IBM Business Rules Management Systems Managing Complex Decisions in your Organisation - Simply

decorumgroveInternet και Εφαρμογές Web

7 Αυγ 2012 (πριν από 5 χρόνια και 9 μέρες)

606 εμφανίσεις

Business Rules Management Systems

Managing Complex Decisions in your Organisation
-

Simply

Richard J.D. Collard

Senior Industry Lead

WW Business Development

SME
-

Transactional Fraud, AML & Risk

richard.collard@uk.ibm.com

+44 7917 612896

Business Rules Management System
-

What Is It????!!!!!!


BRMS
-

more than a Rules Engine


Enabling & Complementary Component & Technology


In SOA, BPM or Stand
-
alone


Empowers Business
-
users


Provides Ability to make Complex Decisions


Consistently, Simply and Quickly


Externalises & Makes Business Logic Available to
Business
-
users


Hard
-
coded Business Logic becomes redundant


Benefits for IT and Lines of Business

Validation

Pricing

Eligibility

Validation Rules

Eligibility Rules

Pricing Rules

Why use Business Rules with BPM?

Business Rules


Automate key decisions in Business Processes by
Externalising

Business Logic


Simplify the Business Process structure


Are invoked from the Business Process as
standards
-
based Services


Are quicker to modify than the Business Process
orchestration

Engage and Support Separate Life Cycles


BPM and BRMS do not typically share the same
velocity of change
.


Business rules typically require
more frequent and more rapid
change cycles

than a process.

6

BPM Alone

BPM + BRMS

Decision Support

Business Decision
Making

Human

Automated / Assisted

Decision
Maintenance

Process re
-
engineering

Rules updates

Decision change
cycle

Months

Days

Decision Ownership

IT

IT + Business

Decision Sharing

Cut and Paste

Central Repository / TDS

Scalability


< 50 Rules
(when rules
enable)

Thousands

Supported decisions

Routing


Business Decisions


Task allocation


Dynamic routing

Combining BPM & BRMS

6

7



Issues


Rules are hidden in code or isolated
within the organization


Changes are hard to track and
maintain over time


Rules used by systems have to be
programmed and require IT resources


Duplication and multiple versions of
the same rules


Lack of auditability, traceability


Decision changes cannot be easily
tested or simulated

Business rules are crucial to operational
systems, and they change over time. The
traditional (ad hoc) approach of dealing
with rule changes leads to…

Traditional Approach for Managing Decision Change


Reduced organizational agility


Reduced employee productivity


Increased load on IT

Where Business Rules Typically Exist

Applications

Processes

People

Documents

8

Where Business Rules Typically Exist

Applications

Processes

People

Documents

B
usiness
R
ule
M
anagement
S
ystem

Rules are
Defined,
Analyzed and
Maintained

User Tools

Rules are
Stored
and
Shared

Rule Repository

Rules are
D
eployed,
Executed and
Monitored

Rule Server

Manage and Automate Decision Logic with BRMS


Eliminate decision silos


Make decision logic
accessible to Business and IT


Implement fine
-
grained,
context
-
specific logic

Business Rules Made Simple



Customizable vocabulary specific to
your organization, industry,
application etc….


Supports language localization


Integrates with external data
sources e.g. list of countries


Drop
-
down lists for customized
domain data


Templates facilitate new rule
creation

Flexible Rule Management for Business Users


Rule Solutions for Office

Text
-
based (point
-
and
-
click + guided editing)

Graphical representations

Technical Environment

Business Environment

Production Environment

Deploy

rules

Deploy

rules

Web
-
based console to monitor
& control rule execution

Rule Execution Server

Single node or cluster

(J2SE & J2EE compliant)

System

Administrator

Decision Validation Service

Testing & Simulation framework

Rule Team Server

Web
-
based Application for

Rule Maintenance

Rule

Manager

Rule Studio

Eclipse
-
based Rule IDE

Rule Service design & development

Rule

Developer

Rule

Repository

IBM ILOG JRules BRMS Product Components

Benefits of a BRMS Approach


Reduce time and resources
required to deploy changes


Author and maintain rules using
non
-
technical language


Express decision logic with
increased precision


Make decisions based on
specific context


Increase decision automation


Improve visibility and
understanding of how decisions
control systems


Lower maintenance costs;
respond quickly to change


Business experts can manage
and validate decision logic


Increase profitability of product,
pricing and promotional offerings


Customize decisions when
possible, standardize if needed


Improve process efficiency


Ensure compliance; enable
sharing/re
-
use of decision logic

What does it enable?

What is the value?

BRMS

13

Precise, Automated Process Decisions with WebSphere ILOG
BRMS

Horizontal: Best/appropriate price, Cross
-
sell/ up
-
sell recommendations, Loyalty
promotions, Exception identification, Risk/fraud assessment, Straight
-
through
processing approvals

Insurance

Banking

Healthcare

Government

Energy/Util./Telco


Claim


Validation


STP approval


Exception routing


Policy/Underwriting


Eligibility


Risk


Pricing


Annuity


Recommendation


Commissioning


Payout calc.


Loan


Eligibility


Risk


Pricing


Account


Cross
-
sell


Fraud/Alerts


Credit Card


Mktg Offers


Fraud


Credit limit



Patient Care


Drug interaction
risk warnings


Follow
-
up alerts


Member


Services
recommendation


Eligibility


Benefit
calculation


Provider


Patient eligibility
for services



Benefits


Eligibility


Calculations


Tax Payer


Classification


Audit flagging


Citizen


Program(s)
recommendation


Land/Permits


Conveyance
processing


Contract
compliance


Service Mgmt


Service
prioritization


SLA alerts


Maintenance
alerts


Order
configuration


Selection of clearing
channels


Minimizing C&S
costs



Consistent selection of least
-
cost channel (by consolidating
all routing decisions in a centralized decision service)


Rapid adaptation to price changes (e.g. new pricing scheme
of channels, “on us” settlement channel after acquisition)

Semantic validation
of messages


Increase straight
-
through processing


Separate valid messages that can be processed in STP from
messages that need exception handling

Use case

Customer

Description

ILOG BRMS customers in payments

Pricing




JRules
-
based pricing engine for all payments products


Differentiated pricing and complex bundle pricing


Reduction of time
-
to
-
market for pricing policies from 6
months to 10 days


ING Belgium considers its business as unique and cannot
manage its pricing policies using a package

Validation of
account numbers


Increase straight
-
through processing (country
-
specific validation rules)


Integration with EPP
-
based payments hub


Other use cases being analyzed: customer
-
specific
conditions for corporates

Use case

Customer

Description

ILOG BRMS customers in payments

Anti
-
Money
Laundering


Top
-
10 US Bank


Rule service performing country
-
specific AML
checks


Integration with EPP
-
based payments hub


Standard
-
based (ISO20022) service definition

Card fraud
detection


Fraud patterns captured as business rules to react
quickly to new fraudulent behaviors

Online fraud
detection

Global e
-
Auction


Fraud patterns captured as business rules to react
quickly to new fraudulent behaviors

Merchant
-
level
fraud detection


Merchant
-
level fraud detection provided as a value
-
added service to banks


Business rules as a way for TSYS to exploit value by
extracting knowledge from huge volume of card
transactions

Use case

Customer

Description

ILOG BRMS customers in card and online payments

Message validation,
calculation of
merchant fees &
internal routing


150M transactions processed in a 8 hour time window


Shorter time
-
to
-
market for new products


Specific rules for various regions

Clearing and
Settlement
System


Expose business policies for validation, interchange and
pricing


Speed of change; defect analysis, handling late change


Flexibility; ability to re
-
order rules or change flows
quickly


Quicker learning curve for new starters

17

Using BPM and BRMS Together


Chose
the claims process

as the best target for Nordic
-
level
unification and automation


Combination of WebSphere Process Server and WebSphere
ILOG JRules was based on the positive results from a test
project in Finland:


Decreased the cost of claims incurred


Increased of customer satisfaction through faster turnaround


Substantial savings of employee work


Uses of BRMS in this solution:


Claims submission validation


Decision automation for liability and compensation


Payment calculation


Straight
-
through processing determination


Manage and automate decision variation across countries,
processes and systems


Accelerate implementation of decision changes

Case Study: European P&C Insurance Company

Soci
é
t
é

G
é
n
é
rale


Customer Risk Management


SG : Major French retail bank, 9M+
customers, 2500 Branches, 13,000 branch
advisors


Branch Advisors spend 1,5 hr each day to
understand their customer’s account
situation and make decisions on what to
do next


They apply their own knowledge to decide
which customer situation to fix first and
how to fix it


There is a fairly high turnover of staff, so
there is little control on this knowledge


The Bank’s Management needs to
understand how many and which
situations are solved, handled, and which
ones may become critical


Management has identified there is a
possibility to reduce risk, whilst improving
the advisors’ productivity and support





Project High
-
level Specification


Identify risky situations on customer groups


A risky situation is


An non compliant situation that requires an action from the bank


It is generated, modified, closed upon reception of an event


It applies to an individual or to a service


It stays open until resolution


Risky situations are presented to branch advisors on a daily basis to help them


analyze the level of customer risk


assess which decisions should be taken.


Assist Branch Advisors by recommending a course of action


Explanations


Integrated rule based pre built customer letters, «

click and send

»


Integrated interview management


Notification of actions to other applications


Build an history of actions taken by the advisor for each customer


Provide a synthetic presentation of all open situations, by level of urgency/priority


Provide a synthetic view of open situations at the Branch level

Business case & targeted ROI


“High risk customer profile” process:











Benefits:


Manage 200 risk management profiles in 7 categories


Allow business users to manage & understand the risk profiles


Gain 1 hour/day/advisor

Get High Risk Profiles

When logging in

4,000,000 events on

9,000,000 accounts

Night

400,000 High Risk profiles


about 5% of customers

Advisor’s working day

Get explanations

And actions plan

13,000 advisors

2,500 branches


50,000 letters


per day

Get list of letters

For a customer

Update actions plan

Société Générale


Risk Management

21

Situation


Long time spent each day by branch officers to assess customer and account situations requiring
actions


Complexity of cases and related actions


Inconsistent operations leading to unmanaged situations

Goals

Solution


Service Oriented Architecture sharing 400 business rules


7 rule services ( Batch and Transaction processing programs) detecting and managing


400,000 high risk customer cases


Determine status and priority of the risky situations


Recommend and explain client actions to branch financial advisors


Generate personalized mails as actions ( 100,000 to 200,000 mails per day)

Benefits


Manage 200 risk management profiles in 7 categories


Allow business users to manage and understand the risk profiles


Gain 1 hr/day/advisor (13,000 advisors in 2,500 branches)



Improve risk detection and management



Increase branch financial advisor efficiency managing risk profiles



Automate action recommendations

BNP Paribas
-

Project OPRA Business Requirements


Offer and Product Repository Applications


BNPP wanted to improve its understanding and management
of the multiple banking offers that were being made to
customers


A
-

Flexible solution to allow definition of banking offers with


B
-

Ability to understand relationship between offers and


C
-

Determine conditions for appropriate, ethical sale


Requirement to allow Client Advisors to look into the product
catalogue and filter by
customer eligibility and profile with
differential pricing tariffs applied


Need to distribute offers to operational systems


mainframe
and distributed

BNP Paribas
-

Project OPRA Implementation


OPRA is based on IBM InfoSphere MDM Server for Product
Information Management and……..


…….IBM WebSphere ILOG JRules / BRMS


InfoSphere MDM Server provides offer authoring and
relationship management…….


….thereby ensuring offer package compatibility, base tariff
enforcement and conditional eligibility


ILOG BRMS provides the capability for BNPP to model eligibility
and pricing rules


The SOLUTION allows BNPP to build / maintain a catalogue of of
product / offer packages while managing collaborative
processes associated with implementation / deployment of
Business Rules

BNP Paribas
-

Project OPRA Benefits


OPRA is based on IBM InfoSphere MDM Server for Product
Information Management and……..


…….IBM WebSphere ILOG JRules / BRMS


Together providing:


Improved Time
-
To
-
Market for new products and services


Ability for non
-
technical users to generate and manage offers


Accurate and complete view of offer / product information


Multi
-
channel servicing initiatives


Personalised offerings


Evolution of offers according to market dynamics


Reduced likelihood of mis
-
selling

03/02/09

26

Automating decisions along a payments flow

Routing

Product

configuration

Validation

Fee
calculation


Validation & auto
-
repair


Validation of account
numbers (country
-
specific rules)



Determine product
and configure
payment attributes
based on
instructions and
customer profile


Least
-
cost routing
in CSM selection


Customer
preferences


CSM operational
constraints, e.g.
type of payments
supported, cutoff
times …


Real
-
time charge
calculation


Monthly discount
calculation


Payment bundles


Agreement and
customer
-
specific
conditions

SLA
-
based
exception mgt

Approval
rules

SEPA DD


Customer
-
specific
and SLA
-
based
limits



Determine what to
do next upon an
exception



Automate SLA
conditions


Mandate
management

Accounting &
reconciliation


Example: How to
reverse a cancelled
payment that is part of
a batch

Customer Success


Simplifying the Process

27

Customer

Process

Pains

BPM
-

Decision
Service

Results

Loan
Origination



Highly manual
process



Delays arose as
applications were
routed through
approvals via
interoffice mail



Lengthy approval
turnaround time



WebSphere ILOG
BRMS to handle a
variety of business
logic for a IBM BPM
-
orchestrated process:


-

Routing


-

Compliance


-

Scoring & rating


Cost savings


$1.5M over 5
years



15 days to 1 approval time
reduction



Improved staff focus and
satisfaction



Social
benefits
eligibility


Fragmented systems
with inconsistent data
meant decisions to
grant benefits were
not consistent from
one center to another




Several departments
but no clear
“interface” for the
citizens



IBM BPM to manage
the application
process end
-
to
-
end




ILOG JRules to
determine the
eligibility of applicants
to social benefits



Better service delivered to
the citizens: fewer
interactions required, clear
and fair application process



Consistency of eligibility
decisions: same criteria and
rules applied across all
agencies and centers


Transparency and
auditability: eligibility rules
can be reviewed and
validated by non
-
technical
users

Capital Markets

Insurance

T & T etc..

Banking

e
-
Gov


IBM ILOG Customers.........................

The Value of Simplification

Five Ways to Fast ROI With BRMS

Cost reduction

1. Reduced application
development


and maintenance

2. improve payments margins

3. Improved regulatory
compliance


Revenue generation

4. Provide value
-
added services

5. Faster time
-
to
-
market

31

Using the Right Tools for Different Types of Rules

Integration with other Components & Services

Rules Types

Rule Examples

Where are
Rules
Defined?

Right Tool

Business

Event Stream

Pattern
-
Match

Event pattern: person on the CIA
watch list, purchases chemicals ,
purchases a one
-
way plane ticket, uses
cash.

Event
-
stream
specific

Business Events

(CEP)

Process Flow / Task
Assignment

Assign loan review task to the
Underwriter with the lowest number of
tasks
-
in
-
queue.

Process
specific

BPM

Process Monitoring
(SLAs)

Trigger notification / process when
“Response Time” KPI for Final Approval
task exceeds maximum threshold
value.

Process
specific

BPM
-
BAM

Variable, Complex
Business Decisions
and Policies

Best Offer (pricing and promotions),
Eligibility Determination, Risk
Assessment and Mitigation, Straight
-
through Processing Determination

Decoupled
from
applications

BRMS

Business Event Processing/BRMS Combined Use Case
-

Retail

Promotional Offers


Use BEP to
detect

opportunity


Customer with regular history of book purchases visits
website after 3 months absence


Customer browses to 2 successive book categories with no
shopping cart activity


Raise opportunity alert


Use BRMS to
determine/decide

whether to offer a targeted
special, and if so, for what


Apply customer information against promotions rule
-
set


Determine best offer (e.g., no shipping charges for any
purchases in the next 1 hour)

Anti
-
Money Laundering


Use BEP to detect situation at risk


Successive near net
-
zero cash deposits/transfers


Absence of previous history of this pattern, or special
circumstances


Use BRMS to determine if a genuine violation, no problem, or a
“watch”


Evaluate customer history and events data against fraud
score and regulatory policies


Evaluate overriding factors


If not an actual violation, determine likelihood that activity is
suspicious


set watch or investigate status

Business Processing & BRMS Combined Use Case
-

AML

BRMS for Legacy Applications

Rules are
generated as
COBOL copybook
for execution in
IMS, CICS, batch

B
usiness
R
ule
M
anagement
S
ystem

Rules are
Defined
,
Analyzed and
Maintained

User Tools

Rules are
Stored
and
Shared

Rule Repository

New in v7.0


Simplified creation of Business Object Model for
COBOL applications


Enhanced ability to import, verbalize and
manage the business model for the COBOL
project


Additional JRules rule authoring capabilities for
generation of enhanced COBOL code


Increased agility for management of automated business
decisions in COBOL applications


Improved visibility of decision logic


Enables progressive application modernization strategy

Business Initiative Presentation to Agricorp

DEPLOY

JRules and Lombardi Teamworks: Standards
-
based
Integration

Windows/Unix/Linux Server

Communication Interface / ESB

Rule
Execution
Server

Interface Flexibility:



Web Services



Messaging



Java APIs

Teamworks
Process

Map to JRules Rule
Services via standard
Teamworks Integration
Services

Rule
Management

User Tools

Rule Repository

FileNet BPM/ECM & WebSphere ILOG BRMS

Windows/Unix/Linux Server or Mainframe

Web Services / APIs

Messaging Queues

FileNet Business Process

Communication Interface

Rule Studio

Rule
Execution
Server

Business
Rule
Management

Rule
Repository

Multi Platform Support:



Java/J2EE



z/OS


.NET

Integration Flexibility



Web Services



Messaging



Java/.NET APIs

Rule Team Server

Deploy

FileNet Content Management

Content

Repository

36