Decision Support & Business Intelligence Systems – Chapter 1 Study Notes
Learning Objectives
Understand today's turbulent business environment and how organizations survive and excel
Recognize the need for computerized support of managerial decision making
Recall an early framework for managerial decision making (Gorry & Scott-Morton / Anthony-Simon taxonomy)
Grasp the conceptual foundations of Decision Support Systems (DSS)
Comprehend Business Intelligence (BI) methodologies, relate them to DSS, and see the evolutionary path from DSS to BI
Business Pressures–Responses–Support Model
Three interacting pillars:
Business Environmental Factors/Pressures
Organizational Responses
Decisions & Support (often computerized)
Goal: align real-time analytics and integrated computer-based decision support with strategy, agility, and productivity
Business Environment Factors
Markets
Strong competition, expanding global markets
Flourishing electronic markets (Internet), innovative marketing methods
IT-supported outsourcing opportunities, need for real-time/on-demand transactions
Consumer Demand
Customization expectations, product quality & diversity, faster delivery
Customers are empowered yet less loyal
Technology
Rapid innovation in products/services, high obsolescence rate
Information overload, social networking (Web 2.0→beyond)
Societal Factors
Greater regulation & deregulation, diverse/aging workforce, more women
Homeland security/terrorism concerns, Sarbanes-Oxley & similar legislation
Corporate social responsibility & sustainability emphasis
Organizational Responses to Pressures
Be reactive, anticipative, adaptive, proactive
Typical managerial actions:
Strategic planning, innovative business models, business process restructuring
Alliances & partnership improvement, superior information systems
Encourage creativity & innovation
Enhance customer service/relationships, transition to e-commerce
Shift to make-to-order & on-demand production/services
Exploit new IT for communication, data discovery, collaboration
Rapid reaction to competitors (pricing, promotions, products)
Automate white-collar tasks & certain decision processes
Employ analytics to improve decision quality
Closing the Strategy Gap
Computerized decision support aims to reduce the discrepancy between current performance and desired performance (mission → objectives → goals → strategy)
Managerial Decision Making
Management Process & Resources
Management: using resources (inputs) to achieve organizational goals (outputs)
Success metric: \text{Efficiency} = \frac{\text{Outputs}}{\text{Inputs}}
Decision making = choosing the best alternative from \ge 2 options
Mintzberg’s 10 Managerial Roles
Interpersonal: Figurehead, Leader, Liaison
Informational: Monitor, Disseminator, Spokesperson
Decisional: Entrepreneur, Disturbance Handler, Resource Allocator, Negotiator
Four-Step Scientific Decision-Making Process
Define the problem/opportunity
Construct a model of the real world
Identify & evaluate potential solutions
Compare, choose, & recommend the best solution
Challenges Making Decisions Today
Explosion of alternatives (tech, IS, search engines, globalization)
Heightened uncertainty (regulation, compliance, instability, terrorism, competition, shifting demand)
Need for speed, frequent changes, high cost of errors → limits trial-and-error learning
Computerized Decision Support Systems (DSS)
Why Use DSS?
Rapid computations & what-if analysis
Better communication & collaboration (groups/teams)
Higher individual & group productivity
Improved data management & access
Helps overcome human cognitive limits
Enhances decision quality & agility
Web-enabled: anytime, anywhere access
Early Decision Support Framework (Gorry & Scott-Morton, 1971)
Dimensions:
Type of Control: Operational ↔ Managerial ↔ Strategic
Type of Decision: Structured ↔ Semi-structured ↔ Unstructured
Nine cells → match examples (e.g., Order Entry = Operational/Structured; Mergers & Acquisitions = Strategic/Unstructured)
Simon’s Decision-Making Phases
Intelligence: environmental scanning, reports, queries, benchmarking → detect problems/opportunities
Design: creativity, generate & analyze alternatives
Choice: compare alternatives, select best
Implementation: deploy & act on chosen solution
Decision Support Systems—Definitions & Scope
Classical (1970s): interactive computer-based systems coupling data + models to aid unstructured problems (Gorry & Scott-Morton; Keen & Scott-Morton)
Recent: IS that support semi- or unstructured managerial activity (Turban 2007; Arnott & Pervan 2008)
Umbrella view: any computerized system aiding decision making (e.g., company-wide KM, functional DSS)
Narrow view: custom applications built for semi/unstructured problems with components: Data, Model, Knowledge, User, Interface (loosely coupled)
High-Level DSS Architecture
Data (internal & external)
Models (quantitative/qualitative)
Knowledge/Rules
User Interface (UI/API)
Major DSS Types
Model-oriented: leverages quantitative models (decision trees, multi-criteria, optimization)
Data-oriented: derives functionality from large structured databases (file drawers, MIS reports, data warehouses, EIS, spatial DSS)
Evolution toward Business Intelligence (BI)
Definition & Objectives
BI = umbrella of architectures, tools, databases, analytical methods, apps & methodologies aimed at delivering easy data/model access → convert data → information → knowledge → decisions → action
Brief History
1970s: MIS static reports
1980s: Executive Information Systems (EIS)
1990s: OLAP, multidimensional ad-hoc reports → term BI coined (Gartner)
2005+: AI & Data/Text Mining, web portals, dashboards
2010s+: ongoing developments (predictive & prescriptive analytics, mobile BI, cloud)
Evolution of BI Capabilities (chronology)
Query/reporting → ETL/metadata → data warehouses → OLAP → dashboards/cockpits → scorecards → portals → alerts → data/text mining → predictive analytics → workflow/broadcasting → spreadsheet integration
BI Architecture Components
Data Warehouse (DW) – well-organized historical repository
Business Analytics – tools for manipulating/mining/analyzing DW data
Business Performance Management (BPM) – monitor & analyze KPIs
User Interface – browser/portal/dashboard for easy access
High-Level BI Flow
Data Sources → ETL → DW (build, organize, summarize, standardize) → Analytics (manipulate, derive results) → BPM (strategy, performance) → UI (presentation); future: intelligent systems overlay
Styles of BI (MicroStrategy)
Report delivery & alerting
Enterprise reporting (dashboards/scorecards)
Cube (slice-and-dice) analysis
Ad-hoc queries
Statistics & data mining
Benefits of BI (Thompson 2004 survey)
Faster/more accurate reporting 81\%
Improved decision making 78\%
Better customer service 56\%
Increased revenue 49\%
Real-time corporate performance visualization
The DSS–BI Connection
Architectural Similarity: BI evolved from DSS foundations
Direct vs. Indirect Support: DSS = specific decision episodes; BI = information provisioning that indirectly supports many decisions
Orientation: BI leans executive/strategic (BPM, dashboards); DSS often analyst-centric
Build Approach: BI built with commercial tools/components; DSS frequently bespoke
Origins: DSS methods/tools mainly academic; BI tools from software vendors
Tool Overlap: data mining, predictive analytics common to both
Relationship Views:
DSS ⊂ BI (analytical component)
BI = special data-oriented DSS
DSS = original element in BI’s evolutionary chain
DSS–BI Pictorial Summary (Kopáčková & Škrobáčková 2006)
Model-driven DSS ↔ optimization/what-if, etc.
Data-driven DSS (DW, OLAP, DM)
Knowledge-driven DSS (Expert Systems)
Document-driven DSS (text mining)
Communication-driven DSS (email, IM, video-conference)
Collective tools feed modern Business Intelligence environments
Tutorial / Discussion Prompts
Analyze real-world DSS applications: components, functions
Observe organizational decision processes; map suitable DSS category
Compare DBMS vs MBMS (model base management system): similarities (storage, retrieval, management); differences (data vs. model focus, query languages, metadata)