3.

WAITING LINE MODELS OR QUEUING THEORY

LEARNING OBJECTIVES:

After studying the chapter, you should be able to:

Understand the concept of queue/waiting line
Know the application areas of queuing model
Understand the general structure and operation characteristics queuing system
Understand the assumptions of single channel Queuing model
Know the infinite and finite source Queuing Model

INTRODUCTION:

A group or individual items waiting to receive service are known as queue. Queues or waiting
lines are in every day life. Waiting exists when a unit that is to be serviced in some way has to
wait for the service as the service facility is giving service to other units.
The unit joins a queue in which it awaits its turn to be serviced by the service facility.

APPLICATIONS:
Queuing theory / waiting lines / congestion process is the study of the randomly fluctuating
waiting lines that occur in the situations like:
General applications of Queuing theory are as follows:
i) People arriving at cinema ticket window
ii) People arriving at barber shops
iii) Letters arriving at typist’s desk
iv) Machines coming for repair / machine parts to be assembled
v) Ships to be loaded / Unloaded
vi) Cars waiting for traffic lights to turn green
vii) Cars waiting for service at petrol bunks
viii) Workmen waiting for tools
ix) Jobs waiting for processing
x) Calls arriving at a telephone switch board
 xi) Customers waiting for service in supermarket

GENERAL SRUCTURE OF A QUEUING SYSTEM:

A queue system consists of a group of customers waiting to be served and also the customer
being served at particular time. Variables affecting the system are:
i) Arrival pattern or process
ii) Service pattern or process
iii) Number of service counters
iv) Queue discipline
v) Customer behaviour and
vi) Server behaviour
A queue model is developed considering all these variables.

QUEUE DISCIPLINE
The order is which the customers are selected from the queue for the service is known as queue
discipline. This can be one of the following types.
a. First in First Out (FIFO): Under FIFS arrangement selection of customers for service is
done in the order, which they arrive and form the queue.
b. Last in First Out (LIFO): Under LIFS arrangement selection of customers for service is
done in the reverse order in which they arrive and the person entering last is the first to be
selected for service. It is not an appropriate method. Such practice is common in the case
of issue of stores, as the storekeeper finds it more convenient to move the items in LIFS
order while issuing for job.
c. Service in Random Order (SIRO): In case of service in random order customers are
selected at random out of those present at the service station at a particular time.
d. Service with Priority (SWP): Under such arrangement certain customers are given
priorities over the others in selection for the service. The priority can be two types:
i. Non-pre-emptive priority: Where the customer already getting served is allowed
to continue with the service till it is completed even when a priority customer
arrive midway during service.
 ii. Pre-emptive priority: Where service to Non-priority customer stopped as soon as
the priority customer arrives. This is in case of when some critical machine breaks
down in factory, repair of these critical machines is given top priority even by
stopping the repair of less important machines.
Which particular queue system will be followed at a service station will depend upon
the management policy.
e. Shortest processing Time (SPT): Customer with the shortest processing time is selected
first and served.
CUSTOMER BEHAVIOUR
There are two distinct classes based on the size of the customer population
i) Number of customers is very large and the demand correspondingly small, it is
convenient to assume that the customer population is infinite.
ii) If the number of customers is small, the finite case must be used. In general if the size
of the calling customer population for service is less than 250, the finite case will be
used.
Various customers while in queue behave differently. Their behaviour pattern can fall in one of
the following categories.
i) Balking: The customers are said to balk if they do not join a queue because of their
reluctance to wait.
ii) Collusion: One person would join the queue and demand service for several customers.
iii) Jockeying: Customers keep on switching from one queue to another in shorter length in
multiple services.
iv) Reneging: impatient customers who could not wait beyond a certain time and leave the
queue.
Server behaviour although the service timings might have been specified, the server may not be
available through the entire span of time. He may leave the service center for attending to his
personal needs.

QUEUEING THEORY TERMINOLOGY & CHARACTERISTICS:

i. Customers: A unit coming for service to the service station is known as Customer. This
can be a person, machine, telephone call, or demand for some commodity.
 ii. Queue or Waiting Line: A line formed by customers waiting to get services is known as
Queue or Waiting Line. The customer already getting the service shall not be included in
queue.
iii. Service Channel: A system or Channel consisting of service points, which provide
service to the incoming customer, is known as Service Channel.
iv. Arrival Rate: The rate at which the customer arrives at the service center for service per
unit of time is known as Arrival Rate. It is denoted by Lamda (λ). The arrival rate is
calculated by dividing the total number of arrival by total units of time. If the rate of
arrival is in Poisson distribution then the distribution of time between two arrivals is
exponential. Thus, the arrival rate is the mean of time intervals will be 1/λ.
v. Service Rate: The rate at which service is provided at the service center, per unit of time
is known as Service Rate. It is denoted by µ (Mue). This is calculated by dividing the
total number of customers serves by total units of time. Service rate should be grater than
the arrival rate. Otherwise the queue will never terminate.
vi. Traffic Intensity or Utilization Rate: The rate at which the available service facility is
utilized is known as Traffic Intensity or Utilization Rate. It is denoted by ρ.
a. If ρ < 1 it indicates that the service facility not fully utilized.
b. If ρ is nearer to 1 that will mean that the ρ is very high and consequently the
waiting time will be more.
c. If ρ > 1 it indicates that queue will never terminated.
vii. Idle Rate: The rate at which service facility remains utilized is known as Idle Rate. Idle
rate is denoted by ρ0.
viii. Expected number of customers in the system (Ls): Expected number of customers in
the system equal to the number of customers in the queue plus those getting the service. It
is denoted by Ls.
ix. Expected number of customers in the Queue or Waiting Line (Lq): Expected number
of customers in the queue is equal to the expected number of customers in the system
minus those who are getting the service. This is known as Queue length and denoted by
Lq .
 x. Expected Time Spent By Customer in System or Waiting Time in the System (Ws):
The expected time spent by customer in system is equal to the expected time while
waiting for service plus service time.
The derivation of these formulae is a mathematical task and beyond the scope of this
book. As such that aspect has not been dealt in detail.
xi. Expected Waiting Time in Queue (Wq): The expected time spent by the customer in the
queue is equal to the expected time spent by the customer in the system minus time spent
at the service station.
xii. Probability of Zero Customers Waiting (P0): The probability of zero customers
waiting means a customer arriving for service does not have to wait in queue or gets
immediate service. This is depend upon the traffic intensity or the idle rate of the system.
According to the probability of zero customer waiting is determined by the equation
P0 = 1- ρ