How to Perform a Gauge R&R using the Average and Range Method (Part 2)

แชร์
ฝัง
  • เผยแพร่เมื่อ 23 พ.ย. 2024

ความคิดเห็น • 98

  • @manalisalvi211
    @manalisalvi211 8 หลายเดือนก่อน +7

    Absolutely helpful and such great easy to understand language!

    • @CQEAcademy
      @CQEAcademy 8 หลายเดือนก่อน

      Thanks for the awesome feedback!!!

    • @greenbeltacademy
      @greenbeltacademy  2 หลายเดือนก่อน

      Glad you think so!

  • @MehariA-h4g
    @MehariA-h4g ปีที่แล้ว +3

    amazing job, you made the life of GR&R very easy, but as you promised i am still waiting for the 3rd video

    • @CQEAcademy
      @CQEAcademy ปีที่แล้ว

      Thank you for the feedback, that 3rd video should be coming out in a few weeks!

  • @carlaluciagomezbaez8146
    @carlaluciagomezbaez8146 ปีที่แล้ว +1

    Watching from the Dominican Republic! The information has been beneficial for my CQE Preparation. I appreciate the time and effort!

  • @ZeyiTan
    @ZeyiTan 10 หลายเดือนก่อน +2

    Look forward to 3rd part.

  • @ayushrajrules
    @ayushrajrules ปีที่แล้ว +2

    Very informative video series. Thanks for putting in so much effort

  • @dannifeng8339
    @dannifeng8339 ปีที่แล้ว +2

    I have been waiting for this video for weeks. Thank you.🌹

  • @littlefatkz3929
    @littlefatkz3929 ปีที่แล้ว +1

    i'm waiting for your 3rd GRR video, appreciate, amazing

    • @greenbeltacademy
      @greenbeltacademy  ปีที่แล้ว

      You're welcome, that 3rd video should be out in the next month or so.

  • @guillermoarvizu1723
    @guillermoarvizu1723 10 หลายเดือนก่อน +1

    Awesome content sir, Any plan to continue uploading more? Cheers from Arizona

    • @greenbeltacademy
      @greenbeltacademy  10 หลายเดือนก่อน

      Yes, most definitely, and I apologize to everyone who's been waiting for the 3rd video! It's in process!

    • @guillermoarvizu1723
      @guillermoarvizu1723 10 หลายเดือนก่อน +1

      @@greenbeltacademy awesome!! I will keep my eyes on it

  • @HomeownerHacks
    @HomeownerHacks ปีที่แล้ว +1

    Andy I passed my Green Belt yesterday. Thanks again for the videos!

    • @greenbeltacademy
      @greenbeltacademy  ปีที่แล้ว

      Oh wow that's awesome!!! Congratulations!!!

  • @CobaReview
    @CobaReview 9 หลายเดือนก่อน +1

    Great video, this material very very help me to make the calculation for the GRR. Thanks..

    • @greenbeltacademy
      @greenbeltacademy  9 หลายเดือนก่อน

      You're absolutely welcome!

    • @CobaReview
      @CobaReview 9 หลายเดือนก่อน

      @@greenbeltacademy I want to ask for the minute 17:32, "Thus when the GRR includes 5 samples then d2 = 2.4812 and for 10 samples then d2 = 3.1791". Could you please give the link for the table for more samples other than 4 samples? Thank you...

    • @CQEAcademy
      @CQEAcademy 8 หลายเดือนก่อน

      @@CobaReviewI don't have a table to share, but if you google it you should be able to find something very quickly

  • @Letter2Mgmt
    @Letter2Mgmt 4 หลายเดือนก่อน +1

    Brilliant video. Thanks so much

  • @helpmepls5698
    @helpmepls5698 ปีที่แล้ว +3

    Absolutely great explanation. Looking forward the third part.

    • @CQEAcademy
      @CQEAcademy ปีที่แล้ว +1

      Thank you!! I"m working on part 3 right now!

    • @kavadapurohith8696
      @kavadapurohith8696 11 หลายเดือนก่อน

      Yes me too!

  • @tumelomrkhudu.1444
    @tumelomrkhudu.1444 ปีที่แล้ว +1

    Watching from South Africa ,Thank you Andy...I'm doing my Certified quality technician. CQT Training course...I find you videos helping me a lot..
    The fact is u make clear example.... hopefully I will pass my CQT exams

    • @greenbeltacademy
      @greenbeltacademy  ปีที่แล้ว

      You're absolutely welcome and I really appreciate the positive comment!!!!!!!!

  • @jjfan485x
    @jjfan485x 9 หลายเดือนก่อน

    Hello! Great videos. Is Part 3 coming soon? Thank you!

  • @ashrafanan8176
    @ashrafanan8176 ปีที่แล้ว +1

    thank you so much , You are amazing, You are doing an amazing efforts

    • @greenbeltacademy
      @greenbeltacademy  ปีที่แล้ว

      You're absolutely welcome and I"m glad to help!!

  • @paulharms1168
    @paulharms1168 9 หลายเดือนก่อน +1

    I just started watching your videos which are very informative. Where can I find your Excel template(s) if they are still available? Thanks

    • @greenbeltacademy
      @greenbeltacademy  9 หลายเดือนก่อน

      You can get them here:
      www.Greenbeltacademy.com/GaugeR&R

    • @paulharms1168
      @paulharms1168 9 หลายเดือนก่อน

      Thanks Andy, much appreciated. ​@@greenbeltacademy

  • @sahilray8458
    @sahilray8458 ปีที่แล้ว

    It was refreshing to watch your video. I wanted to brush up my fundamentals about the Gauge R&R and I watched multiple videos on TH-cam, but your video and explanation was detail, to the point and easy to understand. I am eagerly waiting for part 3 of the series. When are you planning to release?
    P.S. : - I used to work for Alcon in Houston Facility and had an opportunity to talk to you couple of times. It's good to see your here again online. I wish you the best. :D

    • @sahilray8458
      @sahilray8458 ปีที่แล้ว

      Also, I'm subscriber now.

  • @victorramon7852
    @victorramon7852 ปีที่แล้ว +1

    Great, Andy!

  • @mattsheehan843
    @mattsheehan843 11 หลายเดือนก่อน +2

    Thanks so much.

    • @CQEAcademy
      @CQEAcademy 11 หลายเดือนก่อน

      You're welcome!

    • @greenbeltacademy
      @greenbeltacademy  11 หลายเดือนก่อน

      You're welcome!

  • @JohnCelley-xl9lx
    @JohnCelley-xl9lx 9 หลายเดือนก่อน +1

    Thanks for this. What do you do if you can’t collect a sample of parts that span your tolerance. ? Can you just isolate the inspection tool by grabbing the min and the max of the sample parts?

    • @greenbeltacademy
      @greenbeltacademy  9 หลายเดือนก่อน

      Hey John, that'll get you fairly close because as long as you've got the Min/Max parts, then your part range will reflect that process variation.

  • @ahyungrocks5509
    @ahyungrocks5509 7 หลายเดือนก่อน +1

    Thank you for this excellent tutorial!
    I do have one question at 14min 10s where you calculate the reproducibility value. How did you come up values for n * r?

    • @greenbeltacademy
      @greenbeltacademy  7 หลายเดือนก่อน

      When looking up up the d2 factor for reproducibility you stay in the n=1 row, and then your m-value is equal to the number of appraisers.
      So in my case, I had 3 operators, so my m-value = 3, and my d2 factor = 1.912

    • @ahyungrocks5509
      @ahyungrocks5509 7 หลายเดือนก่อน +1

      @@greenbeltacademy
      Sorry, I should be little more elaborate on my question. For the Appraiser Variation (AV) equation, the 2nd half of the equation is (EV^2)/(n*r). How did you come up with the value for n and r? I saw you had 10 for n and 3 for r. Can you explain what do they stand for?

    • @greenbeltacademy
      @greenbeltacademy  7 หลายเดือนก่อน

      @@ahyungrocks5509 Oh gotcha, okay so in that equation n = the number of samples (10), and r is the number of replicate measurements (3).

    • @ahyungrocks5509
      @ahyungrocks5509 7 หลายเดือนก่อน +1

      @@greenbeltacademy thank you

  • @drollette08
    @drollette08 ปีที่แล้ว +1

    Great video

  • @pawandeepsingh8380
    @pawandeepsingh8380 7 หลายเดือนก่อน +2

    At 14:09 why you took value of N equals to 1?
    because n represent sample size and in our case it is equals to 10.

    • @CQEAcademy
      @CQEAcademy 7 หลายเดือนก่อน

      This is a tricky concept to explain, but when we calculator the operator range, we evaluated only 3 different values (Average of A, Average of B and Average of C). So in that calculation we don't consider the 10 samples that were measured. Which is odd I know, and that m-value reflects the number of operators.

    • @pawandeepsingh8380
      @pawandeepsingh8380 7 หลายเดือนก่อน +1

      @@CQEAcademy so in all case we took the value of n equals to 1 while measuring reproducability?

    • @CQEAcademy
      @CQEAcademy 7 หลายเดือนก่อน

      @@pawandeepsingh8380 Yes!

  • @ciaranhahessy1405
    @ciaranhahessy1405 6 วันที่ผ่านมา +1

    When choosing parts or samples should you choose known out of tolerance parts?

    • @greenbeltacademy
      @greenbeltacademy  6 วันที่ผ่านมา

      Yes! I would try to include samples in your experiment that are non-conforming.
      The logic here is that you should want to ensure precision for true failures that you expect to see during routine production.

  • @vijayendraputtakkalavar7666
    @vijayendraputtakkalavar7666 หลายเดือนก่อน +1

    Thank you , super explanation , why 10 parts, why not 30

    • @greenbeltacademy
      @greenbeltacademy  หลายเดือนก่อน +1

      Good question!
      10 parts is sort of the industry standard.
      Outside of that, I’d say that 10 samples is generally enough samples to capture the entire design specification range, while also providing enough measurement values to estimate the repeatability and reproducibility.

  • @RUMAHAQUE-r6g
    @RUMAHAQUE-r6g ปีที่แล้ว

    I have taken samples of 100x on part and another data set of 10x 10 parts.
    For calculating repeatability(R-double bar/d2) ,I'm using d2=3.088 on the 10x 10parts.
    what d2 should I use for the 100x on one part?
    Thanks

  • @frankip3124
    @frankip3124 6 หลายเดือนก่อน +1

    What if R0 is super small where AV turns into a negative number before the sqrt?

    • @greenbeltacademy
      @greenbeltacademy  6 หลายเดือนก่อน

      Hey Frank, that sounds hypothetical, and I've never seen that before. Technically, if the EV is large the operator range will reflect/capture/include that variation (hence the subtraction), and operators tend to create variation in a measurement system, so I've never seen AV turn into a negative number.
      A negative number is impossible, you can't have negative variation.

  • @moosemoss2645
    @moosemoss2645 6 หลายเดือนก่อน +1

    If the measurement itself is a combination of operator decision making and the tool is the analysis still valid?

    • @greenbeltacademy
      @greenbeltacademy  26 วันที่ผ่านมา

      Yes absolutely, and you can quantify the variation that originates from the operators (and their interpretation) in that reproducibility value

  • @abdelrahmanrashad8293
    @abdelrahmanrashad8293 ปีที่แล้ว +1

    Please can you tell me the link for the excell sheet and when will the third video be available ?

    • @greenbeltacademy
      @greenbeltacademy  ปีที่แล้ว +1

      The third video will come out in the next 6-8 weeks!
      Here's the link for the excel spreadsheet: www.Greenbeltacademy.com/GaugeR&R

    • @achouchazeddine7336
      @achouchazeddine7336 ปีที่แล้ว +1

      helo any update on the third video@@greenbeltacademy

    • @greenbeltacademy
      @greenbeltacademy  ปีที่แล้ว +1

      Hey @@achouchazeddine7336 !!! I'm working on a few other commitments right now, but should have it done in the next month or so.

  • @syedmukarram9454
    @syedmukarram9454 ปีที่แล้ว +1

    How come operator C almost always had a lower range value. Could it be that person has more experience/training with the gage.

    • @CQEAcademy
      @CQEAcademy ปีที่แล้ว

      Hey Syed! You're using the right approach when looking at the data! That would be a good hypothesis for that situation, and you could prove that theory by observing the technique of that operator, relative to his/her peers, and evaluating their different levels of training/experience

  • @ahamedthanseeks1440
    @ahamedthanseeks1440 7 หลายเดือนก่อน +1

    In Appraiser Variation formula, what is r ? you have calculated n*r as 10*3. Is r same as m ?

    • @greenbeltacademy
      @greenbeltacademy  7 หลายเดือนก่อน +1

      r is the number of replicate measurements, which in this example I took 3 replicate measurements of each sample.

    • @ahamedthanseeks1440
      @ahamedthanseeks1440 7 หลายเดือนก่อน +1

      @@greenbeltacademy thanks for the reply. Nice video again man. Its really helpful.

    • @greenbeltacademy
      @greenbeltacademy  7 หลายเดือนก่อน +1

      @@ahamedthanseeks1440 You're welcome!

    • @huanlam2346
      @huanlam2346 24 วันที่ผ่านมา

      @@greenbeltacademyhi, can you please explain why instructed to always use n=1, but in formula shows nxr = 10x3? Shouldn’t it be 1x3? Thank you.

    • @greenbeltacademy
      @greenbeltacademy  20 วันที่ผ่านมา

      @@huanlam2346 That n = 1 comment applies for the Pr (Part Range) which requires a different d2 factor than the repeatability calculation, where the d2 factor is evaluated differently.

  • @LeonCui-rt9xb
    @LeonCui-rt9xb 6 หลายเดือนก่อน

    I still cannot understand the AV calculation; why d2=1.912? If the subgroup size=3, the number of combinations of parts and operators (g) = 10 parts and 3 Operators = 10*3=30. Based on the d2 table, we can know that the d2=1.693 instead of 1.192. Can you explain why d2=1.192 in detail in your case study?

  • @nanichowdary397
    @nanichowdary397 10 หลายเดือนก่อน

    Sir, firstly thank you for the videos very informative.
    Question when doing the AV @ 14:15, you said use n=1, but in the video it shows n=10...just checking and want to make sure if this is correct.

    • @greenbeltacademy
      @greenbeltacademy  10 หลายเดือนก่อน

      Great question! okay, so in that equation, when looking up the d2 factor, the "n" value that we use is always equal to 1, which is why in the equation I use 1.912 as the d2 factor.
      However, in that same equation (and I get how confusing this is), when calculating AV, that n value is still 10 because there are 10 samples in the gauge R&R

    • @loganfarrow9811
      @loganfarrow9811 3 หลายเดือนก่อน +1

      @@greenbeltacademy But *why* are we forcing "n" to equal 1 when selecting a d2 value? The table includes options for when multiple samples are used, which on first glance would indicate we use an "n" of 10. Why is it, then, that we only use the "n" value when calculating the AV and *not* for selecting the correct d2 value?

    • @greenbeltacademy
      @greenbeltacademy  3 หลายเดือนก่อน

      ​@@loganfarrow9811 Hey Logan, great question. To be honest, I've never seen an amazing explanation on this approach. I've always understood the sub-group sample size (m) correlating with the number of operators in the experiment because that's the number of measurements that you're including in the range calculation. However, I would assume that the sample size is based on the number of range values included in that calculation, which is simply just 1 since we only calculate the operator range one time.

  • @subasa83
    @subasa83 9 หลายเดือนก่อน

    Thank you for such a nice video! However, there is one question that lingers in my mind after watching this. So basically in Gauge R&R, you want to measure standard deviation (SD), which can be directly computed from your measured data. Thus my question then is, why bother calculating it "indirectly" from the values of range (R) and d2. Thanks!

    • @subasa83
      @subasa83 9 หลายเดือนก่อน +1

      Another question is the meaning of the R double bar. This is used to calculate the inherent variation of WITHIN the instrument. However, R double bar is calculated by taking the average of the range values from ALL OPERATOR and also from DIFFERENT ITEMS. Does this means that the R double bar also carries with it the variance of "operator to operator" and "item to item"? It kind of does not make sense here.

    • @CQEAcademy
      @CQEAcademy 8 หลายเดือนก่อน

      Hey There! Okay, great question about range versus standard deviation. The answer to this question (I believe) is somewhat historical. Years ago, when the GR&R method was first developed, calculations were done by hand, and range values are easier to calculate than standard deviation. Hence the method today.
      Now that we have minitab, ANOVA has become a much more popular method for evaluating GR&R, and the ANOVA method does calculate standard deviation directly instead of the range.

    • @CQEAcademy
      @CQEAcademy 8 หลายเดือนก่อน

      @@subasa83No, R-double bar doesn't not include the variation between operators because that variation blocks out any contribution from the operator or the parts.
      Remember, those range values that we use to calculate R-double bar, is ONLY the variation within a single operator, within a single sample. So any variation that is expressed into those range values comes from the measurement equipment itself.

    • @subasa83
      @subasa83 8 หลายเดือนก่อน +1

      @@CQEAcademy Tks a lot for your spot-on answer. How about my first question? Why don't we calculate SD directly from the data?

    • @greenbeltacademy
      @greenbeltacademy  8 หลายเดือนก่อน

      @@subasa83 hmm that's weird, I thought I had already replied to that question.
      So honestly the answer has to do with history.
      Historically, Quality Engineerings made these calculations by hand, and the range value is so much easier to calculate by hand than the standard deviation.
      Now that we have software like minitab, we use ANOVA which calculates the SD instead of the range value.
      Hopefully that makes sense!

  • @LeonCui-rt9xb
    @LeonCui-rt9xb 6 หลายเดือนก่อน

    Why n=10 but d2=1.912? If n=10 in your calculation, does that mean d2=1.716?

    • @CQEAcademy
      @CQEAcademy 6 หลายเดือนก่อน

      Great question - and that's a tricky part of the whole process, so in that reproducibility calculation we use n = 10 because that sample size is used as an adjustment factor to essentially eliminate EV from the AV calculation.
      However when using the table to look up the d2 factor for AV you use a different perspective.

  • @BodaoSBC
    @BodaoSBC ปีที่แล้ว

    What about video 3 guys 😢?

    • @greenbeltacademy
      @greenbeltacademy  ปีที่แล้ว +1

      I'm working on it now, should be done in the next 4 weeks

  • @egwinharianto2851
    @egwinharianto2851 11 หลายเดือนก่อน

    you miss explaining n and r

    • @greenbeltacademy
      @greenbeltacademy  11 หลายเดือนก่อน

      Hey There! Which N and R value are you referring too?

    • @egwinharianto2851
      @egwinharianto2851 11 หลายเดือนก่อน

      @@greenbeltacademy thank you for the reply , at 13:54

    • @greenbeltacademy
      @greenbeltacademy  11 หลายเดือนก่อน

      Thanks @@egwinharianto2851 !
      Okay, so that n is the number of samples used in the Gauge R&R (10 in this example), and r is the number of replicate measurements (3 in this example).

  • @chikamutiaraf
    @chikamutiaraf หลายเดือนก่อน

    thank you for the videooosss 🥹🥹🥹 help me a lot