Transcript Slide 1

Loss Prevention,
Auditing & Safety Conference 2009
Title Sponsor:
Improving Organizational Safety
Through Predictive Modeling
Kris Russell – Sr. Manager Risk Strategy
Insights, Research & Analysis
Wal-Mart Stores Inc.
Agenda
 Introduction
 Predictive Modeling Defined
 Predictive Modeling in the Retail World
 What to Expect
 Summary/Conclusion
 Q&A
Agenda
 Introduction
 Predictive Modeling Defined
 Predictive Modeling in the Retail World
 What to Expect
 Summary/Conclusion
 Q&A
Predictive Modeling Defined
 Predictive Modeling – *Deloitte’s Definition
 Data Mining
 Algorithms
 Segmentation
 Segmentation
 Vulnerable Store Identification
 Focused Resource Deployment
*Deloitte Touche Tohmatsu
How Predictive Modeling Works
Multiple
Claim
Variables
 Score
Predictive
Equation/
Calculation
Indicator
Score
How Predictive Modeling Works
 Score Based Groups
 Skill Matches Group
 Better Initial Assignment
Predictive Modeling Concept – Example
Claim begins to exhibit traits that make it
suspicious to fraud investigators
Claim Investigation
Investigation Benefit
Traditional Fraud
Identificaiton Process
Fraud investigation
initiated
Claim is opened
When the model scores the
claim, it is flagged for
investigation
Investigation Benefit
Claim Investigation
Predicive Model Fraud
ID Process
Claim is opened
 Early ID
Fraud investigation
initiated
Benefit
 Prevention vs. Prosecution
Claim is closed
Claim is closed
Agenda
 Introduction
 Predictive Modeling Defined
 Predictive Modeling in the Retail World
 What to Expect
 Summary/Conclusion
 Q&A
Wal-Mart’s Predictive Modeling Philosophy
 Combine Multiple Models
 Produce Consolidated Score
 Overall Claims Evaluation
Wal-Mart Litigation Model Case Study
 Traditional Process
 Random
 Time Consuming
 Goal: Flag High Potential Claims
 Claim Opening + 30 Days
 Identification
Claim Management
Wal-Mart Litigation Model Case Study
 Uses 26 variables
 The Question:
Science = Experience?
 Outcome:
‘Lift’ in Identification
Wal-Mart Litigation Model Case Study
 ‘Lift’?
 Traditional Approach ->
Total Claim Pool
(1,000's of
Claims)
Desired
Claim
Pool
Search Includes The Total Claims Data
Population
Total
Claims ID’d
Early
(% of Total
Population)
 Predictive Modeling ->
Total Claim Pool
(1,000's Claims)
Desired
Claim
Pool
Narrow Sample
Search – Early
ID
More Cases
ID’d Early
Wal-Mart Litigation Model Case Study
 Adjuster 1: 14 Years Experience
 7 of 25
 Adjuster 2: 25 Years Experience
 6 of 25
 Model
 7.5* of 25
Agenda
 Introduction
 Predictive Modeling Defined
 Predictive Modeling in the Retail World
 What to Expect
 Summary/Conclusion
 Q&A
Predictive Modeling Life Cycle
Business
Understanding
Deployment
Evaluation
Data
Understanding
Data
Preparation
Modeling
Data Approach Choices
 Decentralized vs. Centralized
 Ownership of data
What to Expect
 Data is Key
 Ask For Help
 Use an experienced actuary
Agenda
 Introduction
 Predictive Modeling Defined
 Predictive Modeling in the Retail World
 What to Expect
 Summary/Conclusion
 Q&A
Summary/ Conclusion
 Predictive Modeling
Proactive Data Use
 Improved Resource Allocation
 Narrow the Window
 Data is Power
Agenda
 Introduction
 Predictive Modeling Defined
 Predictive Modeling in the Retail World
 What to Expect
 Summary/Conclusion
 Q&A