System Analysis and Modeling Notes

Model: A representation of a system for studying its functions and characteristics.
Purpose of modeling includes:
- Understanding
- Learning
- Improvement
- Optimization
- Decision Making

Models are simplifications and abstractions of reality.
They represent an object or idea in a different form than the entity itself.
Composed of symbols and ideas that represent the functional relationships of elements in a system.

Models can be categorized into three levels:
- Instances: Specific examples of a concept.
- Application Classes: Groups that share common traits or functions.
- Meta Classes: Higher-level frameworks that define the structure and rules governing instances and application classes.

Subject System: The system being studied, which requires information gathering.
Usage System: Interactions and contracts related to the subject system.
Development System: System that maintains information about the model being built.

System Object: The main focus within the system.
Property of an Entity: Variables that describe the system's state during a study.
Law: An action occurring over a specified period impacting the system's state.
Instantaneous Occurrence: An event that can change the state of the system instantly.

Iconic Models: Physical representations scaled from real-world dimensions.
Analog Models: Use alternative properties to represent aspects of the original system (e.g., electric resistance modeling fluid friction).
Stochastic Models: Incorporate randomness and probabilities (e.g., weather patterns).
Deterministic Models: Use explicit relationships without randomness (e.g., mathematical equations).
Discrete Models: State variables that change at discrete time intervals.
Continuous Models: State variables change continuously over time.
Combined Models: Merging discrete and continuous state variables (e.g., crop growth and harvest timing).
Mathematical Models: Abstract forms written in equations.
Object-oriented Models: Use real-world object abstractions and relationships.
Heuristic Models: Use rules of thumb for modeling system attributes.

Conceptual Level: High-level understanding of the model’s scope.
Specification Level: Detailed description involving state variables.
Computational Level: Implementation details, potentially including pseudocode or program structure.

Correctness: The model must accurately represent the system.
Relevance: The model should focus on significant aspects.
Cost vs. Benefit: Assessing resource expenditures against outcomes.
Clarity: The model's purpose and function must be easily understandable.
Comparability: Results should be comparable across different models.
Systematic Structure: Organization within the model should be logical and systematic.

Formulate scenarios with stakeholder input.
Extract use cases from scenarios.
Analyze the flow of events.
Generate class diagrams, including class identification, attributes, operations, and associations.
Establish constraints and multiplicities between classes.

Models help explain phenomena that can't be directly experienced. They are vital in both research and communication in scientific contexts.

Relates to models that incorporate time-dependent elements, facilitating interactions over time.
Updates can include new data allowing for ongoing refinement of the model.

Inputs: Land, people, capital, information.
Transformation: Processes that add value, e.g., turning raw materials into products.
Outputs: Goods and services.
Feedback Control: Measurement of outputs against standards for corrective actions.

Models can be hierarchical, where systems encompass various subsystems and elemental parts, indicating a structured organization.

Experiments with the Actual System: Testing in the real world for data collection.
Experiments with a Model: Physical, mathematical, and simulation models can be utilized to predict behavior and interactions.

Stakeholder goals, behaviors, structures, properties, and interconnections must all be accounted for in models.

Advantages: Cost-effective, faster to build, lesser risks, and informative for system behaviors.
Disadvantages: Relying on assumptions that might be incorrect and requiring abstract understanding which can lead to inaccuracies.

Products: Tangible, durable, can be inventoried, longer consumption time, lower customer involvement.
Services: Intangible, perishable, immediate consumption, high customer involvement.

They provide simplifications for real-world study, facilitate discussions, enhance understanding, and enable predictions around behavior and problems.

End of Notes