Ibm odm fraud detection & management system

© 2015 IBM Corporation

Fraud Detection & Management SystemA real time actionable counter fraud decision management system

Antonio Dell’Olio – Senior IT ArchitectBarbara Camandone – Client IT Manager

Headline

• Frauds in insurance context• Main fraud types• Fraud management key factors• Technology adoption in fraud management

• IBM Decision Management approach• Intesa SanPaolo Assicura project

• Customer overview• Customer business needs• Architectural overview

– Front Connection Tier– Middle Integration Tier– Business Logic Rule Tier

• Fuzzy Logic• Working plan

• Benefits of the solution• The “X” factor of project success

• Customer Experience

2

Frauds in insurance context

3

•The insurance industry is significantly affected, especially on the motor liability•Official data amounts to 3% of the total of the fraudulent claims reported•More than one third of people hurt in car accidents exaggerate their injuries•Studies on the subject demonstrate that over 30% of frauds comes from the inside•All insurance companies have anti-fraud measures, no one has an advanced maturity level•Fraud feeds a vicious circle•The Fraud phenomenon is increasing in all regions and is constantly evolving•This costs $13-$18 billion to America’s annual insurance bill

Main fraud types

4

Underwriting Phase Settlement Phase

Customers• False declarations• False documentations• Actions to prepare the fraud pattern

• Claim request about a never occurred event • Claim request about an event described with no clear

details

Body Shop Mechanics

• Invoice reimbursement request about a never occurred expense

• False billing that does not correspond to effectively performed services

Insurance Assessors

• Overestimated assessment or inaccurate damage reported

• Support to fraud scheme implementations

Counterpart Legal

• Support to complex fraud scheme implementations

Fraud management key factors

5

•Ability to identify the “suspected” objects into massive amounts of them, in a timely and accurate way

•Ability to deepen timely analysis of information related to the incident identified as suspicious, leading to the determination of possible fraud

•Ability to analyze even in false positives, false negatives and historical claims, in order to constantly improve the detection methods of undiscovered fraud schemes

•Ability to identify the risks associated with new offers and to realize the necessary protective measures against fraud, before launching these offers on the market

Technology adoption in fraud management

6

• Efficiency: technology can play rapid and massive processing• Effectiveness: Investigators analyze only qualified and highlighted

cases• ROI: significant amounts of resources are held back, compensation is

not paid• Customer intimacy: the end customer receives equity of treatment

and protection of the relationship• Dexterity: higher level of service is delivered in a context where the

risk is adequately managed and minimized

Fine Tuning

Governance

DeterrentDeterrent PreventionPrevention Detection Investigation Sanction & Case Closure

Sanction & Case ClosureDetectionDetection InvestigationInvestigation

Fine TuningFine Tuning

Headline








7

IBM Decision Management approach

8

Operational Decision Management Analytical Decision Management

Business Processes, Applications & Solutions

DecisionServices

BusinessRules & Events

Predictive Analytics & Optimization

Internal & External Data

Policy Regulation Best Practices Know-how

Risk Clustering Segmentation Propensity

Scenario Analysis& Simulation

Scenario Analysis& Simulation

Decision Management is a business discipline that enables organizations

to automate, optimize and govern repeatable business decisions.


9


How can we ensure that business decisions are managed in a controlled environment?

Governance

How can we ensure the right decision is being made at the right time?

Visibility

How can we rapidly respond to evolving market demands, competitive actions and regulatory requirements?

Collaboration


10


Check customer eligibility

ActActDecideDecide

If driver age is less than18 then

set eligibility to NOResult = Yes

Customer is Eligible

to the ”First Class

Car Insurance”

InvokeInvoke

Estimate price Insurance Estimated

Price = EUR 350,00

Call Center

Provide 5% discount to gold customers

Insurance application process

Driver &Car contextDriver &Car context

ResultResult

If the driver has got more than

3 accidents this year

then flag the driver as high riskDriver & Car ContextDriver & Car Context

ResultResult

Invocation of Contextual Decision synchronously from solutions


11


What is a Business Decision ? Combination of contextual and/or time-based rule artifacts

Contextual DecisionsThe externalised decision requires the solution to provide a well defined informational context for applying the action rules to.

The decision requires a vocabulary to interpret the well defined informational context

The result from the action rules is passed back to the calling solution.

The solution is responsible for taking action based on the result

ContextContext

ResultResultAction RulesAction Rules

Action RulesAction Rules

Action RulesAction Rules

RULEFLOW

RULEFLOW

SolutionSolution Externalized Business DecisionExternalized Business Decision

VocabularyVocabulary

Validation Decision - Pass/Fail result ie. Eligibility

Calculation Decision - Calculated result ie. Pricing, Tax

Classification Decision - Multiple results ie. Gold, Silver, Bronze

Validation Decision - Pass/Fail result ie. Eligibility

Calculation Decision - Calculated result ie. Pricing, Tax

Classification Decision - Multiple results ie. Gold, Silver, Bronze

Within the externalized decision 1 or more sets of action rules process the

informational context This is aided by a rule flow to process this in the

appropriate sequence and finalise the result

Different result types require different decision types

Headline








12

Customer overview

13

Intesa Sanpaolo Group – Insurance Pole

Customer overview

14

Distribution Channels

More than 4.000 «sportelli» of Banks of the Group

More than 4.200 Private Bankers

Over 300 agencies giving Personal and Business credits

Online site for Direct Selling

Customer overview

15

Our Mission

Value proposition based on services for families, enabled by technological devices

Bank tellers as a key element to achieve great mass of population

New offer of products approaching customers:«bundled» with Insurance + Services«unblundled» with Insurance or Services stand alone

Few Products, modulars and flexibles, through research of Technology and Services offered by market around us

Advanced and totally integrated IT Platform, completely web-based to improve synergy and working quality

Customer overview

16

Innovative Products

Combines traditional Car Insurance with an assistential component supported by a technological box with two different purposes: •Trace the car and intercept accidents•Direct contact with an operational unit to provide immediate presence on the accident site for every necessity

CAR INSURANCE

Combines traditional Home Insurance with a central device and additional accessories for Safety and Security of your home

Customer business needs

17

• As Intesa SanPaolo Assicura grew, our manual, labor-intensive fraud-review process became increasingly cumbersome

• The process limited our company ability to investigate suspicious filings and reduced its overall efficiency in settling legitimate claims in a timely manner

• We needed to implement a solution that could automate the fraud-detection process to quickly and accurately spot false claims

• Any fraud-identification solution had to integrate with the company claims processing system to automatically stop the payment process for fraudulent claims

Customer business needs

18

•Automate the management of existing anti–fraud indicators

•Identify the suspicions of fraud with a scoring subsystem

•Automate a real-time fraud detection in both phases (underwriting and settlement) of the claim process

•Allow the customer to free itself from the current service providers for the management of anti-fraud indicators

•Allow business users to act directly on the operation of the anti-fraud indicators

•Increase flexibility in the management of the anti-fraud indicators, ensuring to business users a better and wider information usability, in order to improve the Time to Market

The solution needs to:

Architectural overview

19

J2EE Messaging BusPrebuilt Data Model

XSD Schema

Back-end DBMS Repositories

Web ServiceStored Procedure

EJB

Web Service

Front Connection Tier

20

•The agency/website/operator caller invokes a SOAP Web Service exposed by IBM WebSphere Application Server

•A minimum set of key values needed to identify all the claim details is passed through Web Service input interface

•The web application acquires the request data, transforms it into XML message and puts into a processing bus JMS Queue

•The application returns to the caller the outcome of the processing request with a unique correlation identifier needed to the asynchronous callback mechanism

During the synchronous phase:

Middle Integration Tier

21


XSD Schema



Web Service

EJB

Middle Integration Tier

22

•The J2EE Message Driven Bean reads the message from the JMS Queue•The back-end Stored Procedures and Web Services are invoked using the claim identifier data•The application enriches and combines all returned data about claim details, customer and historical information into XML data model described by the XSD Schema•The application invokes the business rules engine (Rule Execution Server) using the collected data through EJB Local Interface•The data are matched against the Business Rules and the customer Scoring Algorithms•The returned results which contain the anti-fraud indicators, the generated score and the request identifier are stored into back-end systems•The transaction integrity and coordination is ensured by the IBM WebSphere Application Server persistence container capabilities

During the asynchronous phase:

Business Logic Rule Tier

23


XSD Schema



Web Service

EJB


24

•The collected claim data are matched against the available Business Rules through the IBM Rete Plus Algorithm•The Rules are expressed into a natural language that is easy to understand and to adopt at every level of the enterprise (Business and IT), so that it will be simpler to enable the change management and the governance of the Business Rule Lifecycle•The Rules combine multiple evidences, light on different markers and compute an overall suspicion score•A claim is classified as normal, abnormal or suspicious depending on its score calculated through the integration of the customer sophisticated algorithm-based scoring system•The rule scoring system contains the logic needed to detect fraud claims and it is inquired each time a new claim has been issued, verified or updated

During the Business Rules evaluation:


25

Some sample implemented Action Rules:

License Plate ControlActivation

%

If the input License Plate appears in at least 3 accidents occurred in the last 18 months then increase the IVASS indicator of 35 points

20%

If the input License Plate appears in at least 1 car accident and its vehicle was classified as destroyed then increase the IVASS indicator of 75 points

1%

If the input License Plate appears in at least 1 car accident happened in the last 5 years where the date of accident is after the effective date of the policy or in the last 15 days of effectiveness of the guaranteethen increase the IVASS indicator of 60 points

29%

Heuristic ControlActivation

%

If there is no crash report in the list of reports of the black box of the vehiclethen increase the GENERAL indicator of 75 points

63%

If the region of accident is not equal to the region of residence of the insuredthen increase the GENERAL indicator of 5 points

5%


26

Decision Tables:

Conditions

Each rowis a Rule

Actions


27

The structure of a Rule Solution:

FunctionTask

Pre/Post Conditions

RuleTask

FlowConditions


28

•IBM Decision Center Portal is available for Business Users in order to give them an easy to adopt rule authoring tool web based•It is possible to modify rules and deploy them in real-time (hot deploy), reflecting the changes immediately on the counter fraud patterns•Integrated security with Intesa SanPaolo Active Directory and CA SiteMinder Single Sign On•Ensures Team Collaboration and Rule Change awareness, specific domain vocabulary terms and real time error detection

Change Management, Rule Governance and Life-Cycle:

Fuzzy Logic

29

•We had to deal with Business Rules like:

the customer has been involved in at least 3 car accidents in the last 18 monthsthe customer has been involved in at least 3 car accidents in the last 18 months

•What happens with customer involved in 3 claims in the last 9 months, or customer involved in 3 claims in the last 20 months ?

the customer has been involved frequently in car accidents sometime around the last 18 months

the customer has been involved frequently in car accidents sometime around the last 18 months

Fuzzy Logic

30

• A “Polygonal Curve of Approximation” has been implemented in the context of the Execution Object Data Model, in order to approximate other curves and boundaries of real-life objects

• The underlying idea was to give to the Business Rules, implemented into IBM Operational Decision Manager, the general capability to use fuzzy quantifiers

• Exposition of new vocabulary terms (verbalization constructs) like “around” or “frequently”

Working plan

31

• Effort Time: 4 months• Total Elapsed Time: 7 months• Periodic progress reports• Test and Staging environments available before the Implementation

phase• Production environment available before the Deployment Support

phase

Work Breakdown Structure

Weeks

Main Phases #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12 #13 #14 #15 #16

Analysis and Design

Implementation

Test

Training & Support

Deployment Support

Training to end users

Business

IT

Milestone

What Now ?

32

• Peaks of 3000 Claims in one day

• 19276 evaluated Claims in 4 months

• Data Analysys on 4 different databases (Registry, Policies, Claims, Blackbox data)

• Over 1 million data per day

• Response time for a complex Claims in less than 3 seconds

• Media of 9000 Claims evaluation a week

Headline



• Customer presentation• Customer business needs• Architectural overview



• Benefits of the solution• The “X” factor of the project success


33

Benefits of the solution

34

•The solution, based on IBM products and the expertise of IBM consultants, has been encapsulated into a repeatable and reusable asset•The solution has a flexible, easy to integrate and adaptive architecture, based on SOA and IBM Decision management•The solution could be used in other insurance contexts, like Life and Home insurance•The solution has extremely increased the fraud detection capabilities, enabling the development of analytical capabilities•The solution is modular, scalable, invasiveness and invariant at any level of the organization•The solution is maintainable by the existing organization, both Business and IT departments

Qualitative benefits:

Benefits of the solution

35

•Medium claim end-to-end evaluation time: around 9 seconds•After 6 months of adoption the solution has generated significant cost savings and real business results•Anticipates a significant increment of saving up to 60% more annually by reducing fraudulent claim payments•Flags erroneous claims in real time, allowing immediate investigation or legal action•Identifies emerging fraud patterns so that the insurer can put new rules and algorithms in place to spot future false claims•Intesa SanPaolo Assicura has begun to increase its operating margins as it pays out fewer suspicious claims•The company has improved its ability to raise customers premiums or not renew insurance policies in case of repeated fake claims

Quantitative benefits:

Headline



• Customer business needs• Architectural overview



• Benefits of the solution• The “X” factor of the project success


36

The “X” factor of project success

37

Customer Experience

• Integration

• Synergy

• Collaboration

• People

Questions ?

Notices and DisclaimersCopyright © 2015 by International Business Machines Corporation (IBM). No part of this document may be reproduced or transmitted in any form without written permission from IBM.

U.S. Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM.

Information in these presentations (including information relating to products that have not yet been announced by IBM) has been reviewed for accuracy as of the date of initial publication and could include unintentional technical or typographical errors. IBM shall have no responsibility to update this information. THIS DOCUMENT IS DISTRIBUTED "AS IS" WITHOUT ANY WARRANTY, EITHER EXPRESS OR IMPLIED. IN NO EVENT SHALL IBM BE LIABLE FOR ANY DAMAGE ARISING FROM THE USE OF THIS INFORMATION, INCLUDING BUT NOT LIMITED TO, LOSS OF DATA, BUSINESS INTERRUPTION, LOSS OF PROFIT OR LOSS OF OPPORTUNITY. IBM products and services are warranted according to the terms and conditions of the agreements under which they are provided.

Any statements regarding IBM's future direction, intent or product plans are subject to change or withdrawal without notice.

Performance data contained herein was generally obtained in a controlled, isolated environments. Customer examples are presented as illustrations of how those customers have used IBM products and the results they may have achieved. Actual performance, cost, savings or other results in other operating environments may vary.

References in this document to IBM products, programs, or services does not imply that IBM intends to make such products, programs or services available in all countries in which IBM operates or does business.

Workshops, sessions and associated materials may have been prepared by independent session speakers, and do not necessarily reflect the views of IBM. All materials and discussions are provided for informational purposes only, and are neither intended to, nor shall constitute legal or other guidance or advice to any individual participant or their specific situation.

It is the customer’s responsibility to insure its own compliance with legal requirements and to obtain advice of competent legal counsel as to the identification and interpretation of any relevant laws and regulatory requirements that may affect the customer’s business and any actions the customer may need to take to comply with such laws. IBM does not provide legal advice or represent or warrant that its services or products will ensure that the customer is in compliance with any law.

Notices and Disclaimers (con’t)

Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products in connection with this publication and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. IBM does not warrant the quality of any third-party products, or the ability of any such third-party products to interoperate with IBM’s products. IBM EXPRESSLY DISCLAIMS ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.

The provision of the information contained herein is not intended to, and does not, grant any right or license under any IBM patents, copyrights, trademarks or other intellectual property right.

•IBM, the IBM logo, ibm.com, Bluemix, Blueworks Live, CICS, Clearcase, DOORS®, Enterprise Document Management System™, Global Business Services ®, Global Technology Services ®, Information on Demand, ILOG, Maximo®, MQIntegrator®, MQSeries®, Netcool®, OMEGAMON, OpenPower, PureAnalytics™, PureApplication®, pureCluster™, PureCoverage®, PureData®, PureExperience®, PureFlex®, pureQuery®, pureScale®, PureSystems®, QRadar®, Rational®, Rhapsody®, SoDA, SPSS, StoredIQ, Tivoli®, Trusteer®, urban{code}®, Watson, WebSphere®, Worklight®, X-Force® and System z® Z/OS, are trademarks of International Business Machines Corporation, registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at: www.ibm.com/legal/copytrade.shtml.

http://www.ibm.com/legal/copytrade.shtml

Thank YouYour Feedback is

Important!

Access the InterConnect 2015 Conference CONNECT Attendee Portal to complete your session surveys from your smartphone,

laptop or conference kiosk.

Documents

Ibm odm fraud detection & management system