iSixSigma

Sampling Question

Six Sigma – iSixSigma Forums Old Forums General Sampling Question

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
    Posts
  • #51786

    Tim
    Member

    I was hoping to get some help in coming up with an effective sampling plan. We have rail cars that carry resin ( in the form of pellets) measured in pounds. Each rail car has 4 cells, each cell containing 50,000 pounds of material. What would be the best sampling plan to use to get a good representation of the population?Thanks in advance

    0
    #180664

    Remi
    Participant

    Tim, there is no formula for the perfect answer (as far as I know).
    You have to take into account the various forms of variation sources you have and make an acceptable choice (for you) based on that.
    And what do you consider to be your population? All pellets produced by AAA during time BBB from location CCC under conditions DDD (etc)
    Variation sources to consider (not complete list!)

    Measurement system
    Between Pellets (size or weight or concentration or …)
    Location/Position in Cell
    Between Cells
    Between Rail cars
    between Batches/Shifts/Days/Weeks/Months
    …?
    Good luck

    0
    #180673

    GB
    Participant

    Confidence level: 95%
    Confidence Interval: 5%
    Pop(n): 50,000 per pallet
    Sample Size: 381 per pallet
    -OR-
    Confidence Level: 95%
    Confidence Interval: 5%
    Pop(n): 200,000 per rail car
    Sample size: 383 per rail car

    0
    #180676

    SandMan
    Member

    Hey HB,
    can you go through how you got those numbers…?
    Did you use the old Z1.4 tables?
    Thanks.

    0
    #180679

    Mario Perez-Wilson
    Participant

    Providing the samples (pellets) an equal probability of being selected!Mario Perez-Wilson

    0
    #180680

    GB
    Participant

    Nist-Sematech is your friend!
    http://www.itl.nist.gov/div898/handbook/index.htm

    0
    #180683

    SandMan
    Member

    :::bowing repeatedly:::

    0
    #180684

    Taylor
    Participant

    Tim
    This is much like sampling grain composites. The sampling equipment we use for grain takes 3 samples from varing locations in truck. The sample bar collects grain from top to bottom with in the feed. Given you have 4 cells in a rail car I would suggest a sample from each cell that would be representive of top to bottom . Dickey John makes great equipment for this. Because the samples come from such a distributed and random location in the car, the sample size needed is much less than computed.
    ย 

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

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