Computer Simulation

1
Computer Simulation

• A Laboratory to Evaluate
• What-if Questions

2
Nature of Computer Simulation

• A system is a combination of elements that
  interact to accomplish an objective.
• A computer simulation model generates estimates
  of system performance.
• An animated simulation allows one to view the
  system activity in accelerated time.
• Discrete event simulation is driven by events
  that occur at certain points in time.
• Systems are both dynamic and stochastic.

3
Process of System Simulation

• Define problem and objectives
• Collect empirical data to determine probability
  distributions
• Begin model development with a process flowchart
• Verify that the model works as intended
• Validate that the model predicts system
  performance based on historical data
• Run scenarios (What ifs) one at a time under
  controlled conditions
• Document and prepare graphic presentations
• Implementation of results

4
Monte Carlo Simulation

• Use random numbers (RN) to generate probabilistic
  events.
• Note that RN is uniformly distributed on the
  interval 0 to 1.
• Random variable are either discrete or continuous

5
GeneratingDiscrete Random Variable
Days x Prob. p(x) Cum. Prob. F(x) RN Assignment
1 0.1 0.1 0.0RNlt0.1
2 0.5 0.6 0.1RNlt0.6
3 0.4 1.0 0.6RNlt1.0
Example of Lead Time Delay for Order
6
Generating Continuous Random Variable

• Uniform Distribution
• Negative Exponential Distribution
• Normal Distribution