4. Simulation

Definition of Simulation

  • A simulation is a computer-based model used to run experiments on a real system.

  • Typically done on a computer.

  • Determines reactions to different operating rules or changes in structure.

  • Can be used in conjunction with traditional statistical and management science techniques.

Major Phases in a Simulation Study

  1. Define Problem: Clearly identify the problem that the simulation aims to address.

  2. Construct Simulation Model: Develop the model that represents the real system.

  3. Specify Values of Variables and Parameters: Determine the values for the variables and parameters used in the simulation.

  4. Run the Simulation: Execute the simulation model, typically using a computer.

  5. Evaluate Results: Analyze the simulation results to draw conclusions.

  6. Validation: Ensure that the simulation model is correct and representative of the real system.

  7. Propose New Experiment: Based on the initial simulation results, propose new experiments to further investigate the problem.

  8. Stop: Conclude the simulation study.

Simulation Methodology: Problem Definition

  • Specifying the objectives of the simulation.

  • Identifying the relevant controllable and uncontrollable variables of the system to be studied.

Constructing a Simulation Model

  • Specification of Variables and Parameters

  • Specification of Decision Rules

  • Specification of Probability Distributions

  • Specification of Time-Incrementing Procedure

Data Collection & Random Number Interval Example

Specify Values of Variables and Parameters

  • Determination of starting conditions.

  • Determination of run length.

Run the Simulation

  • By computer

  • Manually

Evaluate Results

  • Conclusions depend on:

    • The degree to which the model reflects the real system.

    • Design of the simulation (in a statistical sense).

  • The only true test of a simulation is how well the real system performs after the results of the study have been implemented.

Validation

  • Refers to testing the computer program to ensure that the simulation is correct.

  • To ensure that the model results are representative of the real-world system they seek to model.

Proposing a New Experiment

  • Consider changing many of the factors:

    • Parameters

    • Variables

    • Decision rules

    • Starting conditions

    • Run length

  • If the initial rules led to poor results, or if these runs yielded new insights into the problem, then a new decision rule may be worth trying.