Quality Exclusives

The Performance Curve of a Device

May 20, 2011
/ Print / Reprints /
ShareMore
/ Text Size+

A little-know yet important subject, the performance curve of a device is a concept that allows us to evaluate a device’s behavior in several points along the engineering specification range. Through the results of this field of study, it is possible to determine the probability of acceptance of each measured part of the studied device.

This study will be explained from the example below.

  • Bias = - 0,1. The bias is calculated by the difference between the reference value and the average measured value observed.

  • RR = 0,25 (25%). As 5,15 sigma means that 99% of the measurements are within the normal distribution, we can calculate sigma, where:

    5,15 sigma = RR Þ 5,15 = 0,25 Þ sigma = 0,04854

  • Engineering limits: ILE = 80 and SLE = 82

  • Reference values: 79,95; 81; 82; 82,04; and 82,12 (random values, although, near the engineering specification).

    Example of hom to create the performance curve of a device:



  • First Device Performance Curve Point (79,95)

    X = 79,95 – 0,1 = 79,85

    Z = (ILE – X) / sigma = (80 – 79,85) / 0,04854 Þ Z = 3,09 Þ P(Z) = 0,10%

    Z = (SLE – X) / sigma = (82 – 79,85) / 0,04854 Þ Z = 44,29 Þ P(Z) = 0,00%

    ps.: P(Z) = 0.10% is extracted from the normal distribution table for Z = 3,09



    Second Device Performance Curve Point (81)

    X = 81 – 0,1 = 80,9

    Z = (X – ILE) / sigma = (80,9 – 80) / 0,04854 Þ Z = 18,54 Þ P(Z) = 0,00%

    Z = (SLE – X) / sigma = (82 – 80,9) / 0,04854 Þ Z = 22,66 Þ P(Z) = 0,00%



    Third Device Performance Curve Point (82)

    X = 82 – 0,1 = 81,9

    Z = (X – ILE) / sigma = (81,9 – 80) / 0,04854 Þ Z = 39,14 Þ P(Z) = 0,00%

    Z = (SLE – X) / sigma = (82 – 81,9) / 0,04854 Þ Z = 2,06 Þ P(Z) = 1,97%



    Fourth Device Performance Curve Point (82,04)

    X = 82,04 – 0,1 = 81,94

    Z = (X – ILE) / sigma = (81,94 – 80) / 0,04854 Þ Z = 39,96 Þ P(Z) = 0,00%

    Z = (SLE – X) / sigma = (82 – 81,94) / 0,04854 Þ Z = 1,24 Þ P(Z) = 10,82%



    Fifth Device Performance Curve Point (82,12)

    X = 82,12 – 0,1 = 82,02

    Z = (X – ILE) / sigma = (82,02 – 80) / 0,04854 Þ Z = 41,61 Þ P(Z) = 0,00%

    Z = (X - SLE) / sigma = (82,02 – 82) / 0,04854 Þ Z = 0,41 Þ P(Z) = 34,02%



    After accomplishing all the calculations and tracing the curve, for the resulting value of each part, the mode graph shall be used in order to know the probability of acceptance for each related part. In the X axis of the graph, the value found for the part that is being analysed has to be entered and an imaginary line has to be created an then, from the found value, until this imaginary line reaches the graph and, at this moment, it is read in the Y”axis the probability of acceptance.

    The technician responsible for the analysis ha to evaluate the importance of the characteristic to be controlled to define whether the study should be accomplished or not, due to its complexity.

    The performance curve of a device’s objective is solely and exclusively to evaluate the probability of acceptance of each measured part. This study allows us a higher safety margin in decisions such as: to liberate or scrap a batch of manufactured parts as well as to invest in financial resources to improve the RR values or even to reduce the device’s tendency value. The smaller the RR values and the bias are the higher is the probability of acceptance for the values near the engineering limits.

    Did you enjoy this article? Click here to subscribe to Quality Magazine. 

    You must login or register in order to post a comment.

    Multimedia

    Videos

    Podcasts

     In honor of World Quality Month, we spoke to James Rooney, ASQ Past Chairman of the Board of Directors 2013, for his take on quality around the world.
    For more information, read the ASQ Speaking of Quality column.
    More Podcasts

    Quality Magazine

    CoverImage

    2014 September

    Check out the September 2014 edition of Quality Magazine for features!

    Table Of Contents Subscribe

    The Skills Gap

    What is the key to solving the so-called skills gap in the quality industry?
    View Results Poll Archive

    Clear Seas Research

    qcast_ClearSeas_logo.gifWith access to over one million professionals and more than 60 industry-specific publications,Clear Seas Research offers relevant insights from those who know your industry best. Let us customize a market research solution that exceeds your marketing goals.

    eNewsletters

    STAY CONNECTED

    facebook_40.png twitter_40px.png  youtube_40px.pnglinkedin_40px.png