Assignment #1

This assignment is composed of 4 problems, where each problem has sub-problems.

1. Problem 1
This problem is composed of six questions (a, b, c, d, e, f). For the system described below, answer those
questions.

<System>
Three types (gold, silver, and regular) of customers arrive at a single server John Jr burger shop where
they are served one customer at a time. Gold type customers have higher priority than Silver type
customers, and Silver type customers have higher priority than regular type customers. No pre-emption
occurs; in other words, if the server is serving a regular type customer when a gold type customer arrives,
the server must finish serving the regular customer first and then serve the gold type customer.

Customer Arrival Information

 Gold customers: Inter-arrival time of 50 minutes
 Silver customers: Inter-arrival time of 30 minutes
 Regular customers: Inter-arrival time of 10 minutes
 Assumption: At time “zero”, one of each type of customers arrives at the system; 1 Gold, 1
Silver, 1 Regular

Service Information
 It takes 20 minutes for all the customers

Queues and their capacity

 There are three queues (waiting lines) in the system, each of which is dedicated for Gold
customers, Silver customers, and Regular customers, respectively
 Assumption: their capacity is “Infinite”

Ranking rule in the Queue

 Customers are serviced in FCFS (first come first service) order.

Performance measures to be calculated (goals of simulation)

 Average times that each customer types (Gold, Silver, Regular) spend in the system
 Total number of customers (sum of Gold, Silver, and Regular) served at the end of the simulation
end time
 Average waiting times of each customer types (Gold, Silver, Regular) in the queue

Simulation replication (ending) time

 “100” minutes

(a)
For the system described above, define components for a discrete event simulation model. Components
should include 1) entities, 2) resources, 3) attributes of the entities, 4) variables, 5) statistical
accumulators (variables used to calculate statistics), 6) queues, 7) event types, and 8) conceptual logic
(pseudo code) for each event type.
 Entities:
  Resources:
 Attributes:
 Variables:

Statistical accumulators:
 Queues:
 Event types: type A, type B, type C …
 Event type A Logic {
 --- (pseudo code goes here)
 }
 Event type B Logic {
 --- (pseudo code goes here)
 }

Hint: You need to have three different arrival events for Gold customers, Silver customers, and Regular
customers;

(b)
Where in your simulation logic does it guarantee that the correct priority logic (gold customers are served
before silver customers; silver customers are served before regular customers) is implemented? Answer
very briefly.

(c)
Construct a table containing the sample path (which looks similar to Table 2-2 of the textbook or a blank
table shown below) of the simulation that you have developed conceptually in Problem (a). In Problem
1(a), performance measures to be calculated were given. In your table (sample path), you must include
only those variables, attributes, and statistical accumulators that are necessary to calculate the requested
performance measures (do not include unnecessary information).
 In Table below, you need to fill in only the first 5 rows for the “Attributes” column; and you can
leave them blank from the 6th row. However, you must fill in all the rows until the simulation end
time for the other columns (e.g. just finished event, variables, statistical accumulators, and event
calendar).
 Note that you do not have to actually build an Arena model.

PS: Feel free to use Excel.

(d) Based on your simulation results obtained in Problem 1(b), plot the number of customers in the queue
(Q(t)) and the machine status (B(t)) over time.
(e) What is the average numbers of customers in the Gold queue, Silver queue, and Regular queue,
respectively?

 Average numbers of customers in the Gold customer queue: ( )

 Average numbers of customers in the Silver customer queue: ( )
 Average numbers of customers in the Regular customer queue: ( )

(f)
What is the machine utilization (% of time when the machine is busy)?

2. Problem 2

The system considered in Problem 2 is almost same with the one considered in Problem 1, where
the only difference is that there is only one queue shared by all three types of customers. For this
system, answer the following two questions.

(a)
For the system, your conceptual logic (pseudo code) for each event type will be slightly different.
Show me your conceptual logic (pseudo code) for each event type.
 Event types: type A, type B, type C …
 Event type A Logic {
--- (pseudo code goes here)
}
 Event type B Logic {
--- (pseudo code goes here)
}
(b)
Where in your simulation logic does it guarantee that the correct priority logic (gold customers are served
before silver customers; silver customers are served before regular customers) is implemented? Answer
very briefly.

3. Problem 3

The system considered in Problem 3 is almost same with the one considered in Problem 1, where
the only difference is on the ¡§Customer Arrival Information¡¨ (see below). Note that we have
three explicit queues. For this system, answer the following one question.

Customer Arrival Information

 Customers: Inter-arrival time of 30 minutes
 Percentages of arrived customers: 20% for Gold customers, 30% for Silver customers,
50% for Regular customers
 Assumption: At time ¡§zero¡¨, one customer (whose type is not known at that moment)
arrives at the system
(a)
For the system, event types and your conceptual logic (pseudo code) for those event types will be slightly
different. Show me your conceptual logic (pseudo code) for each event type.
 Event types: type A, type B, type C …
 Event type A Logic {
--- (pseudo code goes here)
}
 Event type B Logic {
--- (pseudo code goes here)
}

Hint: You need to have only one arrival event applicable to all the customer types

4. Problem 4

Answer the following “TRUE” or “FALSE” questions.

(a) The above simulation models are stochastic model. 1) TRUE ( ) or 2) FALSE ( )

(b) The above simulation models are static model. 1) TRUE ( ) or 2) FALSE ( )

(c) We understand that the above problem could have been solved without using simulation.
Even if an analytical model can solve the above problem, we still must use simulation tool.

1) TRUE ( ) or 2) FALSE ( )

(d) A continuous modeling approach is preferred to a discrete event simulation modeling

approach for the above system. 1) TRUE ( ) or 2) FALSE ( )

5. Problem 5

(Chapter 2 lecture notes page 2)

A simulation of a similar burger shop is to be conducted by you, with the objective of minimizing the
customer waiting time by improving the customer flow. Provide three iterations, in increasing order of
complexity, of Step 1 and Step 2 in the above simulation process.
6. Problem 6

Assuming the customer waiting time D in a burger shop is determined by three variables A, B and C,
which is defined as D=(A-25B)/(2C). A, B and C are uncorrelated random variables. Variable A is normally
distributed with u=100 and σ2 =400. Variable B is discrete uniformly distributed with a probability
distribution given by p(b) = 1/5 with b=0,1,2,3 and 4. Variable C is distributed in accordance with the
following table:

How would you define all these three, A,B, C inside the simulation model? Explain.