Advantages of Mathematical Models in Decision Making

Data Repositories and Business Intelligence Architectures

• The primary repository of data for business intelligence analysis is the data warehouse.

Decision Support Systems (DSS)

• The model management module in a DSS provides end-users with mathematical models derived from operations research, statistics, and financial analysis.
• The model management module assists knowledge workers in a DSS by enabling them to build, store, and manage mathematical models.

Decision-Making Process

• A decision should abide by ethical standards, in addition to being compliant with the law.
• During the exclusion stage, decision-makers identify and eliminate unsuitable options.
• The main purpose of evaluating feasible alternatives is to select the best option.
• A decision may be unworkable despite being ideal from economic, legal, and social standpoints due to feasibility issues.
• A decision maker should assess the decision's impact on the organization, besides economic, legal, and social factors.
• The major methodological reference proposed in the early 1960s regarding the decision-making process is the rational model.

Mathematical Models

• One advantage of adopting mathematical models in the decision-making process is that they enhance knowledge preservation within an organization.
• Building a mathematical model contributes to knowledge preservation by allowing knowledge workers to document and share their knowledge.

Business Intelligence Systems Architecture

• The primary task related to data sources in a business intelligence system architecture is to integrate data from various sources.
• Data integration techniques include data federation, ETL (Extract, Transform, Load), and data virtualization.
• Correcting data in a data warehouse affects business intelligence systems by ensuring data quality and reliability.
• Typical data sources in a business intelligence system architecture include OLTP (Online Transactional Processing) systems, external sources, and front-office systems.
• Complex queries for OLAP (Online Analytical Processing) analyses are directed to the data warehouse, not operational systems, to minimize the load on operational systems.

Data Warehouse

• Physically separating the data warehouse from operational systems enhances system performance and improves data quality.
• Extending the time span of data in a data warehouse is important for analyzing trends and predicting future outcomes.
• Semantic homogeneity plays a crucial role in data integration within a data warehouse by ensuring consistency in data definitions and format.

Data Sources

• Data from back-office systems usually originates from transactional systems, such as ERP (Enterprise Resource Planning) systems.
• Data originating from call-center activities and customer assistance comes from CRM (Customer Relationship Management) systems.
• Geographic information systems (GIS) provide spatial data as a significant source of external data.
• Front-office systems typically include data from sales, marketing, and customer-facing activities.
• External sources like agencies provide demographic, economic, and environmental data to extend internal databases.
• Back-office systems primarily focus on internal operations, whereas front-office systems focus on customer-facing activities.

Explore the benefits of using mathematical models in decision making processes, beyond just improving effectiveness. Learn how abstract models can deepen understanding of the analyzed domain and lead to long-term advantages.