Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

QUEUING THEORY (WAITING LINE THEORY)

Queuing theory is a branch of mathematics that studies and models the act of waiting
in lines. Waiting lines are the most frequently encountered problem in our daily life.
Queuing theory owes its development to A.K. Erlang’s efforts to analyze telephone
traffic congestion with a view to satisfying the randomly arising demand for services
of the Copenhagen Automatic telephone system in the year 1909. The theory is
applicable to situations where customers arrive at some service stations for some
service; wait (occasionally not); then leave the system after getting the service.
A flow of customers from finite/infinite population towards the service facility forms
a queue (waiting line) on account of lack of capability to serve them all at a time.
Queues may consist of;
 Customers buying milk & other milk products at a milk Parlour
 Machines waiting to be repaired
 Trucks/vehicles waiting at the milk plant
 Patients in a hospital who need treatment,
 Customers waiting for bank teller services at the bank, etc.
1. In general, a queue is formed when either customers, wait for service or service
facilities stand idle & wait for customers. This comes with costs such as;
Waiting costs i.e. costs incurred by customers waiting on the lin. These costs
however reduce as the service level increases.
2. Service costs, i.e. costs incurred when the customer is being attended at the service
facility. These costs increase as the service level increase. Thus, total costs in
queuing theory equals to waiting costs plus service costs.
Note; queuing models are basically relevant to service oriented organizations and
suggest ways and means to improve the efficiency of the service.
Queuing System
This is composed of customer arriving for service, wait for service if it is not immediate
and having been served, they leave the system.
Lq. Ls Service Facility
Served Customer
Arriving Cust.
Components of a queuing System
The following are some of the components that describe a queuing system;
Page 1 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

1. Input process/arrival pattern/calling population; this is a pattern in which

customers arrive and join the system. An input source is characterized by;
 Size of the calling population
 Pattern of arrivals at the system
Customers requiring a service are generated at different times by an input source and
the rate at which they arrive at the system facility is determined by the arrival process.
 Size of the calling population; this is a total number of potential customers who will
require service. The source of customers may either be finite/infinite. Infinite, is
where the number of people is very large e.g. families visiting restaurants, ships
discharging cargo at a dock, etc. whereas, finite is where the number of people
arriving for service is small i.e. customers may arrive for a service individually or in
groups e.g. students arriving at a library counter.
 Pattern of arrivals at the system; customers may arrive in the system facility according
to some certainty & known schedule (deterministic models) & are easier to handle
e.g. one patient every 15 minutes, a candidate for interview every half an hour. On
the other hand, random arrivals (stochastic arrivals), is where customers’ arrivals
are independent of one another and their occurrence can’t be predicted exactly.
2. Service mechanism (service pattern); the service is provided by a service facility (s)
which may be a person (a bank teller, a barber, a machine (elevator, gasoline
pump), or a space (airport runway, parking lot, hospital bed), etc. a service facility
may include one or several people operating as a team.
Aspects of a service system
There are two aspects of a service system, i.e. configuration of the service system and
the speed of the system.
a) Configuration of the service system; this how the service facilities exist. Customer entry
into the service system depends upon the queue conditions. If at the time of
customer’s arrival, the server is idle, the customer will be served immediately.
Otherwise, the customer will be asked to join the queue, which can have several
configurations.
Classification of service systems
Service systems are usually classified in terms of their number of channels or servers,
which may include;

Page 2 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

 Single server (single queue); is a model that involves one queue and one service station
facility where customers wait till the service point is ready to him/her for servicing,
e.g. students arriving at a library counter.
 Single server (several queues); this consists of several queues and the customer may
join any one of the lines but there is only one service channel.
 Several (parallel) servers (single queue); in this type of model, there is more than one
server and each server provides the same type of facility. The customers wait in a
single queue until one of the service channels is ready to take them for servicing.
 Several servers (several queues); this model consists of several servers where each of
the servers has a different queue. E.g. electricity office with different cash counters
where customers can make payments in respect of their electricity bills, different
ticket issue counters in a trade fare, different boarding pass counters at an airport,
etc.
 Service facilities in a series; in this model, a customer enters the first facility and gets
a portion of service & then moves on to the next facility, gets a service & then
moves on to the next facility, etc. & then leaves the system. E.g. in a milk plant,
packaging of milk pouches consists of boiling, pasteurization, cooling & packaging
operations, each of which is performed by a single server in a series.
b) Speed of service; in a queueing system, service system can be expressed in two ways,
i.e. as a service rate & as a service time. The service rate describes the number of
customers serviced during a particular time period whereas, service time indicates
the amount of time needed to service a customer. E.g. a cashier can attend, on
average 5 customers in an hour, the service rate would be 5 customers/hour &
service time would be equal to 12 minutes/customer(60⁄5).
3. Queue discipline; this is a rule determining the formation of queue & a manner
in which customers forming a queue are selected for a service.
Ways in which customers in a queue are served
a) Static queue disciplines; these are based on individual customer’s status in the queue.
The most common queue disciplines here are;
 First-Come-First-Served (FCFS/FIFO); this is where customers are served in the order
of their arrival, e.g. in milk Parlour, railway station, etc.
 Last-Come-Last-Served (LCFS); is where customers are served in the reversed order of
their entry so that the ones who join last are served first. E.g. People who join an
elevator last are the ones to leave first, in a big store (godown) items which come
last are taken out firs.
b) Dynamic queue discipline; this is based on individual customer’s attributes in the
queue and these disciplines may include;
Page 3 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

 Service in random order (SIRO); under this rule, customers are selected for service at
random order irrespective of their arrivals in the service system. Thus, every
customer in a queue is equally-likely to be selected. The time of arrival of a
customer is of no relevance in this case.
 Priority service; under this, customers are grouped in priority classes on the basis of
some attributes such as service time or urgency or according to some identifiable
characteristics & FCFS to provide a service. Treatment of VIPs in preference to
other patients in a hospital is an example priority service.
4. Customer behaviour/attitude; on this basis, customers may be classified as being
patient or impatient & they generally behave in five ways as seen below;
 Balking; this is where a customer may leave the queue because the queue is too long
or waiting space is inadequate for desired service & he/she may decide to return
for service at a later time.
 Reneging; this is where a customer, after joining the queue, waits for some time and
leaves the service system due to intolerable delay or impatience.
 Jockeying; this is where a customer switches from one queue to another hoping to
receive service quickly.
 Colonizing; is where customers waiting for service jump into a newly formed line in
hopes of receiving service quickly.
 Priorities; this is where customer are served before others regardless of their order
of arrival in some applications. This means that, these customers have priority over
others.
Characteristics of a queuing system
i. Queue Length (Lq); this is a number of customers in the queue waiting to get
service. Short queues may mean good customer service or large waiting space while
long queues mean low service efficiency or little waiting space.
ii. System length (Ls); is the average number of customers in the system waiting to be
served including those being served. Long queues imply congestion, potential
customer dissatisfaction, & need for more capacity.
iii. Waiting time in a queue (Wq); is the average time that a customer has to wait in
the queue to get service. Long waiting times may mean a need to adjust the service
rate of the system or change the arrival rate of customers.
iv. Waiting Time in a system (Ws); is the average time that a customer spends in the
system from entry to the queue to completion of service. If this time is more then
there may be a need to change the priority discipline, increase productivity or
adjust the capacity.

Page 4 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

v. Server idle time; this is a relative frequency with which the service system is idle
and which is directly related to cost. Queuing theory analysis involves the study of
system behaviour over time.
vi. Transient & Steady states; a transient state is where the operating characteristics of
a system are dependent on time. While, steady state, is where after sufficient time
has passed, the system becomes independent of the initial conditions and of the
elapsed time (except under very special conditions).
vii. Traffic intensity (Utilization factor); is the proportion of time a server spends with
the customer. A necessary condition for a system to have settled to a steady state is
𝜆
that; 𝜌 < 1 𝑜𝑟 < 1 𝑜𝑟𝜆 < 𝜇 i.e. arrival rate <service rate. If 𝜌 > 1, the arrival
𝜇
rate will be greater than the service rate, and consequently, the number of units in
a queue tend to increase indefinitely as the time passes on, provided the service
rate is not affected by the queue length.
PROBABILITY DISTRIBUTIONS IN QUEUEING SYSTEMS
1. Poisson distribution process (pure birth process); pure birth models are models in which
only arrivals are counted & no departure takes place. In queueing process, birth –
death process is increased by birth or arrivals in the system & decreased by death
or departure of serviced customers from the system. Therefore, if the arrivals are
completely random, then the probability distribution of a number of arrivals in a
fixed time-interval, follow a Poisson probability distribution with parameter
(Lambda/mean) 𝜆. In other words, Poisson point process may refer to a process in
which events occur continuously & independently at a constant average rate.
2. Exponential distribution process. This is the distribution which describes inter-arrival
times in a Poisson point process. Inter-arrivals are the time intervals between two
successive events which will be1⁄𝜆.
QUEUEING MODELS
Queueing models are categorized as deterministic/probabilistic. If each customer
arrival arrives at known interval & service time is known with certainty, then the
queueing model is said to be Deterministic in nature. Whereas, when both arrival &
service rate are unknown & assumed to be random variable, then this type of queueing
model is known as probabilistic in nature.
Classification of queueing models
Queueing models are classified as follows;
Model I: (𝑀|𝑀|1): (∞|𝐹𝐶𝐹𝑆)

Page 5 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

Model II: (𝑀|𝑀|𝑆): (∞|𝐹𝐶𝐹𝑆)

Model III: (𝑀|𝑀|1): (𝑁|𝐹𝐶𝐹𝑆)
Model IV: (𝑀|𝑀|𝑆): (∞|𝐹𝐶𝐹𝑆)
Note; in this handout, model I & II have been discussed while model III & IV are
beyond the scope of this handout.
Model I/Birth & Death model – Single Channel Queueing Model with Poisson arrivals &
exponential service times.
This is a model where arrivals follow Poisson process, service times are exponentially
distributed and there is a single server with infinite capacity (i.e. infinite queue length
& population) & First-Come-First- Served service discipline. This model is one of the
most widely used & simplest models, symbolically denoted as; (𝑀|𝑀|1): (∞|𝐹𝐶𝐹𝑆),
where;
M- Poisson Arrivals (exponential inter-arrivals)
M- Exponential service time (Poisson departure)
1- Single Server
∞- Infinite Capacity (Queue length & population)
FCFS- Service discipline
Assumptions of Single Server Model
This model assumes the following conditions;
i. Arrivals are served on a FCFS basis
ii. Every arrival waits to be served regardless of the length of line.
iii. Arrivals are independent of preceding arrivals, but the average number of
arrivals doesn’t change over time.
iv. Arrivals are described by a Poisson probability distribution & come from an
infinite population.
v. Service time also varies from one customer to the next & are independent of
one another, but their average rate is known.
vi. Service times occur according to the negative exponential probability
distribution.
vii. The average service rate is greater than the average arrival rate.
Notations, Symbols & Basic Formulae Used in Queueing systems

Page 6 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

1. 𝑛~ No. of customers in the system (waiting & in service)

𝜆 𝑛
2. 𝑃𝑛 ~Steady state probability of having ‘n’ customers in the system = 𝑃0 (𝜇)
𝜆
3. 𝑃0 ~Probability of an empty system = 1 − (𝜇)
4. 𝜆~Customers Arrival rate
5. 𝜇~Service Rate
1
𝑚𝑒𝑎𝑛 𝑎𝑟𝑟𝑖𝑣𝑎𝑙 𝑟𝑎𝑡𝑒 𝜆
6. 𝜌~Traffic Intensity (Server Utilization factor) 𝑅ℎ𝑜 = 𝑚𝑒𝑎𝑛 𝑠𝑒𝑟𝑣𝑖𝑐𝑒 𝑟𝑎𝑡𝑒 = 𝜆
1 =
𝜇
𝜇
7. 𝐿𝑠 ~Length of a system (average/expected No. of customers in the system) waiting
𝜆
& in service. =
𝜇−𝜆
𝜆2
8. 𝐿𝑞 ~Queue length (average/expected No. of customers in the queue) 𝜇(𝜇−𝜆)
9. 𝑊𝑠 ~Waiting Time in the system (average/expected waiting time in the system –
1
waiting & in service) =
𝜇−𝜆
10.𝑊𝑞 ~Waiting time in the queue (average/expected waiting time in the queue) =
𝜆
𝜇(𝜇−𝜆)
𝜇
11.𝐿/𝐿 > 0~ Expected/Average length of a non-empty queue (𝜇−𝜆)
𝜌
12.Expected variance of queue length 𝑣𝑎𝑟. (𝑛) = (1−𝜌)2

Worked examples
1. Customers arrive at a Counter being manned by a single individual at a rate of 25
per hour. The time required to serve a customer has exponential distribution with
a mean of 30 per hour.
Required;
Discuss the various characteristics of the queueing system, assuming that there is only
one server.
Solution:
𝐴𝑟𝑟𝑖𝑣𝑎𝑙 𝑅𝑎𝑡𝑒( 𝜆) = 25 𝑝𝑒𝑟 ℎ𝑟.
𝑆𝑒𝑟𝑣𝑖𝑐𝑒 𝑅𝑎𝑡𝑒 (𝜇) = 30 𝑝𝑒𝑟 ℎ𝑟.

i. Traffic intensity (utilization factor)

Page 7 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

𝜆 25 5
(𝜌) = = = = 0.83
𝜇 30 6

ii. Average/expected No. of customers in the system

𝜆 25 25
𝐿𝑠 = = = = 5 𝐶𝑢𝑠𝑡𝑜𝑚𝑒𝑟𝑠
𝜇−𝜆 30 − 25 5

iii. Average/expected No. of customers in the queue

𝜆2 252 50 625 25
𝐿𝑞 = = = = = = 4.2 𝑐𝑢𝑠𝑡𝑜𝑚𝑒𝑟𝑠
𝜇(𝜇 − 𝜆) 30(30 − 25) 30(5) 150 6

iv. Average/expected waiting time in the queue

𝜆 25 25 1
𝑊𝑞 = = = = = 0.16ℎ𝑟𝑠. ≈ 10 𝑚𝑖𝑛𝑠.
𝜇(𝜇 − 𝜆) 30(5) 150 6

v. Average/expected waiting time in the system

1 1 1
𝑊𝑠 = = = = 0.2 ℎ𝑟𝑠 ≈ 12 𝑚𝑖𝑛𝑠.
(𝜇 − 𝜆) (30 − 25) 5

2. In a service department manned by one server, on an average 8 customers arrive

every 5 minutes while the server can serve 10 customers in the same time assuming
Poisson distribution for arrival & exponential distribution for service rate.
Required, determine;
i. Average number of customers in the system
ii. Average number of customers in the system
iii. Average time a customer spends in the system
iv. Average time a customer waits before being served.
Solution:
𝐴𝑟𝑟𝑖𝑣𝑎𝑙 𝑅𝑎𝑡𝑒( 𝜆) = 8⁄5 = 1.6 𝑐𝑢𝑠𝑡𝑜𝑚𝑒𝑟𝑠 𝑝𝑒𝑟 𝑚𝑖𝑛𝑢𝑡𝑒.
𝑆𝑒𝑟𝑣𝑖𝑐𝑒 𝑅𝑎𝑡𝑒 (𝜇) = 10⁄5 = 2 𝑐𝑢𝑠𝑡𝑜𝑚𝑒𝑟𝑠 𝑝𝑒𝑟 𝑚𝑖𝑛𝑢𝑡𝑒.

Traffic intensity (utilization factor)

𝜆 1.6
(𝜌) = = = 0.8
𝜇 2

i. Average number of customers in the system

𝜆 1.6 1.6
𝐿𝑠 = = = = 4 𝐶𝑢𝑠𝑡𝑜𝑚𝑒𝑟𝑠
𝜇−𝜆 2 − 1.6 0.4
ii. Average number of customers in the queue
Page 8 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

𝜆2 1.62 2.56 2.56

𝐿𝑞 = = = = = 3.2 𝑐𝑢𝑠𝑡𝑜𝑚𝑒𝑟𝑠
𝜇(𝜇 − 𝜆) 2(2 − 1.6) 2(0.4) 0.8

iii. Average time a customer spends in the system

1 1 1
𝑊𝑠 = = = = 2.5 ℎ𝑟𝑠 ≈ 2 ℎ𝑟𝑠 & 30 𝑚𝑖𝑛𝑠.
(𝜇 − 𝜆) (2 − 1.6) 0.4

iv. Average time a customer waits before being served.

𝜆 1.6 1.6
𝑊𝑞 = = = = 2 ℎ𝑟𝑠 ≈ 120 𝑚𝑖𝑛𝑠.
𝜇(𝜇 − 𝜆) 2(0.4) 0.8
3. Coca-Cola Company distributes its products by trucks loaded at its only loading
station. Both the Company trucks & the contractor’s trucks are used, of which
30% of the trucks belong to the contractor. It was found that on average, every 15
minutes one truck arrives & the average loading time is 3 minutes.
Required, make suitable assumptions & determine;
i. Probability that a truck has to wait
ii. Average waiting time of a truck
iii. Expected waiting time of the contractor’s truck in the queue per day, assuming
10 hours shift per day.
Solution:
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝐴𝑟𝑟𝑖𝑣𝑎𝑙 𝑅𝑎𝑡𝑒( 𝜆) = 60⁄15 = 4 𝑡𝑟𝑢𝑐𝑘𝑠 𝑝𝑒𝑟 ℎ𝑜𝑢𝑟
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑆𝑒𝑟𝑣𝑖𝑐𝑒 𝑅𝑎𝑡𝑒 (𝜇) = 60⁄3 = 20 𝑡𝑟𝑢𝑐𝑘𝑠 𝑝𝑒𝑟 ℎ𝑜𝑢𝑟.

i. Probability that a truck has to wait/finding a busy system/Traffic intensity (utilization

factor)
𝜆 4
(𝜌) = = = 0.2
𝜇 20

ii. Waiting time a truck spends in the system

1 1 1
𝑊𝑠 = = = ℎ𝑟𝑠 ≈ 3 ℎ𝑟𝑠 & 45 𝑚𝑖𝑛𝑠.
(𝜇 − 𝜆) (20 − 4) 16

iii. Expected waiting time of the contractor’s truck in the queue per day, assuming 10 hours
shift per day.
= (𝑁𝑜. 𝑜𝑓 𝑡𝑟𝑢𝑐𝑘𝑠 𝑝𝑒𝑟 𝑑𝑎𝑦) × (𝑐𝑜𝑛𝑡𝑟𝑎𝑐𝑡𝑜𝑟 ′ %) × (𝐴𝑣𝑔𝑒 𝑤𝑎𝑖𝑡𝑖𝑛𝑔 𝑡𝑖𝑚𝑒 𝑖𝑛 𝑡ℎ𝑒 𝑞𝑢𝑒𝑢𝑒)
30 𝜆
= (4 × 10) × ( ) × ( )
100 𝜇(𝜇 − 𝜆)

Page 9 of 10
 Queueing Theory Study Notes | Emma Charles 0753-236-367/0787-080-333 - Win 10 Pro

4
40 × 0.3 ×
20(20 − 4)
4
12 × = 0.15 ℎ𝑟𝑠 ≈ 9 𝑚𝑖𝑛𝑢𝑡𝑒𝑠 𝑝𝑒𝑟 𝑑𝑎𝑦.
320
Students’ Exercise
4. University Ladies arrive at a saloon according to Poisson process with a mean inter-
arrival time of 20 minutes. Customers spend on an average of 15 minutes for hair
braiding within a period of one hour.
Required;
i. Find the expected number of customers at the saloon & the expected queue
length
ii. How much time will a customer spend in the queue?
iii. What is the probability that a customer will not have to wait for hair braiding?
iv. What is the probability of having at least 2 customers in the system?
5. Dr. Jane Francis a re-known Psychiatrist operates a Psychology Clinic at the City
Centre. Patients arrive at the clinic at the mean rate of 40 patients per hour. The
consultation period taken by Dr. Jane Francis per patient is an average of 4
minutes. Each patient is allowed only one consultation per day.
Required, determine;
i. Average number of patients in a queueing system
ii. Expected number of customers in the queue
iii. The probability that there is no patient in a queueing system
iv. Probability that Dr. Jane Francis is busy/having a queue/not receiving
immediate service
v. Probability of having 6 customers in the system
6. A restaurant has only one cashier. During the peak hours, customers arrive at a
mean rate of 60 customers per hour. The average number of customers that can
be processed by a cashier is 90 customers per hour.
Required, determine
i. The probability that a cashier is idle
ii. Expected number of customers in a queueing system
iii. Expected time a customer spends in the system
iv. Expected number of customers in the queue
v. Expected time a customer spends in the queue waiting for a service.

Page 10 of 10