Aasman_AIDA Intro_and_triple store usage

advertisement
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
Amdocs Intelligent Decision Automation
Overview for Gartner
AIDA and Guided Interaction Advisor to
be announced Feb 14, 2011
Amdocs – Bill Guinn, Mike Lurye, Susan MacCall
December 7, 2010
1
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
The Leader in Customer Experience Systems
Innovation
A Unique Business Model
Service Provider Industry Focus
Leading BSS, OSS & Service Delivery
Products
Strategic Services
2
>
$3.0B in revenue
>
$410M operating income
>
$1.4B in cash
>
More than 19,000 employees
>
Over 60 countries
Leading Telecom
Operations Management
Systems Vendor
Worldwide.
Highest Possible Overall
Rating in BSS Scorecard
and OSS Scorecard
Reports.
(May 2010)
(Dec 2009, Jan 2010)
Information Security Level 1 – Confidential
TM Forum’s “Industry
Leadership” award.
(June 2010)
© 2010 – Proprietary and Confidential Information of Amdocs
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
What Is Amdocs Intelligent Decision
Automation (AIDA)?
>
>
3
Amdocs Intelligent Decision Automation (AIDA) is a closed-loop, selflearning system that lets you…
>
See what happens, when it happens
>
Understand what it means to your business
>
Take action and enforce business policy –
automatically, intelligently and in business real-time
By uniquely combining technologies AIDA is able
to better understand and predict the behavior of
customers, providing individualized treatment to
achieve specific business targets:
>
Increasing RPU by selling them the right products
>
Decreasing churn by treating customers based on
their specific likes, needs, and recent issues
>
Reducing cost of operations by proactively
preventing issues and optimizing cross-channel
care functions
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
Requirements
>
Massive scale, billions of events/day -> 100M customers
>
Business real time infrencing and decisioning (100ms response time)
>
Uncoupled integration – open world model –
>
M:M:M : atomic ->abstraction : structured and unstructured data
>
Cohesive past, present, and real time view World View
>
True user managed and defined schema and decision factors (concepts)
>
User defined policy, decisions, and the logic to manage concepts
>
Closed loop learning: decision->actions->effect->learning
>
Low cost .05 customer/year
4
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
AIDA Runtime Architecture
Actions
Decision Engine
Events
Application Server - Space Based
Amdocs
Event Collector
Event
Ingestion
Events
Semantic
Inference
Engine
and
Container
Business
Container
Rules
Bayesian
Belief
Network
Scheduled
Events
Revenue
Mgt
Ordering
Amdocs
Integration
Framework
CRM
“Sesame”
Other
Other
Other
System
System
Systems
CRM
Performance cache
Other
Other
System
Systems
Network
Business
Intelligence
Data Sources
Any internal or external system
Structured or Unstructured
5
Information Security Level 1 – Confidential
RDF
Triple Store DB
© 2010 – Proprietary and Confidential Information of Amdocs
Operational Systems
Any operational system
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
How the AIDA Inference Engine Works
Semantic Concept Model
Machine learning trained models
Semantic Concept Model
The model defines the concepts including the high
level business concepts
The model contains the relationship between
concepts including the dependencies
Defines concepts and meaning
Defines relationships between concepts
Semantic model capabilities reasoning
BBN probabilistic model
Recommendation model ( Future)
Other Machine learning algorithms
Inference
When an event occurs the event handler rule fires for that event
Evaluates the event message
Evaluates the existing ontology
Determines which semantic instances to create or update
Inference
Engine
and
Business
Rules
When any data changes, the inference engine fires in a “When Then” style of computing, updating all “Automatic” concepts.
Custom concept rules are fired if necessary. This creates a chain
of updates
Bayesian
Belief
Network
When a “on demand” concept is needed the inference engine
finds and computes all of the dependant concepts
Business policy Rules
6
Event
Create semantic concepts from events
Concept
Action
Define a custom business concept
A set of custom rules actioning business policy
Information Security Level 1 – Confidential
Machine learning
When a concept is dependent on “machine learned”
information the inference engine manages the invocation and
timing of interfacing
© 2010 – Proprietary and Confidential Information of Amdocs
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
Why a triple Store, why Allegrograph
Characteristic
Triple
Relational
Store (AG )
No-SQL
Dynamic data model
Inference
Real Time
Scalability
Semantic Capabilities & M:M:M
Unstructured data
TCO
7
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
Multidimensional
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
AIDA Business Composer
Library
Composer Demo Application
RELATIONSHIP MAP
TAG
DEPENDENCY MAP
PATH
Customer
ACTIONS
ENTITIES
PROCEDURES
EVENTS
BUSINESS ENTITIES & CONCEPTS
+
Entity Name
Customer
Acquisition Date
Status
Acquisition Date
Status History
Customer Accounts
Status
Type
Customer Bill
Status History
Call Frequency
+
Type
Churn Propensity
Call Frequency
Collection Risk
Churn Propensity
+
-
Customer
Customer Problems
Interactions
Payment Pattern
Collection Risk
Customer Accounts
Subscriptions
Payment Pattern
Customer Action
Customer Bill
Customer Devices
Customer Problems
CONSTANTS
+
Customer
Description. The role played by an Individual or
Organization in a business relationship with the
service provider in which they intend to buy, buy,
or receive products or services from the service p
8
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
Filter Map
Search Map
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
AIDA Business Composer
Library
Composer Demo Application
RELATIONSHIP MAP
TAG
DEPENDENCY MAP
Filter Map
PATH
Search Map
Customer
ACTIONS
ENTITIES
PROCEDURES
Customer Payment Pattern
EVENTS
BUSINESS ENTITIES & CONCEPTS
+
VIEW
EDIT
DEPLOY
Entity Name
Customer Payment Pattern
Customer
LAST MODIFIED: 02/12/2010
MODIFIED BY: Greg Sloan
COMPUTED:Automatically
VERSION: 1.3
Acquisition Date
DESCRIPTION: Determine the payment pattern for the customer by evaluating payments determining good payer, bad payer,
worsening payer or improving payer
Status
Acquisition Date
Status History
1.
Find
2.
If
all of the
Within Accounts
6
Customer
Previous Bills
Customer Bill
Status History
Call Frequency
+
Type
Churn Propensity
-
Call Frequency
Collection Risk
Churn Propensity
Payment Pattern
all bills
have
+
Payment Timeliness
Customer Accounts
3.
-
then
Customer Payment Pattern is
otherwise
If
all bills
then
Customer Bill
have
Payment Pattern is
otherwise
If
and
5.
at least
later bills
If
and
Customer Problems
good payer
Payment Timeliness
are
At least
all later bills are
75%
of
late
then
50%
Earlier Bills
Payment Pattern is
of
On time or early then
Earlier Bills
are early or on time
worsening payer
are
Payment Pattern is
+
Description. The role played by an Individual or
Organization in a business relationship with the
service provider in which they intend to buy, buy,
or receive products or services from the service p
Information Security Level 1 – Confidential
late
bad payer
Customer
9
equal to
Subscriptions
4.
Customer Devices
early
Interactions
Payment Pattern
Customer Action
equal to
Customer Problems
Collection Risk
CONSTANTS
months
Status
Type
© 2010 – Proprietary and Confidential Information of Amdocs
Late
Improving payer
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
The Applications
Guided Interaction Advisor (GA in May)
Virtual Agent (GA in September)
10
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
Guided Interaction Advisor (GIA)
>
A pre-built Ontology and rule set in AIDA designed
to address key issues in call center interaction
>
Amdocs Guided Interaction Advisor
>
>
>
Business Benefits
>
>
>
>
Eliminates system and agent diagnosis time
Provides consistent and efficient call handling
Increases agent and customer satisfaction
Anticipated benefits based on 100K actual
accounts assessment:
>
>
>
11
Anticipates reason for the customer interaction
Then Automates access to the required information and
Guides the flow of action and decision making
AHT reduction of 10-15%
FCR improvement of 10-15%
CSR training time reduction of 15-20%
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
How Guided Interaction Advisor Reduces AHT
Example
* In addition to displaying the bill, GIA returns multiple “next”
actions such as dispute charge, adjust the fee, make payment
depending on high level concepts and policy
GIA Looks at events including bill sent event
Finds abnormal fee – Returned Check
Probability that the call is motivated by fee vs. other issues
Compute High-level concepts
Examine company policy for treating this customer
Determine actions for each probable issues
Dollars
Payments
Policy action rules
Motivation for call
Bill Due Date is Oct 1
Third Party Charges are normal
Charges are not typical
Assess
Probability
• Abnormal Fee
• Pay Bill
• Device Exchange
• Customer Cancel
47%
23%
12%
5%
LOGIC ZOOM:
VIEW
EDIT
Late
VIEW
EDIT
.
Customer
Graph
Information Security Level 1 – Confidential
advanced
VERSION: 1.3
If
Customer
has a late Fee
and Customer Payment Pattern MODIFIED
is
OWNER:
User Name
BY: User Name
DESCRIPTION: Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed condimentum ante eu turpis placerat
porttitor. Maecenas vitae lectus non justo mollis rutrum at condimentum mi. [Edit]
.
then IfCredit
Fee
Customer
has a. late Fee
• High value customer
• Long time customer
• Improving payer
• Contract expiring
Abnormal fee
simple
VERSION: 1.3
DEPLOY
Abnormal Fee Policy
1
1
High level Concepts
fee is “Returned check”
12
advanced
LOGIC ZOOM:
OWNER: User Name
MODIFIED BY: User Name
DESCRIPTION: Lorem ipsum dolor sit amet, consectetur adipiscing elit. Sed condimentum ante eu turpis placerat
porttitor. Maecenas vitae lectus non justo mollis rutrum at condimentum mi. [Edit]
Bill has charges for fees.
Tax amount is normal
simple
DEPLOY
Good Payer or Improving Payer and Customer Value
Bill has abnormal fee
CIM actions
is not Low
and Customer Payment Pattern is
Good Payer or Improving Payer and Customer Value
then
Credit Fee
Amount Received
Partial Payment
Payment
and
and
and
and
is not Low
.
3
Customer Bill
3
Interactions
Presents contextually relevant info, streamlined action
Other
*
Display bill
Highlight fee on bill
Script message
Prepare one click action
Highlight one click action
Late Fee
Credit Fee
Method
Customer Payment Pattern
Amount Received
Partial Payment
Payment
and
and
and
and
Late Fee
Credit Fee
Method
Customer Payment Pattern
Guided
Interaction
Advisor
© 2010
– Proprietary and
Confidential Information of Amdocs
Pay Bill
Educate free on-line pay
Prepare one click pay
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
Guided Interaction Advisor
Events collected in
real time
Transformed into a connected
graph of business concepts
Predictions & Actions
Plan Overage
Bill greater than last month
Prorates
Interactions
Roaming charges
Third Party Charges
Orders
Bills
Abnormal fee
Payments
Rate increases
Collections
Charge dispute
Charge dispute
First bill
Past due amount
Customer
Pay bill
Pay instructions
Customer Cancellation
Reactivation
Individual
Device activation
Device Activated
Device education
Device heartbeat
Subscriptions
Device changes
Chronology of events
Subjective
"good payer"
Patterns
"always pays 2 days late"
Trends
“improving payer"
Geospatial
“within 5 miles of the tower"
Time
“within 5 minutes of an outage"
Probability
“probably will call about the bill"
Absence of occurrence
“missed payment”
Highly
User
extensible
…
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
Extensible
13
Device exchange
Device Lost
Device not working
Device resume
Service data not working
Service text not working
Service voice not working
Use Case Extensions
AMDOCS > CUSTOMER EXPERIENCE SYSTEMS INNOVATION
Presenting Insight to CSR
Prediction on reason for the
call – ranked by probability
Process opens
relevant screen for
reference and action
Recommended actions –
based on best ROI
Prioritized Recommended
treatment and script
14
Information Security Level 1 – Confidential
© 2010 – Proprietary and Confidential Information of Amdocs
Download