1
Q
what is output data analysis?
A
is the analysis of data generated by a simulation run to predict system performance or compare performance of two or more system designs.
- Output data from simulation is random
* Results may differ greatly from the true characteristics of the system
* Estimate a measure of performance of a simulated system
* Use a confidence interval
2
Q
what are the two types of simulations:
A
- Terminating simulation: runs for a duration of time
eg, bank (terminating event).
- A terminating simulation is one in which the simulation starts at a defined state or time and ends when it reaches some other defined state or time. - Steady-state simulation: runs over a long period of time
eg computer/telecom networks, climate
- A steady-state simulation is used to investigate the steady-state behavior of a system, where the simulation continues until the system reaches a steady-state.
3
Q
what are the issues with output data analysis? 1
A
- Initial / Starting Conditions: Initial observations may not be good estimations of the true conditions of the system, Transient/warm-up period where you either impose initial conditions or
run simulation for a while and discard initial portion
Eg
* A heavily congested system manufacturing system
4
Q
what are the issues with output data analysis? 2
A
- Auto-correlated / Dependent Data: Dependence between events, need to ensure independence
- Classical statistical methods are not directly applicable.
- A queuing system and the waiting time of a customer.
- An inventory system and the amount of beginning inventory on a given day.
5
Q
how do we compare alternative system configurations?
A
- Output from several different simulation models that might represent competing system designs or alternative operating policies
- Very important as the real utility of simulation lies
in comparing such alternatives before implementation in practice
