Scribd Logo
Upload

Welcome to Scribd!

  • 141 views

    Smoker Propensity Project Charter

    Uploaded by

    Houessou Houeto Hervé
    Smoker Propensity Project Charter

    Copyright:

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

    Smoker Propensity Project Charter

    Uploaded by

    Houessou Houeto Hervé
    141 views2 pages
    Smoker Propensity Project Charter

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Smoker Propensity Project Charter

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    141 views2 pages

    Smoker Propensity Project Charter

    Uploaded by

    Houessou Houeto Hervé
    Smoker Propensity Project Charter

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    You are on page 1of 2

    Project Charter: Project 1 1/2

    Case for change / Problem Statement Assumptions

    There is a need for quicker turnaround time, swifter issue of policies and customer Distribution of a smoker & non-smoker is binomial
    buying improvement from life insurance policies. The challenge life and health Age is linear and there is consistency
    insurers face is to refine and simplify products without compromising strict
    underwriting practices and its risk profile. The biggest dilemma is to eliminate the Enough storage/processing space will be given by UConn
    practice of fluid tests, which 1. frustrate the consumer and can potentially result in servers
    that customer going elsewhere (lost profit); and 2. create unnecessary costs for Selection dataset represent sample entire US population
    the company, all while providing the best and most competitive price on the
    market. Top Risks (max. 5)

    Business Case 1. Failure to meet deadline


    2. Target accuracy will not be attained given Company Xs data
    There is a financial value and significant market share for life and health insurers source
    by producing a high accuracy model that would predict smokers. The model would
    help eliminate intrusive testing, such as fluid and other medical tests that result in 3. Not enough information provided by Company X resulting in
    lost customers and ultimately lost profit to the company and additional incurred inaccurate outcomes
    expenditures that could be avoided. Further, the model would make the prospect
    to committed client process faster as well as offer a more competitively priced
    simplified issue life insurance product. In Scope Out of Scope
    Risk assessment during the underwriting process is very cumbersome and
    complicated. In order to ease and shorten the process, a desired accuracy level Use Company Xs proxy as a Non-US population
    must be met. The goal is to exceed Company Xs minimum predictive accuracy benchmark for our model
    and balance both sensitivity and specificity of the model. This would ultimately accuracy
    make the underwriting process more effective and allow for swifter issuing of the
    policy. Exploration of data Predicting Company X
    There are considerable amount of studies done on smoker propensity and approving a customer for a
    smoker cessation. Utilizing this data as complementary information along with policy
    data analytic approaches would be one avenue to help detect fraud as well as
    improve the model and target accuracy. As an example, can data such as Researching forecasting tools External usage of copyrighted
    reviewing the persons motor vehicle report, social media pages, credit reports data
    and time perspective, the propensity to consider short-versus long-term
    consequences of ones actions, aid in improving a smoker propensity model? Leverage consumer/public data
    Would other healthy living indicators serve as predictors for a non-smoker? Can Constraint Matrix
    sources
    Objectives / Goal
    demographics, suchstatements
    as highest smoking rate being Kentucky, attribute to a
    smoker? MOST SOME LEAST
    Achieve a 90% predictive accuracy on the smoking propensity model, which
    would allow for a common proxy for model accuracy.
    Explore any avenues to surpass Company Xs accuracy level using innovative SCOPE X
    data sources publicly available and/or consumer data as predictors

    TIME X
    Project Charter: Project 1 - 2/2
    Deliverables & Milestones (High Level Milestone Plan) Project Organization
    TENTATIVE SCHEDULE
    Sponsor
    Project Start Date: Project End Date:

    Key Deliverable Milestone Name Planned MS


    Steering Board (Advisor/Professor)
    (MN) Date
    Members (UCONN)
    Project Charter Sponsor Status Project Manager
    Meeting

    First-dry run presentation Mid-term Tollgate .


    Signof
    Final Report Final Reporting Role Members
    Team Signature Date

    Sponsor
    Slide deck presentation Final Toll-gate

    UCONN Member

    UCONN Member

    UCONN Member

    UCONN Member

    UCONN Member

    UCONN Member

    You might also like