy of the work sheet about MSA Attribute study? my email: [email protected] or at [email protected]
thanks
I ‘ll be glad if those who already have could share it to me.
thanks again
0December 28, 2004 at 7:28 am #112863Hi Ozler,
ShRLi this, got in the same situation here, can i also have a copy of the work sheet about MSA Attribute study? my email: [email protected] or at [email protected]
thanks
I ‘ll be glad if those who already have could share it to me.
thanks again
0January 14, 2005 at 8:39 am #113461
Agustín AsúaParticipant@Agustín-AsúaInclude @Agustín-Asúa in your post and this person will
be notified via email.I am trying to perform Attribute Measurement Systems Studies, I got MSA 3th Edition, I want to use the Cross Tabn Method and I get lost when I try to find out the formulas used. Can somebody clarify to me the real formula used in MSA pag 128 to132?
0February 1, 2005 at 7:25 pm #114305Ozler,
I would greatly appreciate you sending me a working sheet about MSA Attribute Study too. I have pulled up several articles on the internet and there seems to be tons of info out there ……..Unfortunately it seems that I can get lost in the first 5 minutes of reading.
Thanks, Leona
0February 1, 2005 at 7:28 pm #114307Ozler,
Sorry …..I forgot to send my email address.
Leona
[email protected]
0February 2, 2005 at 12:00 pm #114336Hello Leona,
If you can write me your email address, i can send you the working sheet.
Regards,
Ozler0February 2, 2005 at 4:30 pm #114358Ozler,
My email address is [email protected]
Thank you so much for responding and I look forward to utilizing your worksheet.
Thanks again,
Leona
0April 6, 2005 at 8:08 pm #117315Ozler
also please send to me the workig sheet to my hotmail e-mail address: [email protected] you very much.Regards,0April 7, 2005 at 4:21 am #117340Hi Ozler
Can you send me a copy of Attribute MSA work sheet
[email protected]
Thanks
0April 13, 2005 at 11:05 pm #117698Hi,
Can anyone explain how can I perform a study to verify the capability of a machine operators to inspect product in process. I want to know if they’re capable of detecting the defective parts with the current process, while loading (some operators) and packaging (some others)the parts in the machine at a given rate. I was thinking about performing a % agreement Kappa Attribute Study but I’m not sure how can I apply that to a crew of operators while running the machine. Could anyone recommend other methods???? I’m really new to this and I have not had any kind of Six Sigma training.
Thanks, any help will be appreciated!!0April 28, 2005 at 3:14 am #118569Hi,
I have gone through the MSA Reference Manual, 3rd eddition, pg. 125-140. And, I have tried to key in the data (pg. 127) by using QI Macro, found that the results shown are difference with the MSA Reference Manual. The LSL and USL that I used is 0.45 and 0.55 with 25% R&R. May anyone know why was this happen?
Hi Ozler,
May you please send me a copy of MSA as you mentioned before? My e-mail add. is ” [email protected] “.
Much appreciated if you can send me a copy. Thank you.
Warm regards,
Alice0May 3, 2005 at 4:09 pm #118797
Juan PabloParticipant@Juan-PabloInclude @Juan-Pablo in your post and this person will
be notified via email.Hello,
I am performing a research about the relation between process capability studies and uncertainty in measurement systems.
Could you recomend me some literature that could help me in my research?
Thanks.0May 20, 2005 at 9:10 pm #119886Hi Ozler,
I too am lost like others when I started reading the MSA3rd Edition for Attribute data. Can you pleaseeee send the working sheet for the same.
my email id: [email protected]
Thank you in advance.
Regards
Manju
0June 8, 2005 at 6:48 pm #120964Would you send MSA sample sheet to me? Thanks.
Vincent0June 13, 2005 at 4:25 pm #121300Hi Ozler
Please e-mail me the xls worksheet for attribute gage R&R. I have also got lost to find expected count in the cross tab method.
My e-mail ID is [email protected]
Thanks
Piyush0June 16, 2005 at 5:41 am #121507Hi Alice,
I’m afraid to tell you that the cited manual is plaint of errors, in particular the section you’re talking about. Please, go to http://main.isixsigma.com/forum/showmessage.asp?messageID=69042
to see the explicitly recognized errors, as soon as possible, because AIAG has the habit of change its links. Otherwise, you can access the AIAG’s webpage and make a search for something as “Errata for MSA”
In case of doubt, let me know about, I have all the amended sheets.
__________________
Ozler,
I have no problems at all, but I’m courius about your greated demanded sheet, please send me a copy of your MSA document. My e-mail address is [email protected]
Best regards,
Dan.0June 16, 2005 at 5:51 am #121508Hi Alice,
Ooooopsss!!!
I made a great mistake of the class “copy&paste”
The right address to search for is http://www.aiag.org/publications/quality/msa3_errata.cfm
I’m sorry.0June 16, 2005 at 10:12 am #121529If the measurement is not ranked then use a Kappa study.
If it is ranked use Intra Class Correlation.
I have used both in the past when training and also in practice.
The main problems with them is finding the right tool to make the statistics easy. My own organisation has good excel spreadsheets for both, and I am sure that some will exist on the web.0June 22, 2005 at 1:37 am #121907Hi Daniel,
Much thanks for your valuable knowledge sharing. It’s really under my expectation about the erratum making.
But, I’m still confusing with the “Ref Value” that mentioned on page 127 (Attribute Study Database). From my understanding, the “Ref. value” might not be same for difference experiment. And, have to determined by ourselve. Am I right?
My questions:
1. How could we determine the value since it’s an attribute data?
2. Should we classified the defects in categories, for instance, accepted, marginal accepted, rejected?
Do you have any ideas about the reference value? I’m lost…
Secondly, I’m still getting a diffence answer for the example on page 127. I will send you a copy true your email, [email protected] . May you please help me to have a check? Thank you.
Best Regards,
Alice0June 29, 2005 at 5:47 am #122318Hello Alice!I answered you some days ago the e-mail that you sent to me, I transcribe such answer below, with some aditional comments.
_____________________________________
Hi Alice!
By now, I can’t have access to the AIAG’s manual (That means 5 or 6 weaks). I looked your file and, at a first glance, I noted that formulas in cells AM9 and AR9 are apparently wrong. Also, I loaded your data in Minitab and the results are consistent with those from the AIAG’s manual, page 128.
However, I don’t remember what’s said on page 127. Could you scan this page (and the previous ones, if you considered it pertinent) and send me it by e-mail?
I promise you that I will take me the time necessary to analyze your problem and to write to you an adequate answer.
Best regards,Daniel
__________________________________________________With respect to the value of reference, typically it is related to the value established by an immediately superior instance. That is, for evaluations at floor level, it would be the metrology’s laboratory value of the company; for this one, some regional laboratory, and so on. It will be taken as the “true value” for the quality’s characteristic that it’s being evaluated.
With respect to the second question, perhaps I am a little old minded, but if we are going to take a statistical model from Poisson or Binomial, then there are only place for two, or numerable results: Or it happens to pass or it does not pass, or well it has 1, 2, or any number “n” of defects.0June 30, 2005 at 1:05 am #122374Hello Ozler or Any Good Sumaritan,
I am having a hard time pursuing MSA study for Attribute Gage and have lost when I reach the cross tabulation method of measurement. I would greatly appreciate if you can share any relevant information or formulas that I can use that could help me get back on track. My email address is [email protected]
Thanks a lot.
Hector0July 5, 2005 at 2:02 am #122492Good day, Daniel!
I’ve sent you the copies last week.
I do try to loaded the data in Minitab, there are 2 type of attribute:
1. Attribute Agreement Analysis – results are consistent with AIAG’s manual.
2. Attribute Gage Study (Analytical Method) – FAILED to generate the results caused too few number of trials per part.
According to stat guide, Minitab accepts either summarized or raw data for attribute gage studies, with conditions, the number of trials for each part must be greater than or equal to 15. If using the AIAG method (default) to test the bias of the gage, then must have exactly 20 trials per part.
From my data, if ignored the appraiser, I’ve only 9 trials per part.
May you please consult me in above issue, Daniel?
Looking for your reply, and thanks in advance…
Best Regards,
Alice0July 5, 2005 at 5:02 am #122494Hello Alice!
Indeed, the form in which Minitab works accepts only your data for “Attribute Agreement Analysis”. If I am understanding, and you wish to evaluate the bias with your present data, I believe that you can use the previous method of the same AIAG.
Best regards,
Daniel0August 3, 2005 at 9:12 am #124180Hello Ozler,
I am having a hard time pursuing MSA study for Attribute Gage and have lost when I reach the cross tabulation method of measurement. I would greatly appreciate if you can share any relevant information or formulas that I can use that could help me get back on track.
My email address is serkan.sipahiyazaki-europe.com0September 15, 2005 at 1:47 pm #126890Hi OzlerCan you send me a copy of Attribute MSA work sheet
My e-mail adress is: [email protected]
Thank’s a lot0September 19, 2005 at 3:31 pm #127090Hi OzlerCan you send me a copy of Attribute MSA work sheet.
Thanks in anticipation,
Tony
[email protected]0September 21, 2005 at 4:27 pm #127243I have conducted an attribute R&R study for an inspection proccess (pass/fail). We currently do 100% inspection. Normally for the analysis you assume an alpha and beta risk. Is there a way of actually calculating it? Also, is there a way to assess a trade off between decreased inspection frequency and supplying customer with rejected parts? Thank you
0October 3, 2005 at 2:52 pm #127796Hi!
Can you send me a copy of Attribute MSA work sheet
My e-mail adress is: i.barriolahedisa.com
Thank’s in advance0November 14, 2005 at 10:37 am #129670Hi
I am not able to understand Attribute MSA study ( Both Hypothesis & Signal detection method. Could any of you please send me some info & excel sheet which you all have been referring. Pls send the info to [email protected]
Regards
Naveen0November 14, 2005 at 1:17 pm #129673
Engine LadyParticipant@Engine-LadyInclude @Engine-Lady in your post and this person will
be notified via email.Hi Ozler,
I too am lost like others when I started reading the MSA3rd Edition for Attribute data. Can you pleaseeee send the working sheet for the same.
my email id: [email protected]
Thank you in advance.
Regards
Susan0November 14, 2005 at 1:19 pm #129674
Engine LadyParticipant@Engine-LadyInclude @Engine-Lady in your post and this person will
be notified via email.Tom: I have had no luck in getting any of the addressesyou posted to work? Any suggestions?
Engine Lady0November 14, 2005 at 5:17 pm #129683Hi Ozler,
I have read the string of postings here and I am also interested in using a better method of Attrubute GR&R
Could you please send me a copy of your steadsheet.
My email address is [email protected]
Thank you in advance.
Don0January 13, 2006 at 5:56 pm #132394I am not a six sigma certified so need some basic help with doing Gage R&R or Attribute gage analysis on a gage that does not really do direct measurement. The gage is used to inspect voltages, currents and Temp on another equipment when the gages voltage, current, and Temp is being manipulated ( increase voltage or temp etc.). So how can I do Gage R&R or Attribute Gage study on this is the gage does not really measure anything? I need to show that the gage can show repeatability and reproducibility for this gage.
0March 31, 2006 at 3:56 am #135722
CharlezParticipant@CharlezInclude @Charlez in your post and this person will
be notified via email.Dear Ozler,
Greeting. I would like to get your advise to set up a proper MSA system in my company especially in attribute data.
I am very new in this MSA study and I am the person who has been assigned to setup the system in my company. When went through the AIAG 3rd edition MSA Manual. I have doubt when reach to page 128 which I am confusing about the cross table method. I cannot figure out from the manual how should I setup the study, especially the calculation and working sheet of the MSA attribute study. Can you send me a copy of yours and provide me some advise or guideline?
Thank you
Regards,
Charlez Wu C. M.
0March 31, 2006 at 4:02 am #135723
CharlezParticipant@CharlezInclude @Charlez in your post and this person will
be notified via email.Dear Ozler,
Greeting. I would like to get your advise to set up a proper MSA system in my company especially in attribute data.
I am very new in this MSA study and I am the person who has been assigned to setup the system in my company. When went through the AIAG 3rd edition MSA Manual. I have doubt when reach to page 128 which I am confusing about the cross table method. I cannot figure out from the manual how should I setup the study, especially the calculation and working sheet of the MSA attribute study. Can you send me a copy of yours and provide me some advise or guideline?
my email address is [email protected]_schweizer.com
Thank you
Regards,
Charlez Wu C. M.
0April 2, 2006 at 9:17 pm #135789Dear Ozler,
Could you please send me a copy of your steadsheet on attribute MSA
my e-mail address is [email protected]
Thank you,
Michael0April 21, 2006 at 8:28 am #136644
Phil SedmanParticipant@Phil-SedmanInclude @Phil-Sedman in your post and this person will
be notified via email.Hi
Looking for a relatively simple spreadsheet for compilation of Attribute GR&R. Understand you may have a suitable spreadsheet. Grateful if you could send me a copy.
my e-mail address is [email protected]
Thanks in advance.
Phil
0April 24, 2006 at 7:39 pm #136769Hi
I have a format for this study
On my memory stick sorry I dont have in this moment
But I can send tomorrow in the morning0June 13, 2006 at 1:04 pm #139000
andy daviesParticipant@andy-daviesInclude @andy-davies in your post and this person will
be notified via email.Dear Ozler,
Could you please send me a copy of your steadsheet on attribute MSA
my e-mail address is [email protected]
Thankyou,
Andy0June 13, 2006 at 8:13 pm #139049
Lynda TalbotParticipant@Lynda-TalbotInclude @Lynda-Talbot in your post and this person will
be notified via email.Hi Andy or Ozler,
Please send me a copy of your famous spreadsheet as well. I also am having great difficulty understanding this crosstab method.
Greatly appriciated,
thx,
Lynda
My email address is [email protected]0June 14, 2006 at 2:06 pm #139109While it has been a while on this thread, I would also appreciate a copy to [email protected]. Thanks.
0June 17, 2006 at 7:26 pm #139269
brayanneParticipant@brayanneInclude @brayanne in your post and this person will
be notified via email.Hi Oz… I understand you have this famous file. I am the SPC engineer for the plant and is currently doing MSA studies. MSAs for attributes are quite a challenge for me. If you don’t mind, I would like to have your file as well so I can make a good comparison from you. Thank you!
here’s my email address: [email protected]
will be waiting for it. Thank you in advance!0July 6, 2006 at 11:12 pm #140041Hi Ozler,
I too am very lost like others when I started reading the MSA3rd Edition for Attribute data. Can you pleaseeee send the working sheet for the same.
my email Old_School_2k6hotmail.com
Thank you in advance.
Regards
SPCimen0July 7, 2006 at 6:06 am #140048Matt,If you have access to Minitab v14, and even if you don’t perhaps this link might help:http://www.minitab.com/resources/tutorials/AttributeAgreementAnalysis/default.aspxThe MSA manual by AIAG is of little use and very inaccurate. I suggest you consider a second use for it as a door stop.If you want to understand how to do an MSA correctly consider looking at http://www.spcpress.comVP
0September 6, 2006 at 1:23 pm #142846
p.ilayarajaParticipant@p.ilayarajaInclude @p.ilayaraja in your post and this person will
be notified via email.Dear sir,
I have msa study ( 4 study) procedure & formats are request to sent my mail.
Regards,
P.ILAYARAJA.
9841905214.0September 6, 2006 at 9:48 pm #142874
La FinneParticipant@La-FinneInclude @La-Finne in your post and this person will
be notified via email.Then I would recomend a comic. Much more fun than MSA.
0November 10, 2006 at 9:45 pm #146889
BrainCellParticipant@BrainCellInclude @BrainCell in your post and this person will
be notified via email.Stan,
please go back and read your post to the original question. This is typical of you. You are a complete idiot. If you don’t have the intelligence or knowledge to give a better answer than this, why don’t you get lost.
Nobody who needs or wants to learn has time for crap like this.0November 15, 2006 at 2:52 pm #147254
PatriciaParticipant@PatriciaInclude @Patricia in your post and this person will
be notified via email.Ozler,
Can you please send me the table?
Patrícia0November 21, 2006 at 6:42 pm #147694Ozler,
I too would be interested in using your MSA attribute working sheet sheet as a reference.
Gary0December 4, 2006 at 8:02 am #148372Hi Ozler,
I would appreciate a lot if you would also include me sending your MSA attribute working sheet.
Thanks,
Dave0December 5, 2006 at 9:23 pm #148490Hi Ozler,
I’m working on resolving an attribute gauge issue and your spreadsheet appears to be of much help to many. Could you please send a copy? Thanks much.
My address is [email protected]0December 14, 2006 at 7:08 am #148961
frodhelParticipant@frodhelInclude @frodhel in your post and this person will
be notified via email.Hi Ozler
please do share with me a copy of your working attribute MSA worksheet . this would be of great help on my study of the subject, which also i’m confused.
thanks in advance
my e-mail add: [email protected]
frodhel0December 15, 2006 at 3:39 am #149046Hi, Joy
I wonder if you still do the good samaritan way to those who are in need. Just yesterday that i’ve found this site and very much appreciated the basic guidelines you reply to Matt 4 years ago.
I am new to MSA and particularly assigned to attribute study. I relied on the MSA 3rd but also got lost on cross-tab method. It would be of great help if you have further info to easily understand the formula used or other example to ellaborate the argument presented.
Thank you in advance.
lhed
0January 31, 2007 at 9:28 am #151306
vincent w. chenMember@vincent-w.-chenInclude @vincent-w.-chen in your post and this person will
be notified via email.Hi Ozler,
May I have ur Attribute GRR worksheet? Appreciate a lot.
[email protected] (my e-mail)
regards,
vincent0April 20, 2007 at 8:08 am #155022
david carbonellParticipant@david-carbonellInclude @david-carbonell in your post and this person will
be notified via email.Dear Mr. Ozler,
we are very interested in your MSA sheet, and we would very grateful if you could send us this document to my email adress : [email protected]
Thank you in advance your help.
Best regards,
David
0April 26, 2007 at 2:57 am #155275
nrsivakumarParticipant@nrsivakumarInclude @nrsivakumar in your post and this person will
be notified via email.msa study formet
0May 24, 2007 at 5:13 pm #156492
EgnaldoParticipant@EgnaldoInclude @Egnaldo in your post and this person will
be notified via email.I am new in msa too and wold like to get more information about
your wok sheet
can you send it to me
thanks
Egnaldo
0June 14, 2007 at 6:35 am #157407Hi Ozler
I want to do a MSA study on a vision system. Could you send me a copy of your excel sheet?
thanks in advance
[email protected]
Armand
0June 26, 2007 at 9:22 pm #157941Please send me a copy of your spread sheet. I would greatly appreciate it.
Thank you,
Sally0July 4, 2007 at 8:17 pm #158220Hi Ozler,
Can you please e-mail me attribute MSA worksheet?
My e-mail is [email protected]
Many thanks. Tom.0October 25, 2007 at 7:33 pm #163876
The RockmanMember@The-RockmanInclude @The-Rockman in your post and this person will
be notified via email.Could you please send me a copy of the attribute gage study worksheet. Thanks. [email protected]
0November 13, 2007 at 8:55 pm #164721
Al LavinParticipant@Al-LavinInclude @Al-Lavin in your post and this person will
be notified via email.Could you also send me a copy.
Regards
AL0November 28, 2007 at 4:34 pm #165393
John LanooyParticipant@John-LanooyInclude @John-Lanooy in your post and this person will
be notified via email.Hey Armand,
I live in the south of the Netherlands and also want to do an MSA on a vision system. Maybe we could share some experiences? Did you get the file from Ozler? Could you forward the file to me?
Thanks,
John Lanooy0December 28, 2007 at 2:26 am #166601The MSA v 3 manual on Attribute Gage R&R is about the worst I’ve seen. They leave out a lot of links in the calculations — such as conversion of the “count data” to the probabilities, estimation of the Prob of chance, etc. And they left out how they arrived at the Miss Rate and False Alarm Rates.
A better article is “When Quality is a Matter of Taste, Use of Reliability Indexes.” by David Futrell (Quality Progress, May, 1995) p81. This is a good overview of the method and provides examples. Also, crosstabulation methods are available in just about every stat textbook which includes nonparametric methods. But these usually don’t cover the “Kappa” calculation.
I don’t have any insight or reference covering the “Miss Rate” and “False Alarm Rates”.0May 4, 2008 at 11:55 am #171733
N.SugumarParticipant@N.SugumarInclude @N.Sugumar in your post and this person will
be notified via email.Dear Sir
Kindly Send To Me formet And Exp For All Details About Attribute Study
Thankng You
Regard
N.Sugumar0April 25, 2009 at 4:29 am #183699I recently 100% inspected some parts for attribute failures. The supplier had shipped approximately 1 or 2 defective parts per 100 parts for three previous shipments. The defect was a threaded hole was not in the key way as shown on the print. When I inspected these parts it appeared that all the the threaded holes were centered in the key way. I was later shown that their were burr’s on at least 60% of threaded holes which was called out on the notes as no burr’s. I was called negligent by my employer in regards to my inspection. My question is can someone be predetermined or their behavior changed to inspect for something they have seen before and miss an obvious defect? Does anyone have any data, especially quantitative data to support this theory?
0April 25, 2009 at 9:32 am #183702Attribute MSA is part of Six Sigma ,it is not best way to apply with the simple problem .There is another applied MSA on this as if you can send your detail of your problem to me at [email protected]
Regards /Wissanu0 -
AuthorPosts
The forum ‘General’ is closed to new topics and replies.