lOMoARcPSD|21942857
Output Analysis Practices
Simulation Models in Industrial Engineering (International University - VNU-HCM)
Scan to open on Studocu
Studocu is not sponsored or endorsed by any college or university
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
OUTPUT ANALYSIS: PRACTICE EXERCISES
Question 1: Estimation of absolute and relative performance
Output analysis is the examination of data generated by a simulation. What are the
purposes of estimating absolute performance and relative performance when doing
output analysis?
Question 2: Absolute measures
Compute a measure of error (95% confidence interval) and a measure of risk (95%
prediction interval) for a production system. Given the overall average of the daily
average production time on 60 replications of the production system simulation is 4.5
hours, with a sample standard deviation of 1.2 hours. Make a promise on the long-run
average of the daily average production time and the average production time on a
particular day.
Question 3: Type of simulations with respect to output analysis
What are terminating simulation and steady-state simulation? Explain the purposes of
each type of simulations and provide each of them with a corresponding example.
Question 4: Terminating simulations (statistical background)
Given results of six independent replications of a convenient store. Compute an overall
point estimator, its estimated variance, and the 95% confidence interval half-width for the
mean queuing time.
Replication Average Queuing Time
1 0.65
2 1.23
3 1.05
4 0.82
5 1.64
6 1.86
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
Question 5: Steady-state simulation: confidence intervals with specified precision
Given results of 10 replications of a service system as follows:
Replication Average Service Time (minutes)
1 15.13
2 12.50
3 21.05
4 26.31
5 18.42
6 19.67
7 20.75
8 17.83
9 23.17
10 21.46
Consider the above data as an initial sample of size R0 = 10. Suppose that it is desired to
estimate long-run average service time to within ϵ = 1 minutes with 95% confidence.
Compute the initial estimate of the number of replications needed to achieve the desired
precision.
Question 6: Comparison of system designs
Perform a statistically valid comparison of two system designs using common random
numbers, given alpha = 5%
Response Time for System Design (hours)
Replication Y1 Y2
1 0.3561 0.3997
2 0.3741 0.4148
3 0.3823 0.4473
4 0.3990 0.4606
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
5 0.3790 0.3872
6 0.3640 0.4450
7 0.4039 0.4748
8 0.3903 0.4645
9 0.3888 0.3963
10 0.4184 0.4058
Question 7: Arena report
Suppose system output of simulation is as follows. Compute:
a) The standard deviation from the 95% C.I. half-width
b) The 90% C.I. half-width
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
APPENDIX
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
lOMoARcPSD|21942857
Downloaded by Jamy Nguy?n (tramttn0287@gmail.com)
