iSixSigma

Ppk

Viewing 15 posts - 1 through 15 (of 15 total)
  • Author
    Posts
  • #48916

    speed
    Member

    Can any one tell me the sigma multiple value,
    Case I : if Ppk= – 0.10
    Case II: if Ppk is a negative number.
     

    0
    #166157

    Taylor
    Participant

    cannot be negative, impossible

    0
    #166159

    melvin
    Participant

    I:  for most purposes  3*Ppk = Z.lt   so your Z.lt is roughly -0.30 sigma
    II: negative numbers indicate that your mean is outside of your spec limits

    0
    #166160

    Lolita
    Participant

    Yes
    Correct

    0
    #166161

    Lolita
    Participant

    wrong

    0
    #166162

    Haugen
    Participant

    Question – How/when/why do you decide to change identities? Same question regarding when do you create a whole new identity, vs switching back and forth between the current identities you are using? You seem to enjoy creating and using new identities, but do not put any energy at all into masking the posting traits that make you so easily identifiable.Just curious…….

    0
    #166163

    Lolita
    Participant

    It  could  be  negative

    0
    #166165

    Brandon
    Participant

    Jim – I’ve noticed it happening when we’ve out’d someone. That is, when they post either inaccurate, off-base, contradictory or whatever other off-base posts.
    They seem to have to abandon that identity and attempt to join as someone else. But, like you say…the written word is nearly as identifiable as a voice or a face.

    0
    #166167

    BTDT
    Participant

    Brandon:I posted for a long time under the nom de guerre, BTDT, mostly to be assured that my advice was only given the weight of its own merit, and to avoid the misconception that I was only doing so to promote myself. I admit that I would sometimes post under a different name when I thought I was being particularly witty.After a while most people found out who I was, so now it’s just me.Cheers, Alastair Muir

    0
    #166168

    Taylor
    Participant

    OR when some idiot post under someone else’s username like happened to me yesterday. see above post on the answer. Obviously not me because I would simply replied to this question “Looks like homework to me”. But I dont plan on changing my username just yet.
    I do find it ironic that someone would have a username “Lolita”

    0
    #166169

    BTDT
    Participant

    Chad:… probably just Jeremy Irons, being ironic and all.Cheers, Alastair

    0
    #166170

    Brandon
    Participant

    BT & Chad – I agree. I use only Brandon – whether my posts are any good or not. What did Popeye say – “i yam what I yam!”… or something like that.
    And, yes, a few eeks ago someone was hassling me by posting under Brandon – you likely recall. I called in the iSS Police and since it hasn’t happened in quite a while I presume they intervened. We can’t fake our IP Addresses!!

    0
    #166206

    Dixit
    Member

    tell me about Ppk
     

    0
    #166215

    Taylor
    Participant

    Vijay
    PpK & Pp are measures calculated using the long term process standard deviation. Since long term process variation is used, these measures take into account subgroup drift in the data as well as, the within subgroup variation. Ppk is a ratio of the distance between the process average and the closest specification limit. Like Cpk, Ppk takes into the centering of the process long term, and is a good indicator of what the customer is seeing in relation to process performance. Pp is a ratio of customer specified tolerance to six standard deviations, and gives you a long term process variation indicator without regard to process centering, otherwise how good the process could be if centered.
     

    0
    #167253

    Tim Behr
    Member

    Nice explanation, Chad. I like to say “overall” versus “long term”. I often calculate both CpK and PpK for a single batch of material if the data is collected in subgroups or if I use a moving average for the CpK calculation (which I normally don’t do). PpK is generally easier to calculate because it’s easy to calculate the overall standard deviation of a data set using Excel. It’s more difficult to calculate a pooled standard deviation for individual subgroups or batches. Minitab and other stats programs can do this easily but you need to define the smaller group, either by subgroup size (consecutive parts), batch size, or moving average size.PpK is always the lowest of the four capability indices: Cp, CpK, Pp, PpK (unless you’re using a moving average for CpK and you’re bouncing back and forth) so if you’re just looking to see if you’re making parts that are out of spec, start there. If the PpK looks low if if you’re trying to identify the source of variation, start looking at the others. A significantly higher Pp indicates you have a centering problem. A significantly higher CpK indicates you have a drifting or unstable process.

    0
Viewing 15 posts - 1 through 15 (of 15 total)

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