attribute gage rr ridgway

Upload: pradeep-joshi

Post on 03-Jun-2018

226 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/12/2019 Attribute Gage RR Ridgway

    1/50

    Attribute Gage R&R

    An Overview

    Presented to ASQ Section 1302August 18, 2011

    by Jon Ridgway

  • 8/12/2019 Attribute Gage RR Ridgway

    2/50

    Overview

    What is an Attribute Gage R&R?

    Why is it worth the work?

    What are the caveats?

    How do I perform it? How do I understand the results?

  • 8/12/2019 Attribute Gage RR Ridgway

    3/50

    My Attribute Gage R&R Experience

    First Data

    Credit Card mailings (packages)

    Airlite Plastics:

    o r nt ua ty Print Quality, Cup & Lid Decoration

  • 8/12/2019 Attribute Gage RR Ridgway

    4/50

    WHY conduct a Gage R&R????

    Some processes require subjective decision

    making: Inspection

    Validation

    Subjectivity creates the potential for variation Measurement System variation impacts process

    capability:

    Type I Errors

    Type II Errors

  • 8/12/2019 Attribute Gage RR Ridgway

    5/50

    So, Why Conduct a Gage R&R?

    To understand:

    How likely Appraiser will agree with himself / herself: WITHIN / Repeatability

    How likely all Appraisers will agree with each other:

    BETWEEN / Reproducibility Understanding R&R allows you to:

    Predict probability (%) of agreement / disagreement

    Implement training to improve that probability Reduce Type I and Type II Errors = $$$

  • 8/12/2019 Attribute Gage RR Ridgway

    6/50

    MSA

    The Foundation of everything in Quality is

    measurement

    Measure for two primary reasons:

    o ma e a ec s on As the basis for process improvement

    Can we trust our measurement system to

    give us reliable data? CONFIDENCE

  • 8/12/2019 Attribute Gage RR Ridgway

    7/50

    Ultimate purpose of the AttributeAgreement Analysis

    To determine if your measurement system

    can distinguish between a good & bad part

    Accuracy & Precision:

    ccuracy: Absence of bias, or agreeing with the standard.

    Precision:

    Ability of different Appraisers to reach the sameconclusion several times.

  • 8/12/2019 Attribute Gage RR Ridgway

    8/50

    Accurate, But Not Precise

  • 8/12/2019 Attribute Gage RR Ridgway

    9/50

    Precise, But Not Accurate

  • 8/12/2019 Attribute Gage RR Ridgway

    10/50

    Which is Easier to Remedy?

    Accurate, but not Precise Precise, but not Accurate

  • 8/12/2019 Attribute Gage RR Ridgway

    11/50

    Gage R&R Review

    Measurement System Analysis (MSA)

    1st R: Repeatability

    2nd R: Reproducibility

    a a n enera : Continuous / Variables

    Attribute / Discrete

  • 8/12/2019 Attribute Gage RR Ridgway

    12/50

    Attribute vs. Continuous

    Attribute Data:

    Categorical, named only, arbitrary scalesAlso known as Discrete Data

    Continuous Data:

    Allows for infinitely finer sub-divisions

    Also known as Variables Data

  • 8/12/2019 Attribute Gage RR Ridgway

    13/50

    Basic Data Types

    Nominal: Literally, name

    Represents categories Ordinal:

    Orderedor ranked data

    Interval: Measured / scaled data: Each position equidistant

    0 can be relevant (temperature)

    Ratio: Numbers compared as multiples of one another

  • 8/12/2019 Attribute Gage RR Ridgway

    14/50

    Hierarchy of Data TypesNominal Ordinal Interval Ratio

    Classified Data Quantified Data

    DISCRETE / ATTRIBUTE CONTINUOUS / VARIABLE

    Non-parametric Parametric

  • 8/12/2019 Attribute Gage RR Ridgway

    15/50

    2 Main Attribute Gage R&R Types1) Binary / Nominal

    GO / NO GO Data are Categorical and mutually exclusive

    Kappa statistic is relevant

    2) Ordinal Rank, not categorical

    Data are not mutually exclusive

    Kendalls statistic more relevant than Kappa

  • 8/12/2019 Attribute Gage RR Ridgway

    16/50

    Kappa Statistic Proportion of agreement between

    evaluators after chance agreement hasbeen removed:

    Kappa = P observed P chance/ P chance

    Expressed as a number: From 0 (expected by chance)

    Up to +1 (complete agreement)

  • 8/12/2019 Attribute Gage RR Ridgway

    17/50

    Kendalls Statistics

    Two different Kendalls for different tests:

    Kendalls Coefficient of Concordance: Rankings without a known Standard

    Rankings with a known Standard

    Expressed as 0 (weaker agreement) to +1(stronger agreement)

  • 8/12/2019 Attribute Gage RR Ridgway

    18/50

    Kappa & Kendalls Summary

    Kappa:

    Nominal / Binary Only

    Match or No Match

    en a s oe c en o oncor ance:Ordinal but not using a known Standard

    Kendalls Correlation Coefficient:

    Ordinal and using a known Standard

  • 8/12/2019 Attribute Gage RR Ridgway

    19/50

    Attribute Gage R&R Considerations

    Study Purpose

    Destructiveness

    Precision vs. Time

    Binomial / Nominal vs. Ordinal

  • 8/12/2019 Attribute Gage RR Ridgway

    20/50

    MSA Factors Impacting Variation

    Gage

    Appraiser

    Method

    Part Environment

  • 8/12/2019 Attribute Gage RR Ridgway

    21/50

    Controlling MSA Factors

    Ideally:

    1. Use the same Assessment Method

    2. Require all Appraisers to assess the same

    3. Conduct the study under the normalassessment conditions

  • 8/12/2019 Attribute Gage RR Ridgway

    22/50

    Controlling MSA Factors, Cont.

    1) Appraisers: Select from group that

    normally appraises the part.2) Number of parts should cover the entire

    .

    3) More than one appraisal per Appraisershould be done.

    4) The presentation of the samples withinthe Trial should be randomized.

  • 8/12/2019 Attribute Gage RR Ridgway

    23/50

    Nominal / Binary Study Two Appraisers

    50 Parts

    2 Trials

  • 8/12/2019 Attribute Gage RR Ridgway

    24/50

  • 8/12/2019 Attribute Gage RR Ridgway

    25/50

  • 8/12/2019 Attribute Gage RR Ridgway

    26/50

    Output: Within

  • 8/12/2019 Attribute Gage RR Ridgway

    27/50

    Output: Within vs. Standard

  • 8/12/2019 Attribute Gage RR Ridgway

    28/50

    100

    95

    95.0% C IPercent

    100

    95

    95.0% C IPercent

    Date of study:

    Reported by:Name of product:

    Misc:

    Assessment Agreement

    Within Appraisers Appraiser vs Standard

    FredLee

    90

    85

    80

    Appraiser

    Percen

    t

    FredLee

    90

    85

    80

    Appraiser

    Percen

    t

  • 8/12/2019 Attribute Gage RR Ridgway

    29/50

    Output: Between

  • 8/12/2019 Attribute Gage RR Ridgway

    30/50

    Output: Between vs. Standard

  • 8/12/2019 Attribute Gage RR Ridgway

    31/50

    Ordinal Case Study: Print Quality

  • 8/12/2019 Attribute Gage RR Ridgway

    32/50

    What did we want to know?

    Do all Appraisers of Print Quality:

    Agree consistently with Themselves?

    Agree consistently with Each Other?

    ,

    Accept

    Accept but Adjust

    Reject

  • 8/12/2019 Attribute Gage RR Ridgway

    33/50

    How was it Done?

    10 samples, Good & Bad

    Random Order, Same for All

    2 Trials per person

    All people in the study Environment

  • 8/12/2019 Attribute Gage RR Ridgway

    34/50

    Ensure Gage R&R Consistency

  • 8/12/2019 Attribute Gage RR Ridgway

    35/50

    Spanish Version

  • 8/12/2019 Attribute Gage RR Ridgway

    36/50

    Vietnamese Version

  • 8/12/2019 Attribute Gage RR Ridgway

    37/50

    Trial Order

    My Checklist

  • 8/12/2019 Attribute Gage RR Ridgway

    38/50

  • 8/12/2019 Attribute Gage RR Ridgway

    39/50

    ResultsSample QA1-1 QA1-2 QA2-1 QA2-2 QA4-1 QA4-2 Standard

    1 1 1 2 2 2 2 2

    2 2 2 2 2 3 3 3

    3 1 1 2 2 2 2 2

    5 2 1 3 3 2 3 2

    6 1 1 2 2 2 2 2

    7 2 1 3 3 3 3 3

    8 2 2 2 2 2 2 3

    9 1 1 2 2 3 3 1

    10 1 1 1 1 2 2 1

  • 8/12/2019 Attribute Gage RR Ridgway

    40/50

    False Alarms & Misses Assess Fail when Standard = Pass:

    False Alarm Type I Error

    ssess ass w en an ar = a :

    Miss

    Type II Error

  • 8/12/2019 Attribute Gage RR Ridgway

    41/50

    ResultsSample QA1-1 QA1-2 QA2-1 QA2-2 QA4-1 QA4-2 Standard1 1 1 2 2 2 2 2

    2 2 2 2 2 3 3 33 1 1 2 2 2 2 2

    4 2 2 3 3 2 2 3

    False

    Alarms

    5 2 1 3 3 2 3 26 1 1 2 2 2 2 2

    7 2 1 3 3 3 3 3

    8 2 2 2 2 2 2 39 1 1 2 2 3 3 1

    10 1 1 1 1 2 2 1

  • 8/12/2019 Attribute Gage RR Ridgway

    42/50

    Minitab 15 Four Results:

    1. Within2. Within vs. Standard

    .

    4. Between vs. Standard

  • 8/12/2019 Attribute Gage RR Ridgway

    43/50

    Check Here

  • 8/12/2019 Attribute Gage RR Ridgway

    44/50

  • 8/12/2019 Attribute Gage RR Ridgway

    45/50

  • 8/12/2019 Attribute Gage RR Ridgway

    46/50

    100

    80

    95.0% C I

    Percent

    100

    80

    95.0% C I

    Percent

    Date of study :

    Reported by:

    Name of product:Misc:

    Assessment Agreement

    Within Appraisers Appraiser vs Standard

    QA-4QA-2QA-1

    60

    40

    20

    0

    Appraiser

    Perce

    nt

    QA-4QA-2QA-1

    60

    40

    20

    0

    Appraiser

    Perce

    nt

  • 8/12/2019 Attribute Gage RR Ridgway

    47/50

  • 8/12/2019 Attribute Gage RR Ridgway

    48/50

  • 8/12/2019 Attribute Gage RR Ridgway

    49/50

    Two Big Lessons You cant trust your data until it is proven

    to be trustworthy. A single, one-time Gage R&R study is

  • 8/12/2019 Attribute Gage RR Ridgway

    50/50

    Questions? Thank You!