You are previewing Systems of Insight for Digital Transformation: Using IBM Operational Decision Manager Advanced and Predictive Analytics.
O'Reilly logo
Systems of Insight for Digital Transformation: Using IBM Operational Decision Manager Advanced and Predictive Analytics

Book Description

Systems of record (SORs) are engines that generates value for your business. Systems of engagement (SOE) are always evolving and generating new customer-centric experiences and new opportunities to capitalize on the value in the systems of record. The highest value is gained when systems of record and systems of engagement are brought together to deliver insight.

Systems of insight (SOI) monitor and analyze what is going on with various behaviors in the systems of engagement and information being stored or transacted in the systems of record. SOIs seek new opportunities, risks, and operational behavior that needs to be reported or have action taken to optimize business outcomes. Systems of insight are at the core of the Digital Experience, which tries to derive insights from the enormous amount of data generated by automated processes and customer interactions. Systems of Insight can also provide the ability to apply analytics and rules to real-time data as it flows within, throughout, and beyond the enterprise (applications, databases, mobile, social, Internet of Things) to gain the wanted insight. Deriving this insight is a key step toward being able to make the best decisions and take the most appropriate actions.
Examples of such actions are to improve the number of satisfied clients, identify clients at risk of leaving and incentivize them to stay loyal, identify patterns of risk or fraudulent behavior and take action to minimize it as early as possible, and detect patterns of behavior in operational systems and transportation that lead to failures, delays, and maintenance and take early action to minimize risks and costs.

IBM® Operational Decision Manager is a decision management platform that provides capabilities that support both event-driven insight patterns, and business-rule-driven scenarios. It also can easily be used in combination with other IBM Analytics solutions, as the detailed examples will show. IBM Operational Decision Manager Advanced, along with complementary IBM software offerings that also provide capability for systems of insight, provides a way to deliver the greatest value to your customers and your business. IBM Operational Decision Manager Advanced brings together data from different sources to recognize meaningful trends and patterns. It empowers business users to define, manage, and automate repeatable operational decisions. As a result, organizations can create and shape customer-centric business moments.

This IBM Redbooks® publication explains the key concepts of systems of insight and how to implement a system of insight solution with examples. It is intended for IT architects and professionals who are responsible for implementing a systems of insights solution requiring event-based context pattern detection and deterministic decision services to enhance other analytics solution components with IBM Operational Decision Manager Advanced.

Table of Contents

  1. Front cover
  2. Notices
    1. Trademarks
  3. IBM Redbooks promotions
  4. Preface
    1. Authors
    2. Now you can become a published author, too!
    3. Comments welcome
    4. Stay connected to IBM Redbooks
  5. Chapter 1. Introduction
    1. 1.1 Decision making
      1. 1.1.1 Systems of record
      2. 1.1.2 Systems of engagement
      3. 1.1.3 Systems of insight
      4. 1.1.4 Types of decisions
    2. 1.2 Motivation and business challenges
      1. 1.2.1 Decision-making process in general
    3. 1.3 Decision automation
      1. 1.3.1 Decision automation versus decision optimization
    4. 1.4 Business use cases
    5. 1.5 Example of system of insight provided in this book
  6. Chapter 2. Systems of insight solutions
    1. 2.1 Real-time solutions
      1. 2.1.1 Event-driven situation detection
      2. 2.1.2 Decision automation
      3. 2.1.3 Business activity monitoring
      4. 2.1.4 Stream processing
      5. 2.1.5 Next best action
      6. 2.1.6 Advanced patterns
    2. 2.2 Retroactive solutions
      1. 2.2.1 Manual analytic investigations
      2. 2.2.2 Big Data analytics
    3. 2.3 Proactive solutions
      1. 2.3.1 Predictive analytics
      2. 2.3.2 Prescriptive analytics
  7. Chapter 3. “A model for systems of insight
    1. 3.1 Systems of insight capabilities
    2. 3.2 Consume and collect
      1. 3.2.1 Consume
      2. 3.2.2 Collect
    3. 3.3 Analyze and report
      1. 3.3.1 Analyze
      2. 3.3.2 Report
    4. 3.4 Detect and decide
      1. 3.4.1 Detect
      2. 3.4.2 Decide
    5. 3.5 Integration
      1. 3.5.1 Data and transactions
      2. 3.5.2 Integration within a system of insight
      3. 3.5.3 Action execution
  8. Chapter 4. Consume and collect
    1. 4.1 Consuming data in motion
      1. 4.1.1 Event sources
      2. 4.1.2 Event distribution and filtering
    2. 4.2 Collecting data at rest
      1. 4.2.1 Batch processing
      2. 4.2.2 Database triggers
      3. 4.2.3 Big data and business rules
  9. Chapter 5. Analyze and report
    1. 5.1 Descriptive analytics
    2. 5.2 Predictive analytics
      1. 5.2.1 The predictive analytics process: The CRISP-DM data mining process
      2. 5.2.2 Predictive analytics with SPSS
      3. 5.2.3 SPSS Modeler capabilities
      4. 5.2.4 The predictive modeling process with SPSS
    3. 5.3 Real-time analytics
    4. 5.4 Prescriptive analytics
    5. 5.5 Cognitive analytics
    6. 5.6 Reporting and monitoring
      1. 5.6.1 Reports
      2. 5.6.2 Dashboards
  10. Chapter 6. Detect and decide
    1. 6.1 Introduction
      1. 6.1.1 Detect
      2. 6.1.2 Decide
      3. 6.1.3 Interaction between detect and decide
      4. 6.1.4 Decision management
      5. 6.1.5 ODM in the context of systems of insight
    2. 6.2 Request-driven decisions in ODM Decision Server Rules
      1. 6.2.1 Solution components
      2. 6.2.2 Decision Service projects
      3. 6.2.3 Modeling and authoring artifacts
      4. 6.2.4 Rule project artifacts
    3. 6.3 Situation-driven decisions in ODM Decision Server Insights
      1. 6.3.1 Decision Server Insights components
      2. 6.3.2 Business model definition
      3. 6.3.3 Agents
      4. 6.3.4 Lifecycle of an event
      5. 6.3.5 Building context
      6. 6.3.6 Using context
      7. 6.3.7 Scheduling rule execution
      8. 6.3.8 Integration and connectivity
  11. Chapter 7. Taking action
    1. 7.1 Actionable insight
    2. 7.2 Common actions
      1. 7.2.1 Asynchronous Interactions
      2. 7.2.2 Synchronous interactions
      3. 7.2.3 Feedback
    3. 7.3 Continuous improvement
  12. Chapter 8. Integration examples with ODM Advanced
    1. 8.1 Scenario overview
    2. 8.2 Architecture
    3. 8.3 Solution setup
      1. 8.3.1 Solution definition
      2. 8.3.2 Model definition
      3. 8.3.3 Agents definition
    4. 8.4 Consume and collect example
      1. 8.4.1 Message queue configuration
      2. 8.4.2 InfoSphere Streams configuration
      3. 8.4.3 DSI configuration (connectivity and server configuration)
    5. 8.5 Analyze and report example
      1. 8.5.1 Create scenario model
      2. 8.5.2 Store model as a stream on the server
      3. 8.5.3 Scoring model configuration
      4. 8.5.4 Predictive analytics agent implementation
    6. 8.6 Detect and decide example
      1. 8.6.1 Creating the decision service
      2. 8.6.2 Creating the OSGi service
      3. 8.6.3 Mapping the OSGi service to a BOM entry
      4. 8.6.4 Starting the service in a rule agent
    7. 8.7 Taking action example
      1. 8.7.1 Decision Server Insights outbound integration with IBM Integration Bus
      2. 8.7.2 Decision Server Insights outbound integration with IBM Business Process Manager (IBM BPM)
  13. Chapter 9. Building systems of insight with ODM Advanced
    1. 9.1 Adoption
      1. 9.1.1 Identify use case
      2. 9.1.2 Maturity model
    2. 9.2 Request-driven decisions development cycle
      1. 9.2.1 Rule analysis
      2. 9.2.2 Modeling
      3. 9.2.3 Authoring
      4. 9.2.4 Validation
      5. 9.2.5 Deployment
      6. 9.2.6 Monitoring
      7. 9.2.7 Execution patterns
      8. 9.2.8 Concept of operations
      9. 9.2.9 Applicability in a service-oriented architecture (SOA)
    3. 9.3 Situation-driven decisions development cycle
      1. 9.3.1 Analysis
      2. 9.3.2 Model
      3. 9.3.3 Solution development
      4. 9.3.4 Connect
      5. 9.3.5 Test
      6. 9.3.6 Deploy
      7. 9.3.7 Execute
      8. 9.3.8 Monitor
      9. 9.3.9 Concept of operations
      10. 9.3.10 Applicability in an event-driven architecture
    4. 9.4 Considerations for planning and designing DSI deployments
      1. 9.4.1 Designing for performance
      2. 9.4.2 Configuring for performance
    5. 9.5 ODM Decision Server Insights topologies
      1. 9.5.1 Component definitions
      2. 9.5.2 Minimum configuration
      3. 9.5.3 HA and CA configuration
      4. 9.5.4 Selecting your topology
    6. 9.6 Transactionality and rollback
  14. Appendix A. Integration of ODM Decision Server Insights with custom services
    1. A.1 OSGi service implementation
    2. A.2 XOM to BOM mapping
  15. Related publications
    1. IBM Redbooks
    2. Online resources
    3. Help from IBM
  16. Back cover