SIMUL8 Corporation : Support : Survey On Warm Up, Run length and Replication Number

Warm Up, Run Length & Replication Number Survey

Warwick Business School and SIMUL8 Corporation are working together to create an automated tool that can analyze your simulation and provide recommendations on warm-up, run-length and number of replications. We would be grateful if you could complete this short survey to help us design a tool that best meets our customer's needs.

Name

optional

Company

optional

1. Do you know what a warm-up period is?

Yes

I've heard of it

2. How confident are you that you know when a warm-up should be used?

Totally confident

Fairly confident

Not confident

Have no idea

3. Do you use a warm-up period in your own simulations?

Yes

Sometimes when applicable

4. If your answer to question 3 was No, is this because:

warm-up is not applicable to your type of models

You don’t know what a warm-up period is

Not aware that you need a warm-up period

5. If you sometimes use a warm-up, how do you decide on the length of warm-up?

Statistical analysis

Use a graph

Guess

Rule of thumb (e.g. always use 1 day)

Other:

6. How do you decide on a run-length?

From context of the problem I want to solve

Statistical analysis

Guess

Rule of thumb (e.g. always use 1 week)

Other:

7. Do you use more than one run (replication) i.e. do you run a trial?

Yes

Sometimes

8. If so, do you use the default value set by the simulation software?

Yes

Sometimes

9. If you sometimes set the number of runs in a trial yourself, how do you decide on the number of runs to use?

Statistical analysis

Guess

Rule of thumb (e.g. always run 10)

Other:

10. Are you aware of the importance of using the correct:

Run Length:	Yes	No	Vaguely
Warm-up Period:	Yes	No	Vaguely
Number of runs in a trial:	Yes	No	Vaguely

11. If so, how did you become aware?

Degree

Simulation Training Course

Simulation Literature, (books, newsletters…)

Other:

12. Which of these statements do you think best describes a 95% confidence interval?

An interval where there is a 95% chance that the true mean lies within it.

The interval computed that, if the study was repeated multiple times, would contain the true mean 95% of the time.

A range of values that the mean estimated by the simulation could fall within.

13. Do you output and use confidence intervals for your simulation results?

Yes

Sometimes

14. If simulation software could provide recommendations for warm-up length, run-length and number of replications, would you make use of this facility?

Yes

Sometimes

15. Please give a brief summary of your simulation background and experience: e.g. O.R. degree, 5 years experience in consulting.

15. Any other comments?