SIMULATION OF QUEUING

SYSTEMS
 QUEUING SYSTEMS

• Whether it is a bank, or a theater or waiting for a bus, we

find queues everywhere in our day to day life.

• Theory of queuing is to sort out such problems. Agner Krarup

Erlang, a Danish engineer who worked for the Copenhagen
Telephone Exchange, published the first paper on queuing
theory in 1909.

• A queuing system involves customers arriving at a constant

or variable time rate for service at a service station.

• Customers can be students waiting for registration in college,

aeroplane queuing for landing at airfield, or jobs waiting in
machines shop.
 QUEUING SYSTEMS

• The basic concept of queuing theory is the optimization of wait time, queue length and the service
available to those standing in a queue.

• Cost is one of the important factors in the queuing problem.

• Waiting in queues incur cost, whether human are waiting for services or machines waiting in a machine
shop.

• On the other hand if service counter is waiting for customers that also involves cost. In order to reduce
queue length, extra service centers are to be provided but for extra service centers, cost of service
becomes higher.

• On the other hand excessive wait time in queues is a loss of customer time and hence loss of customer
to the service station.

• Ideal condition in any service center is that there should not be any queue. But on the other hand
service counter should also be not idle for long time.

• Optimization of queue length and wait time is the objective of queuing

SYMBOLS USED
 Elements of a Queuing System
1. Calling source. or the population from which customers are
drawn.
Calling source may be finite or infinite.
When queue is so long that arrival of one more customer does
not effect the queue length, we call it infinite source of
customers.
A reverse of this situation, when queue length is not long and
incoming or outgoing of one-customer affects the queue; we
call it a finite source of customers.
2. The input or arrival process. This includes the distribution of
number of arrivals per unit of time, the number of queues that
are permitted to be formed, the maximum queue length and
the maximum number of customers desiring service.
3. The service process. This includes time allotted to serve a
customer, number of servers and arrangement of servers.
 KENDALL’S NOTATION
• We will be frequently using notation for queuing system, called Kendall’s notation,
that is, V/W/X/Y/Z,
V arrival pattern,
W service pattern,
X number of servers,
Y system capacity
Z queue discipline.

• The symbols used for the probability distribution for inter arrival time and service
time are, D for deterministic, M for exponential.
• Similarly FIFO (First in First out), LIFO (Last in First out), etc., for queue discipline.
• If the capacity Y is not specified, it is taken as infinity, and if queue discipline is not
specified, it is FIFO (First in First Out).
• For example M/D/2/5/FIFO stands for a queuing system having exponential arrival
times, deterministic service time, 2 servers, capacity of 5 customers, and first in first
out discipline.
If notation is given as M/D/2 means exponential arrival time, deterministic service
time, 2 servers, infinite service capacity, and FIFO queue discipline.
 PRINCIPLE OF QUEUING THEORY
The operating characteristics of queuing systems are determined largely by two statistical properties,
i) probability distribution of inter arrival times
ii) probability distribution of service times.

For the case of simplicity, we will assume for the time being, that there is single queue and only one
server serving the customers.
We make the following assumptions.
• First-in, First-out (FIFO): Service is provided on the first come, first served basis.
• Random: Arrivals of customers is completely random but at a certain arrival rate.
• Steady state: The queuing system is at a steady state condition.
The above conditions are very ideal conditions for any queuing system and assumptions are made
to model the situation mathematically.

First condition only means irrespective of customer, one who comes first is attended first and no
priority is given to anyone.
Second conditions says that arrival of a customer is random and is expected anytime after the elapse
of first mean time of interval(τ say).
In a given interval of time (called mean time of arrival τ, between two customers) only one
customer is expected to come. This is equivalent to saying that the number of arrivals per unit time
is a random variable with a Poisson’s distribution.
 Poisson’s Arrival pattern.
• Number of arrivals per unit time is a random variable with a Poisson’s
distribution
 QUEUE LENGTH
• Within the single server and multiple servers there are two
categories queue length
1. Finite queue length
2. Infinite queue length

The infinite queue length model assumes, that every person

who comes joins the line. There is no restriction on the
number of people who are actually waiting or there is no
restriction on the length of the queue.

In finite queue length models we try to restrict the queue

length to a certain limit after which we say that if this
threshold limit is reached, people who come into the system
do not join the system. A good example of a finite queue Car
repairing garage
 Population
• Finite population models: If we take the
example of the doctor or the reservation system
or the car mechanic they all come under the
category of what are called infinite population
models
• Infinite population models: If we have a factory
and if this factory has about thirty machines
and if there is a dedicated maintenance team,
this maintenance team attends to calls every
time these machines breakdown
Relationship between Lambda and Mu
• Lambda usually denotes the arrival rate in a queuing system
• Mu denotes the service times in a queuing system
• The value of lambda by mu is less than 1, particularly, if we have an
infinite queue length models.
• For example, we assume lambda equal to 5 per hour and mu is
equal to 6 per hour, then lambda by mu is less than 1.
• If mu were not 6 per hour and mu were 4 per hour. Then, every
hour five people on an average enter the system and four people on
an average leave the system, so, the queue length will automatically
increase by 1 every hour and this means that somebody who joins
the queue will never get served.
• For a queuing system to be efficient, particularly, when we have
infinite queue length, lambda by mu has to be less than 1
 Key terms :
• Balking:If an arrival does not join the system and leave, then the arrival is
set to balk. The person balks when the person does not join the line and
leaves.
Balking can again be of two types: Forced balking and an unforced balking

• Reneging: The person joins the system, but after some time decides not to
continue and simply moves out of the system, called reneging

• Jockeying: There will be a tendency to shift from one line to another line
for a while and come back depending on what we think which is the rate
at which people in these lines are moving. Such a
phenomenon is called jockeying.
QUEUING ARRIVAL-SERVICE MODEL