CS620 Midterm Modules 1-125 by Ali Abbas

CS620 – Modelling and Simulation

1. Simulation: Simulation is the imitation of the operation of a real-world system or process to study its behavior under different scenarios.
2. When Simulation is Appropriate: Used when experimentation with the real system is difficult, dangerous, or expensive.
3. When Simulation is Not Appropriate: Not suitable when simple analytical solutions or direct calculations are possible.
4. Advantages of Simulation: Helps analyze complex systems, test 'what-if' scenarios, and train people without real-world risks.
5. Disadvantages of Simulation: Can be expensive, time-consuming, and results are only as good as the model's accuracy.
6. Application Areas: Manufacturing, healthcare, military, supply chain, traffic systems, business processes.
7. Military Applications: Used for training, logistics, battle planning, and weapon system evaluation.
8. System & Environment: System: focus area; Environment: external factors affecting the system.
9. System Concepts: Includes homeostasis, adaptation, feedback loop, throughput, etc.
10. Components of a System: Entity, Attribute, State, Activity, Event, and Delay.
11. Discrete vs Continuous: Discrete: changes at specific times (bank queue). Continuous: smooth change (water level in tank).
12. Model of a System: A simplified representation of a real-world system to study it.
13. Types of Models: Static/Dynamic, Deterministic/Stochastic, Discrete/Continuous.
14. Discrete Event Simulation: Simulation where state changes at discrete points using event list.

15–22. Steps of Simulation Study: Planning → Model → Data → Verification → Experiment → Reporting → Implementation.

23. Simulation Case Study Steps: Define goal → Build model → Validate → Run → Analyze.

24–27. Spreadsheet Simulations: Use RAND(), simulate randomness (coin toss, arrivals, service time).

28–29. Simulation Table: Organize input, random data, logic, and result columns in table form.

30–34. Discrete Event Framework: Defines Clock, Events, Activities, State, and Entity interactions.

35–38. Event Scheduling Algorithm: 3 phases: A (remove event), B (execute), C (trigger new events).

39–40. World Views: Three views: Event Scheduling, Activity Scanning, Process Interaction.

41–43. Manual Simulation: Simulation done step-by-step using paper or spreadsheet.

44. Statistical Models: Use distributions (Poisson, Exponential, Normal) to model randomness.

45–48. Review Terms: Entity, Attribute, State, Activity, Event, Delay, Clock, System, Environment.

49–50. Useful Models: Queueing and inventory simulation models used in business and industry.

51–55. Queuing Models: Used to simulate lines of people, items, or requests waiting for service. Includes arrival rate, service rate, queue discipline.

56–60. Inventory Systems: Simulate ordering, stocking, and selling products. Helps find optimal stock level and reorder policies.

61–63. Delay in Simulation: Delays occur when entities must wait for a resource. Modeled using probability or event dependencies.

64–66. Activity vs. Event: Activity: takes time (e.g., service). Event: instantaneous change (e.g., arrival, departure).

67–70. Simulation Clock: Keeps track of simulation time. It jumps to next event time in discrete event simulation.

71–73. Entity Attributes: Attributes define properties like arrival time, priority, or service type for each entity.

74–76. Future Event List (FEL): A time-ordered list of all scheduled events in the system. Drives simulation progress.

77–79. Random Numbers: Use uniform distributions (0–1) with functions like RAND(). Convert to specific distributions for modeling. 80–83. Probability Distributions: Exponential, Poisson, Normal – used to simulate arrival, service, and failure times.

84–85. Validation Techniques: Compare model results with real-world data, use expert opinion, or run statistical tests.

86–88. Verification Techniques: Ensure model logic is implemented correctly. Use walkthroughs, trace events, and code checks.

89–91. Animation in Simulation: Visualization of system behavior (e.g., moving cars, customer flow). Enhances understanding and debugging.

92–94. Statistical Output Analysis: Analyzing results from multiple runs: mean, variance, confidence intervals.

95–96. Terminating vs Non-Terminating Systems: Terminating: simulation ends (e.g., 8-hour shift). Non-Terminating: runs indefinitely (e.g., ATM).

97–98. Warm-up Period: Ignore initial output to allow the system to reach a stable state. Helps avoid bias.

99–100. Use of NetLogo: NetLogo is a tool for agent-based simulation. Used for modeling decentralized systems (e.g., ant colonies, fire spread).

Topic-wise Notes (101–125)

101–103. Agent-Based Modeling (ABM)

ABM is a simulation method that models the behavior of individual agents (like people, animals, or machines) that act independently and interact with each other to create system-level behavior.

104–106. Characteristics of ABM

Agents act autonomously Local rules, no central control Agents may adapt, learn, and evolve Collective (emergent) behavior arises from interactions

107–108. Benefits of ABM

Models individual behaviors Captures complex interactions Visual and flexible Useful when system behavior is not easily predictable

109–110. Applications of ABM

Used in:

Biology (cells, immune systems) Sociology (crowd behavior)