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
Define Problem: Clearly identify the problem that the simulation aims to address.
Construct Simulation Model: Develop the model that represents the real system.
Specify Values of Variables and Parameters: Determine the values for the variables and parameters used in the simulation.
Run the Simulation: Execute the simulation model, typically using a computer.
Evaluate Results: Analyze the simulation results to draw conclusions.
Validation: Ensure that the simulation model is correct and representative of the real system.
Propose New Experiment: Based on the initial simulation results, propose new experiments to further investigate the problem.
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.