SIE 431 Simulation Modeling and Analysis

Midterm Exam, May 15 2021

Name ___________________________

Instructions

1. This exam is closed book, closed notes, exam.

2. Show all your work in detail. Answers without supporting work will be down graded.

3. This exam is 8 pages including this cover sheet. Check to see whether you have all the pages.

Question Points

1 10
2 15
3 20
4 15
5 20
6 10
7 10

Total 100

1
1. (10 pts)

An overview of how a discrete event simulation works is depicted in Figure below. Answer the
following three questions.

(a) (4 pts)
What go to 1 and 2 in the “Main” console? Note that the sequence of answers is important.
(1. ); (2. )

(b) (2 pts)
What goes to 3 (name of the module/routine)?
(3. );

(c) (4 pts)
What are two major functions of the “Timing Routine”?

2
(4. )
(5. )
2. (15 pts)

A simulation consultant visited a local bank and collected 1000 data of customer’s service
time. The simulation consultant analyzed the collected data, and the summary is shown below.

Time Range 0~10 min 10~20 min 20~30 min 30~40 min
Probability 0 1/2 1/2 0

(a) (5 pts)
Construct a CDF based on the data shown in the above table. Make sure to show values in the
“Y-axis”.

(b) (5 pts)
Given the above data, the simulation consultant decided to use an empirical distribution for
the customer’s service time. Choose only one of the following syntax and fill it in with
appropriate parameters.

CONT ( )
DISC ( )

(c) (5 pts)
.
Using the above linear congruential generator and the empirical distribution that you have
derived in problem 2 (b), what are the service time of the first three customers? Assume your
seed value is 7. Show all the steps of your answers.

i Zi Ui Service Time

0 7 N/A N/A

3
 3
3. (20 pts)

(a) (5 pts)
An Arena model for M/M/1 priority system with one shared queue is shown below. Which
module is used to specify the priority logic between high parts and low parts (viz. high parts
enter the queue ahead of low parts)?

( ) Module

(b) (5 pts)
In Arena, what is the name of module in the “Elements” template which is used to specify the
current number of parts waiting in the queue?

( ) Module

(c) (5 pts)
An Arena model for M/M/1 priority system with two explicit queues is shown below. What is
the queueing rule that is used among high parts (e.g. low value first; first come first serve)?

Your Answer: ( )

4
(d) (5 pts)
In the Arena model in Problem 3(c), which module(s) is (are) related with “scheduling future
events in the event calendar? List all of them.

4. (15 pts) Answer the following questions

(a) (3 pts)
A stream in random number generation corresponds to this. What is this?

Your Answer: ( )

(b) (4 pts)
A graph shown below represents the queue length for MACHINE 1 over time obtained from a
simulation run for 10 time units. What is the average queue length for MACHINE 1?

Your Answer: ( )

(c) (4 pts)
There are two types of statistics in simulation. What is the name of the type which the statistic in

5
Problem 4(b) belongs to?

Your Answer: ( )

(d) (4 pts)
Three pieces of information are needed to specify an event in the event calendar. What are they?

( ) ( ) ( )

5. (20 pts)

(a) (10 pts)

(G_Arrive)
Check resource availability
If resource is idle
Make the resource status to busy
Schedule ( )
else
Increment the number of customer in the gold_queue by one
Schedule ( )

(S_Arrive)
Check resource availability
If resource is idle
Make the resource status to busy
Schedule ( )
else
Increment the number of customer in the silver_queue by one
Schedule ( )

(R_Arrive)
Check resource availability
If resource is idle
Make the resource status to busy
Schedule ( )
else
Increment the number of customer in the regular_queue by one
Schedule ( )

6
 (Depart)
Check the gold_queue; If there is any customer in the gold_queue
Schedule ( )
else
Check the silver_queue; If there is any customer in the silver_queue
Schedule ( )
else
Check the regular_queue; If there is any customer in the regular_queue
Schedule ( )
else
Make the resource status to idle
Schedule ( )

(b) (5 pts)
In Problem 5(a), which of the four events (“G_Arrive”, “S_Arrive”, “R_Arrive”, and “Depart”)
ensure the priority (gold customer > silver customer > regular customer) logic?

( ) Event

(c) (5 pts)
Currently, three arrival events (“G_Arrive”, “S_Arrive”, “R_Arrive”) are defined in Problem
5(a). If inter-arrival times have been collected for “customers” together with percentages of
gold, silver, and regular types, how many arrival event(s) is(are) needed?

Your Answer: ( )

6. (10 pts)

An Arena model is shown below. Note that ten entities are created at time 0, the inter-arrival
time is “infinite” (meaning ten entities are created only once), and the resource capacity is
“1”. Suppose that we run this model for one replication. I would like to know the system state
at the end of the simulation (replication) run. Fill in the table below with the status
information at the end of the simulation run. Note that there is no “Seize” block nor
“Release” block before or after the “Delay” block. Note also that “Count” block increases
“#_produced” variable by one when each entity passes by.

7
 Machine status
TNOW #_produced NQ(queue1)
(busy or idle)

7. (10 pts)

Let’s suppose we have performed data collection of the time that customers spend at a bank.
Sixteen (16) data have been collected, and they are shown in the following table.
0.1 0.3 0.4 0.5 0.9 1.4 1.8 1.9 2.3 2.4 2.6 2.7 2.8 3.1 3.8 3.9

Let’s also suppose that the Input Analyzer has suggested the linear PDF shown below as the
best-fitted distribution for the collected data. Using the “Chi-square test”, determine whether
the collected data are independent samples from this fitted distribution or not at level  = 0.05.
Let’s assume the number of interval (k) used in the “Chi-square test” is 4. Let’s also assume
that the range of the values is 0 to 4. Note: 23, 0.05 is 7.81. Note that the chi-square statistics is

defined by: 2 = , where Nj = number of Xi’s in the jth interval, and Ej =

expected number of Xi’s in the jth interval if the PDF was correct.

8
 Figure. Fitted linear distribution for Problem 7

i Nj Ej
1
2
3
4

Therefore, the collected data (choose one: 1) are, 2) are NOT)