iSixSigma

The MM Experiment

Six Sigma – iSixSigma Forums Old Forums General The MM Experiment

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
    Posts
  • #30644

    billybob
    Participant

    I have read numerous posting how M&Ms are used for statistical training and so on.  Now that I know about 6S everytime I see a bag of M&Ms in a vending machine I really know they are full of defects that make me, the customer, unhappy so I choose another product of higher-quality to snack on.  Man can’t live by taste alone, we want and demand quality junk food!
    Does anyone know if the M&M company will be applying 6S to reduce the number of defects in their product?  Or are they happy selling defects to people are willing to pay for them?
    Later,
    Billybob

    0
    #80057

    Mikel
    Member

    Billybob,
    Those who use M&M’s in their training as examples of defects probably do not really understand the customer’s expectations.
    Go ahead and eat them with pride, but don’t forget about a balananced diet. The basic grease group is just as important as the basic chocolate group.

    0
    #80059

    Robert Butler
    Participant

      If your post is in earnest and you aren’t just having some Monday fun I don’t think the object of using M&M’s in statistics training is to emphasize defect detection.  All of the exercises that I have taught use M&M’s to illustrate the concepts of distribution and variation.  Properly done, the use of M&M’s can really help people understand the issues that arise when dealing with data.
      As I’ve mentioned on other threads, the M&M experiment is so clear and concise that you can use it to teach 3rd and 4th grade students advanced statistical concepts.  So the next time you look at a Plain M&M packet don’t think “defect” think “distribution”.

    0
    #80074

    Mike Carnell
    Participant

    Robert,
    I haven’t met a lot of 3rd and 4 th graders that have any use for advanced statistical concepts. I have met a lot of waves of training which do need to do a little inclass application of attribute gage R&R. In that application with basic criteria of colorful and clearly marked that anyone can supposedly understand allows people to do an actual exercise in class that doesn’t have the class screaming that it is another manufacturing example.
    I am sure you are correct that it can be used to teach statistics. Does that really mean if you are using them for another demonstration you are doing something wrong? I don’t think so.
    Cayapults are great for DOE. Does that mean I am misusing them when we do hypothesis testing and capability analysis. I don’t think so.

    0
    #80088

    Robert Butler
    Participant

    Mike,
      I’m left with the impression that we are talking past each other.  Billybob indicated that his experience with M&M’s in training exercises dealt with defects-I’m assuming defects/bag.  All I said was that the only experiments involving M&M’s of which I was aware focused on helping people understand the concepts of distribution, variation, standard deviation, etc.
      The reference to 3rd and 4th grade students was not meant to denigrate, merely to emphasize the fact that by using something such as the M&M experiment you could make these concepts understandable to almost anyone.
      What I do find interesting is that apparently you and Billybob are aware of yet another experiment involving M&M’s that focuses on defects and introduces the concept of attribute gage R&R.  If this has been written up could you provide a citation?  If it hasn’t could you tell me how it is done?
      By way of recirpocity, if you are interested in fine tuning your teaching techniques with respect to using M&M’s, check the past issues of the American Statistical Association’s STN (Statistics Teachers Network)  newsletter.  There is a four part article “Some Students and a t-test or Two” that may give you some ideas for using M&M’s for the purposes mentioned in the first paragraph.

    0
    #80091

    billybob
    Participant

    Hello,
    Our M&M experiment was to find defects from a control sample.  Was the coating color uniform? Any chips in the coating was a defect, parting lines were defects.  Was the M&M logo present and clear? 
    But we didn’t work out of small individual bags, just one big family size so the total population was in question on that manufacturing day as the bag was filled at the factory.  We only had one big-bag to work from and not many small samples to predict any expected range of colors.  Thanks for opening my eyes that the training we were given was flawed, I wonder what that bad training cost my company?
    It would have been more interesting walking out to the parking lot and checking cars for damage, or having 4 of the same brand tires, measuring tread depth, or even doing tire pressure tests.
    But I stand by the statement that we shouldn’t expect high quality M&Ms from them.  We are purchasing defects that don’t arrive to us at the same quality level as they left the manufacturer.  If the coating is broken then they will “melt in my hand..hence A DEFECT.”  Maybe thats the real lesson you consultants should be teaching?
    After the possums over heard me telling Ma about this they won’t even eat M&Ms now.  Maybe I can pawn what I have left on the kids Trick-or-Treating Thursday night.
    Later,
    Billybob

    0
    #80094

    Joy Cowling
    Participant

    Robert,
    I’ve used M&M’s for attribute R&R and basic stats (histograms, run charts, paretos, etc).  I’d like more info on using them for hypothesis testing.  Do you have a link to the newsletter that you mentioned (if it is online)?
    For the R&R, give each team 20 M&M’s.  Help them to set up their worksheets in Minitab for two or three inspectors, two trials each.  Give them the criteria (vague is usually good for this exercise) for a defect.  We have them be responsible for material handling, numbering their samples, etc.  When they are done with their inspections, have an instructor inspect them as the “standard”.  Then do the analysis in Minitab.  They usually get a terrible result.  The message is that for attribute decisions we often give operators no more information than that.  We give them vague standards and poor instructions.  In debrief, talk about what would have made the results better (visual standards, better description, more training) etc.
    If you want more info, email me at [email protected].  I’ll be happy to help.
    Joy

    0
    #80095

    Billybobismyidol
    Participant

    I can appreciate your humor and the point you are making. For the others – always try to leave at least a minute a day when you don’t take your self so seriously.  Chill out and enjoy life!!!!!!!!!!!

    0
Viewing 8 posts - 1 through 8 (of 8 total)

The forum ‘General’ is closed to new topics and replies.