Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 55 views

    MGT 6421 Quality Management II

    Uploaded by

    MustakiMipa Regresi
    This document discusses methods for measuring the effectiveness of control charts, including operating characteristic (OC) curves and average run length (ARL). The OC curve shows the probability of observations falling within control limits for different process means. ARL is the average number of points plotted until an out-of-control signal and is related to the OC curve. The document provides examples of calculating OC curves and ARL for X-charts and attributes control charts to evaluate control chart performance.

    Copyright:

    © All Rights Reserved
    Download as DOC, PDF, TXT or read online from Scribd

    MGT 6421 Quality Management II

    Uploaded by

    MustakiMipa Regresi
    55 views8 pages
    This document discusses methods for measuring the effectiveness of control charts, including operating characteristic (OC) curves and average run length (ARL). The OC curve shows the probability of observations falling within control limits for different process means. ARL is the average number of points plotted until an out-of-control signal and is related to the OC curve. The document provides examples of calculating OC curves and ARL for X-charts and attributes control charts to evaluate control chart performance.

    Original Description:

    statistical quality control

    Original Title

    Performance

    Copyright

    © © All Rights Reserved

    Available Formats

    DOC, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses methods for measuring the effectiveness of control charts, including operating characteristic (OC) curves and average run length (ARL). The OC curve shows the probability of observations falling within control limits for different process means. ARL is the average number of points plotted until an out-of-control signal and is related to the OC curve. The document provides examples of calculating OC curves and ARL for X-charts and attributes control charts to evaluate control chart performance.

    Copyright:

    © All Rights Reserved
    Download as DOC, PDF, TXT or read online from Scribd
    Download as doc, pdf, or txt
    55 views8 pages

    MGT 6421 Quality Management II

    Uploaded by

    MustakiMipa Regresi
    This document discusses methods for measuring the effectiveness of control charts, including operating characteristic (OC) curves and average run length (ARL). The OC curve shows the probability of observations falling within control limits for different process means. ARL is the average number of points plotted until an out-of-control signal and is related to the OC curve. The document provides examples of calculating OC curves and ARL for X-charts and attributes control charts to evaluate control chart performance.

    Copyright:

    © All Rights Reserved
    Download as DOC, PDF, TXT or read online from Scribd
    Download as doc, pdf, or txt
    of 8

    MGT6421

    QualityManagementII

    ControlChartPerformance

    Whenselectingcontrolcharts,weshouldbeinterestedinhoweffectivethey
    are.Ourobjectivesfortodayaretoknowhowtomeasuretheeffectiveness
    ofthechartswehavediscussed.

    Therearetwocommonlyusedmethodsforcomparingcontrolchart
    performance.Thefirstmethodistodeterminethechartsoperating
    characteristic(OC)curve.Thesecondistodeterminetheaveragerun
    length(ARL).
    TheOCCurve:
    TheOCCurveshowstheprobabilitythatanobservationwillfallwithinthe
    controllimitsgiventhestateoftheprocess.Thisisverymuchlikefinding
    powercurvesinhypothesistesting.Forexample,forthe X chart,wecan
    findtheprobabilityofbeingwithinthelimits(Pa)forparticularvaluesof
    theprocessmean:
    Pa P( LCL X UCL| ) .

    Implementation and Performance - 1

    MGT6421

    QualityManagementII

    ThecurveisgeneratedbycalculatingPafordifferentvaluesof.Below
    wecalculatetheoperatingcharacteristiccurveforthe X chartwhenthe

    standardsareknown.Inthiscasethecontrollimitsare 0 3
    .Wewill
    n
    letthemeanshiftinincrementsofthestandarddeviationof X (i.e.,

    0
    ).Ifthisisthecase,then
    n

    Pa P 0 3
    X 0 3

    n
    n

    0 3

    Pa P

    ,so





    0
    0 3 0
    n
    n X
    n
    n

    n
    n
    n

    or
    Pa=P(3Z3).

    Example:

    Implementation and Performance - 2

    MGT6421

    QualityManagementII

    ThetablebelowshowssomevaluesofPafordifferentvaluesof.TheOC
    curveisalsodrawn.

    0 2
    n

    0 3
    n
    0 1

    Pa
    0.9973
    0.9772
    0.8413
    0.5000

    1.0000
    0.9000
    0.8000
    0.7000
    0.6000

    Pa 0.5000
    0.4000
    0.3000
    0.2000
    0.1000
    0.0000
    0

    Implementation and Performance - 3

    MGT6421

    QualityManagementII

    TheRunLength

    AnothermeasureofperformancethatiscloselyrelatedtoOCcurvevalues
    istherunlength.Therunlengthisarandomvariableandisdefinedasthe
    numberofpointsplottedonthechartuntilanoutofcontrolconditionis
    signaled.Thebeginningpointatwhichwecountthenumberofplotted
    pointsdependsonwhetherwearefindingtheincontrolrunlengthorthe
    outofcontrolrunlength.

    Theincontrolrunlengthmeasuresthenumberofplottedpointsfromthe
    beginningofthemonitoringperioduntilanoutofcontrolsignal,giventhat
    therehavebeennochangesintheprocess.Wewanttheaverageincontrol
    runlengthtobehigh.

    Theoutofcontrolrunlengthmeasuresthenumberofplottedpointsfrom
    thetimeofaprocesschangeuntilanoutofcontrolsignalisgiven.Its
    valuedependsuponthesizeoftheshift.Wewanttheaverageoutof
    controlrunlengthtobesmall.

    Assumingthatthestatisticbeingplottedisindependentovertime(i.e.,one
    plottedpointisindependentofotherplottedpoints),therunlengthfollows
    ageometricdistribution.

    Implementation and Performance - 4

    MGT6421

    QualityManagementII

    TheAverageRunLength

    Theaveragerunlength(ARL)istheaveragenumberofpointsplottedon
    thechartuntilanoutofcontrolconditionissignaled.Itistheexpected
    valueoftherunlengthdistribution.ItisrelatedtotheOCcurveasfollows:
    ARL

    1
    .
    1 Pa

    Forourexample:

    Implementation and Performance - 5

    MGT6421

    QualityManagementII

    TheARLisprobablyusedmoreoftenincontrolchartliteraturethanOC
    curvevalues,becauseitisconceptuallyeasiertounderstand.Thetable
    below showstheARLvaluesforthe X chart,bothwithandwithoutrun
    rules*.

    Shift
    0
    1
    2
    3
    * Rule

    1:
    Rule 2:
    Rule 3:
    Rule 4:

    Rule 1
    370.4
    43.89
    6.3
    2

    Rules 1 & 2
    225.44
    20.01
    3.65
    1.68

    Rules 1, 2 & 3
    132.89
    10.95
    3.14
    1.67

    Rules 1, 2, 3 & 4
    91.75
    9.22
    3.13
    1.67

    One observation outside the control limits


    Two of three outside of 2 standard deviatioins
    Four of five outside of 1 standard deviatioin
    Eight in a row on one side of the center line.

    Example:

    The source for these numbers is Champ, W.C., and Woodall, W. H. (1987).
    "Exact Results for Shewhart Control Charts With Supplementary Runs Rules,"
    Technometrics, Vol. 29, No. 4, 393-399.

    Implementation and Performance - 6

    MGT6421

    QualityManagementII

    ARL with Attribute Control Charts

    WecanalsoapplytheARLideatoattributecontrolcharts.Theprinciples
    areallthesame.Allthatchangesaretheprobabilitycomputations.

    Examples:

    Implementation and Performance - 7

    MGT6421

    QualityManagementII

    ProbabilityLimits

    WecanalsousetheideaofOCcurveswithattributecontrolchartsto
    designlimitsthatbettermeettheassumptionsofthestatisticbeingplotted.
    Inotherwords,thechartsthatwehavediscussedtothispointalluse3
    sigmalimits.Butifthebinomialorpoissondistributionsarenotclosetoa
    normaldistribution,theprobabilityofatypeIerrormaybefarfrom.0027.
    HencewedesiretodesigncontrolchartsthathavesmallertypeIerrors(or
    alternatively,thathavebetterincontrolARLvalues).

    Examples:

    Implementation and Performance - 8

    You might also like