Pareto Analysis - · PDF file 2010-01-09 · III. Pareto Analysis: Frequency Versus...

Click here to load reader

  • date post

    14-Jul-2020
  • Category

    Documents

  • view

    15
  • download

    0

Embed Size (px)

Transcript of Pareto Analysis - · PDF file 2010-01-09 · III. Pareto Analysis: Frequency Versus...

  • 1

    Pareto Analysis

  • 2

    Seven Basic Quality Tools

    1. Process Mapping / Flow Charts* 2. Check Sheets

    3. Pareto Analysis 4. Cause & Effect Diagrams 5. Histograms 6. Scatter Diagrams (XY Graph) 7. Control Charts

  • 3

    Topics

    I. Pareto Principle

    II. Performing a Pareto Analysis Constructing a Pareto Chart

    III. Pareto: Frequency Vs. Cost Analysis

    IV. Pareto Analysis Exercise Using QETools

    V. Pareto Drill Down

  • 4

    I. Pareto Principle

    Pareto* Principle provides the foundation for the concept of the “vital few” and a “trivial many”.

    Examples: Quality – a small percentage of defect categories (causes) will constitute a high % of the total # defects. Cost – a small percentage of components will constitute a high % of total product cost. Others: Inventory, absenteeism, downtime.

    *Note: Wilfredo Pareto – 19th Century Italian economist studying wealth who observed that a large proportion of wealth is owned by a small percentage of the people. Pareto principle was later applied to quality by J.M. Juran.

  • 5

    80/20 Rule

    Pareto principle is sometimes referred to as the 80/20 rule.

    In quality, this rule suggests that ~20% of defect categories will account for ~80% of the total number of defects.

    “Vital Few – Trivial Many”

  • 6

    II. Pareto Analysis Ranking of data by importance in descending frequency (highlights most significant concern)

    Example: Reasons for Delays in Preparing New Product Bids

    Pareto - New Bid Delays

    0 10 20 30 40 50 60

    In su

    ffi ci

    en t c

    us to

    m er

    s pe

    c' s

    R eq

    ui re

    m en

    t c ha

    ng e

    by c

    us t

    U nk

    no w

    n te

    st re

    qu ire

    m en

    ts

    W ai

    t f or

    a pp

    lic at

    io n

    re vi

    ew

    P ric

    in g

    in fo

    n ot

    a va

    ila bl

    e

    R es

    ea rc

    h si

    m ila

    r p ro

    du ct

    pr ic

    in g

    W ai

    t f or

    e ng

    in ee

    rin g

    re so

    ur ce

    s

    W ai

    t f or

    s al

    es re

    so ur

    ce s

    Q uo

    te p

    ac ka

    ge in

    co rre

    ct

    D at

    a ba

    se s

    ea rc

    h er

    ro rs

    Fr eq

    ue nc

    y

  • 7

    First, Obtain Frequency Sum Data for Each Category

    From check sheet, create a table of categories and occurrences (i.e., frequency). Example: Reasons for Delays in Preparing New Bids

    Reasons for Delays Frequency Insufficient customer specifications 56 Internal pricing information not available 18 Wait for application review kickoff 5 Requirement change by customer 30 Quote package filled out incorrectly 45 Wait for engineering resources 8 Wait for sales processing resources 10 Research for similar product pricing 10 Unknown test requirements 11 Data base search errors 3

    Total 196

  • 8

    Second, Create a Pareto Table Sort in Descending Order (by Freq or Relative Freq) Compute Relative and Cumulative Frequencies

    Relative Frequency ~ Frequency / Total (56/196=29%) Cumulative Freq % ~ Running total of % (29% + 23% = 52%)

    Reasons for Delays Freq Rel Freq, % Cum Freq Cum Rel Freq, % Insufficient customer spec's 56 29% 56 29% Requirement change by cust 45 23% 101 52% Unknown test requirements 30 15% 131 67% Wait for application review 18 9% 149 76% Pricing info not available 11 6% 160 82% Research similar product pricing 10 5% 170 87% Wait for engineering resources 10 5% 180 92% Wait for sales resources 8 4% 188 96% Quote package incorrect 5 3% 193 98% Data base search errors 3 2% 196 100%

    Total 196 100%

  • 9

    Pareto Chart

    Left Y-axis – Frequency or Relative Frequency Right Y-axis – Nothing or cumulative percentage line.

    0%

    5%

    10%

    15%

    20%

    25%

    30%

    Ins uff

    icie nt

    cu sto

    me r s

    pe c's

    Re qu

    ire me

    nt ch

    an ge

    by cu

    st

    Un kn

    ow n t

    es t re

    qu ire

    me nts

    Wa it f

    or ap

    pli ca

    tio n r

    ev iew

    Pr icin

    g i nfo

    no t a

    va ila

    ble

    Re se

    arc h s

    im ila

    r p rod

    uc t p

    ric ing

    Wa it f

    or en

    gin ee

    rin g r

    es ou

    rce s

    Wa it f

    or sa

    les re

    so urc

    es

    Qu ote

    pa ck

    ag e i

    nc orr

    ec t

    Da ta

    ba se

    se arc

    h e rro

    rs

    R el

    F re

    q %

    0%

    20%

    40%

    60%

    80%

    100%

    120%

    C um

    ul at

    iv e

    Fr eq

    %

  • 10

    III. Pareto Analysis: Frequency Versus Cost (or Severity)

    Pareto Analysis may be performed using: Frequency of occurrence (expressed as a frequency count or relative frequency %), Or Total cost, Or Severity, adverse outcome, or avoidability

    Note: the most frequently occurring item may not be the most important item to address first.

  • 11

    Assessing Cost Impact

    Suppose a hospice has the following Pareto Frequency Analysis for Medicare denials.

    If the cost for an occurrence varies by category, one may weigh the categories by multiplying the frequency by estimated cost per occurrence (e.g., average cost).

    Category Frequency Cost per

    Occur Total Cost

    Inc supervisory visit 113 362 40906 Not recipient 46 536 24656 Unsigned election 12 650 7800 Non-terminal disease 11 882 9702 Unsigned Certification 8 13790 110320 Unmet Level of Care 6 31851 191106 Unmet Plan of Care 4 1289 5156

  • 12

    Pareto: Cost Vs. Frequency

    Would the priorities be different based on a cost analysis?

    TOTAL 389688

    Category Total Cost Relative Frequency Cumulative Frequency

    Unmet Level of Care 191103 49.0% 49.0% Unsigned Certification 110313 28.3% 77.3% Inc supervisory visit 40924 10.5% 87.8% Not recipient 24678 6.3% 94.2% Non-terminal disease 9711 2.5% 96.7% Unsigned election 7802 2.0% 98.7% Unmet Plan of Care 5157 1.3% 100.0%

    TOTAL 200

    Category Frequency Relative Frequency Cumulative Frequency

    Inc supervisory visit 113 56.5% 56.5% Not recipient 46 23.0% 79.5% Unsigned election 12 6.0% 85.5% Non-terminal disease 11 5.5% 91.0% Unsigned Certification 8 4.0% 95.0% Unmet Level of Care 6 3.0% 98.0% Unmet Plan of Care 4 2.0% 100.0%

    By denial

    By cost of denial

    Medicare billed for Higher cost than expected based on criteria.

  • 13

    IV. Lecture Exercise: Pareto Analysis for Loan Turndowns

    Defect Categories for Loan Turndowns Closing costs too high, selling home, change in marital status, change in job status, not saving enough, lost interest, interest rate is too high, miscellaneous.

    Using the data file, pareto.xls create a pareto table of frequency, relative frequency (%), and cumulative frequency, and then a Pareto Chart.

  • 14

    Exercise: Pareto Analysis

    Step 1: Sum by Defect Category Using QE Tools – perform Binary Cross Tabulation to obtain frequency counts for each category for loan check sheet data.

    Step 2: Run a Pareto Analysis Using the Sum Data from Step 1, create a pareto table and chart.

  • 15

    Step 1: Binary Cross Tabulation

    Using check sheet data (see sample of data below) for the different loan turndown categories, select: QETools >> Tabulation >> Binary Cross Tabulation

    Note: only first 9 rows are shown from file pareto.xls

  • 16

    Binary Cross Tabulation Example

    From binary cross tabulation, QETools automatically creates new data columns for categories and frequency counts in “Datasheet”

    Auto Save Categories Sum to “Datasheet”

  • 17

    Step 2: Pareto Analysis

    Select: QETools >> Graphical Tools >> Pareto

  • 18

    Pareto Table: Results

    TOTAL 132

    Category Frequency Relative Frequency Cumulative Frequency

    HighClosingCosts 63 47.7% 47.7% Selling_home 37 28.0% 75.8% Change_marital 16 12.1% 87.9% Change_job 5 3.8% 91.7% Insuff_Saving 4 3.0% 94.7% Lost_interest 3 2.3% 97.0% rate-too-high 3 2.3% 99.2% miscellaneous 1 0.8% 100.0%

  • 19

    Pareto Chart by Relative Frequency (with Cumulative Frequency Line)

    Pareto Chart

    0.0%

    10.0%

    20.0%

    30.0%

    40.0%

    50.0%

    60.0%

    Hig hC

    los ing

    Co sts

    Se llin

    g_ ho

    me Ch

    an ge

    _m ari

    tal Ch

    an ge

    _jo b

    Ins uff

    _S av

    ing Lo

    st_ int

    ere st

    rat e-t

    oo -hi

    gh mi

    sc ell

    an eo

    us

    Category

    R el

    at iv

    e Fr

    eq ue

    nc

    0.0%

    10.0%

    20.0%

    30.0%

    40.0%

    50.0%

    60.0%

    70.0%

    80.0%

    90