Minitab Tutorial - Measurement System Analysis

Gage R&R using minitab. A look at MSA and how minitab facilitates simple analysis...
FREE DMAIC DOWNLOAD! click the link
paul-allen.mykajabi.com/dmaic...
#Lean Six Sigma
#Six Sigma

[@kalyanbasak5344]

Best MSA explanation using Minitab that I have come across so far. Brilliant 🎉

[@Tkirby1405]

This video has been a huge help - I have ran a MSA by watching this and feel very confident now. Thank you Paul - you have a new subscriber here looking forward to seeing more of your videos :)

[@ankuragari]

Thanks! It really is a great informative video. You have explained and demonstrated the functionality of MSA in Minitab .It has given a great clarity.

[@paulallen5321]

You are welcome!

[@nave2n]

I started to watch your video from yesterday. I feel you teach more practically than theoretically.

[@paulallen5321]

Naveen - thanks for your positive comments...

[@Vs17Oct72]

Great reference, my first go to for all SPC concepts..

[@mamtakathuria6653]

Really Wonderful Explanation!! Mr. Paul

[@paulallen5321]

Cheers Mamta - Glad you liked it...

[@plantzengineering9925]

Hi Paul! Thanks for the comprehensive tutorial! I know you mentioned crossed R&R tests are for NON-destructive testing, however I believe they are also utilized for destructive batch testing (as opposed to single part measurement). I may be mistaken, so your insight would be much appreciated!

[@07warka]

Brilliant Paul

[@Daniel-gw6ss]

Great videos! Thank you

[@paulallen5321]

Daniel - Thankyou for your positive comments. If you'd like more of the same my book 'drink tea and read the paper' can be bought from LULU.COM. It contains all the principles of the video's...

[@PerfectlyLegitUser92]

great stuff, thank you so much!

[@paulallen5321]

Cheers Bob - Thanks for the support...

[@saarakiriwellage5572]

That was great explanation.

[@paulallen5321]

Saara - thanks for your positive comments. If you're into this kind of thing you might like my book. Drink tea and read the paper - you can get it direct from LULU.COM

[@MUSICAL_E]

Great explanation Paul. Please suggest is answers to three of your questions can be called inferences drawn?

[@paulallen5321]

Jaspreet - I think you could say that, the first 2 questions are answered with definitive numbers so maybe not an inference. This last question is an inference though...

[@elsonspornraft2388]

Nice content. Thanks

[@paulallen5321]

Elson - Thanks for the positive comment... if you like my video content take a look at my book 'Drink tea and read the paper' you can get it from LULU.COM...and keep watching!

[@abouzidnabil2749]

Good Tutorial Really Efficient

[@internationalleansixsigmai6828]

Paul Allen is back up and running!!

[@paulallen5321]

Adouzid - Thanks for the comment, if you like my video's please take a look at my book 'dink tea and read the paper' on LULU.COM. I think there is discount currently....

[@abderahimmoustahssin9559]

Thanks Paul , it 's very interesting, one question more, do you have any example with nested method? Thanks

[@paulallen5321]

Hi Abderahim - I don't unfortunately, I'm not a big fan of this technique as I'm not sure it really tells you anything. My first question is do you have a CPK problem?

[@anthonyvenuto9416]

well done

[@paulallen5321]

Cheers Anthony.....

[@DucNguyen-df6pz]

Thanks for the video, Paul! However, just wondering how can it be blind to the part number thru out different times of measurement? We need to make sure that our system is measuring exact the same part in order to see the difference comes from the measurement system. Right?

[@paulallen5321]

Hi Duc - yes the system is blind to the sample, but you know exactly which sample it is measuring, so you can compare repeated measurements

[@puravchoksi3984]

Hello Paul ! Great explanation. I really enjoy watching your videos.They are very helpful and I learnt a lot. Can you please let me know are there any specific reason in keeping acceptance criteria to 10%? I mean why most industry standard is 10% and not 5% or 15%? Why specific 10%? Thanks!

[@paulallen5321]

Purav - It's to do with the shape of the normal distribution, the tail is very thin out at 10%. So although you have error it is applying to very few items....

[@puravchoksi3984]

@@paulallen5321 Understood. Thank you for responding!

[@NguyenChienSurvivalNTF]

thanks a lot sir

[@paulallen5321]

Most welcome

[@stockmarketguru5155]

Thanks for your explanation through MINITAB...Can we take 5 or 4 standard deviation in calculation ??

[@paulallen5321]

Hi Guru - please ask the question again I'm not sure I uderstand...

[@jorgecampo7962]

Hello Paul, I hope you are doing great? would you mind explaining me agian the diferenece between ANOVA and XBar and R analysis? I dont really catch get it, what do you mean when you said measurement system that is doing one or two repeats? thank you.

[@paulallen5321]

Jorge - No problem, sometimes when you do an MSA you only have one peice of equipment or system to test. So if you have an automatic measuring system and there is only one of them then you'll only do a single system repeat. So more than likely you'll have a pair of measurements from one system to evaluate. 2 data points not 4 when you 2 x 2 data points. When you only have one measurement system like this you can only use the X Bar R method.

[@jorgecampo7962]

@@paulallen5321 Perfect, thank you!

[@zekkez]

Hi Paul, should the study coordinator define the sample measurement size before giving them to Appraisers to measure? Also, when would a tolerance (USL-LSL) be necessary to apply? Thank you

[@paulallen5321]

Hi there - no there is no need to know the standard, that happens when you calibrate. This is a test of repeating the same measure for the same part...

[@prasadkorde9892]

What does c4 by operator indicates? is that mean operator 2 is slight inconsistent with measurements?

[@paulallen5321]

Prasad - good to hear from you, no C4 is just to column where the data we collected is sitting in the spreadsheet..

[@jaygomez2320]

Hello sir! Thank you for this! :) Can get a copy of the dataset? THankyou!

[@paulallen5321]

Hi Jay - unfortunately I don't have that data file any more I think I only created it for the video. if you stop the video at 9 mins you can copy the data from there. To be honest given have complicated minitab can be I think practicing is the set up is the most important thing to do.

[@francisdantas8513]

Hi Paul. Just if the Problem in Between the system, than which graph do i have to see to find out what i have to do

[@paulallen5321]

Francis - there should be a graph that plots the averages for each part/operator. If these averages are different then the problem is between not within system...

[@abhipatel6757]

@@paulallen5321 in that case, Should we change/update measurement system right? Is there any another way to solve this between measurement system?

[@bmanojreddy]

Can you help me how to populate data in Minitab if we have more than 2 or 5 measurement data?

[@paulallen5321]

best way is to put the data in a block and then calcuate row data - average. The analyse that one column...

[@AnthonyTilleyVideography]

Do these 10 parts have to all be the same parts?

[@donnymac575]

Can machine or the calibration system replace operator for measurements that are automated with no human operator?

[@paulallen5321]

Hi Don - lots of people replace people with automated systems. But when they do an MSA on it. They still don't work perfectly each time. Whether is machines or not the MSA will tell you how good the system you've chosen is...

[@donnymac575]

@@paulallen5321 Thanks Paul!

[@achinesemac8577]

Much easier to understand

[@paulallen5321]

Thanks for the positive comments - If you like the advice in the video's take a look at my book 'Drink tea and read the paper' you can get it from LULU.COM

[@renatoribas5172]

Finally a British accent in a fantastic subject !

[@paulallen5321]

Renato - thanks for the support and ther is another 200 video's for you to watch...

[@renatoribas5172]

@@paulallen5321 This is great man, I'm gonna head over to GD&T and listening skills whilst "take time to smell the roses" haha

[@prasadkorde9892]

what does C4 BYPARTS INDICATES AND ALSO X BAR CHART? IN YOUR CASE PART TO PART VARIABILITY IS HIGHER SO WHY IT IS NOT CONSIDERED?

[@paulallen5321]

Prasad - X bar means average chart and part to part variability is of course what we want. All the variability should be in the parts not in the measurement system. What we're looking ar is if the variability coming from the measurement system is too large...

[@prasadkorde9892]

@@paulallen5321 Thank you for your reply. but % study variation in case of Part to part is 94.43% which is higher that repeatability and reproducibility , so do we want higher % of part variability for a good system ? and what is reason behind it ?

[@abhipatel6757]

@@paulallen5321 is there any thing related for part to part measurement? I mean if we have 20 percent of part to part then is there any problem? I guess it just use to show how many different parts we selected for measurement sustem

[@paulallen5321]

Abhi - if you have only 20% from part to part, that must mean 80% of variation is coming from the measurement system. That would be a very poor measurement system, the best is that 100% is part to part and 0% is meaurement error...

[@diarychannel4529]

What the aplication sir? I want to learn about Minitab.

[@basilpaul7248]

I have a doubt? Is this a measurement system analysis to check the quality of the product or the quality of the weighting machine?

[@paulallen5321]

Basil - the MSA checks the quality of the measurement system. that is the equipment plus method plus material plus people using it...

[@basilpaul7248]

@@paulallen5321 thanks for your reply, actually I have sent a email regarding my doubts. I need a experimental design. Can you help me for that?

[@basilpaul7248]

I mean how to check the quality of product.

[@paulallen5321]

Basil - I will reply to that as well...

[@paulallen5321]

The MSA maths will separate the difference created by parts and the difference created by the measurement system...

[@jaesunglee8565]

Repeatability has more variability than reproducibility in the case you shared in video. It's meaning that more variability is coming from machine itself not operators, isn;t it? But, you pointed out standardization of SOP and training to OP2 as a corrective action which are colsely related with reproducibility. ???

[@paulallen5321]

Hi Lee - no you connot seperate the Machine and operator like that. The variability is within system and the system is made of the m/c the operator the method and the variability on the parts...

[@HardikPatel-re9wz]

@@paulallen5321 i think ans of 3rd question- should also include calibration of tool 😅 bcz problem is repetition of measurement by measuring tool !

[@paulallen5321]

Hi Hardik - Calibration unfortunately cannot control variation in the measurement system, it simply controls precision..

[@suec.1387]

Mr. Allen, what version of Minitab are you using in your most recent tutorials?

[@internationalleansixsigmai6828]

Paul Allen is back up and running!!

[@paulallen5321]

Sue - if you like my video's please take a look at my book 'dink tea and read the paper' on LULU.COM. I think there is discount currently....

[@suec.1387]

@@paulallen5321 , good day! I do have your book! I purchased it about a year ago. Thanks for the helpful video content!