International Journal of Scientific Research in _______________________________ Research Paper .

Multidisciplinary Studies E-ISSN: 2454-9312

Vol.4, Issue.11, pp.01-05, November (2018) P-ISSN: 2454-6143

On The Application of Queuing Model in Nigeria Banking Sector: a Case

Study of Zenith Bank Plc, Abakaliki Branch
O. C. Asogwa1, C. M. Eze2 & M. T. Edeaja2*
1
Federal University Ndufu Alike Ikwo, Ebonyi State
2*
Federal University Ndufu Alike Ikwo, Ebonyi State
2
University of Nigeria Nsukka

Corresponding Author: [email protected], Tel: +234 8036785840

Available online at: www.isroset.org

Received: 13/Nov/2018, Accepted: 20/Nov/2018, Online: 30/Nov/2018

Abstract - This study focused on the queuing system in Nigerian banking sector. The sources of data were primary and were
collected from Zenith Bank Plc, Abakaliki Branch at the withdrawal section of the banking hall for four days in a period of one
month interval. The data were collected based on the arrival pattern and the service pattern of customers. The methodology
employed followed by the birth and death Markovian process. The results obtained showed that the arrival rate is 0.2219 and
the service rate is 0.1524. However, the probability that the servers are idle is 0.1574 and the cost incurred from waiting is
N656.76. Besides, any increment on the number of servers by the bank management will help reduce the time customers spend
on queue and also reduce cost incurred from waiting. It was also suggested that the management of the concerned Bank should
employ only capable servers as to reduce customers wait on service, that is when the cause of long time taken to serve one
customer is not due to network or power supply.

Keywords- Bank, queue, arrival time, service time, incurred cost.

I. INTRODUCTION frequent breakdown of such computerization and fraudulent

activities. The problem of long queue persisted in almost all
Queue is an act of joining a line or waiting. Queues are banking systems. Time is money and waiting is a non-value
mostly formed when customers (arrivals) demanding added activity which can cause prolonged discomfort and
service have to wait because their number exceeds the economic cost to individuals coming into the system.
number of servers available or the facility does not work Therefore, it is always a desire of every customer to obtain
efficiently or takes more than the time prescribed to service an efficient and prompt service delivery from a service
a customer. It is a part of everyday human experience at system.
banks, filling stations, barber shops, salon shops, bus-stops,
supermarkets, telephone booths, computing and design of II. RELATED WORK
factories, toll gates, food canteens etc. According [1], they
described queuing theory as a study that tends to help the There are many applications of the theory of queues, most of
owners of different organizations to analyze the percent at which have been well documented in the literature of
which customers wait for their services to be given to them probability, Operation research and Management science.
and improve the percentage of services in this manner. [2] a Some of the applications are machine repair, tool booths,
Danish Mathematician did the first study on Queuing theory inventory control, the loading and unloading of ships,
which resulted into the worldwide acclaimed Erlang scheduling patients in the hospitals, in computer field’s etc.
telephone model. He examined the telephone network However, [3] did a comparative analysis of Service Delivery
system and tried to determine the effect of fluctuating by ATM in Two Banks with the application of Queuing
service demands on calls on utilization of automatic dial Theory. The study found that the average arrival rate,
equipment. Over the years, there have been technological average service rate, average time spent in the queue for
advancements such as Internet Banking, Mobile or Access bank are 2.01, 1.65, 0.5 respectively and UBA are
Telephone Banking, Automated Teller Machine (ATM) etc. 3.28, 1.75, 1.67minutes, respectively. The study concluded
by banks in attempt to minimize waiting line problem at the that the average number of idle time obtained for the two
withdrawal point and other service points in the bank. These banks were 3minutes and 7minutes respectively. [4]
measures have not yielded the much desired result due to examined the application of queuing models to customer’s

© 2018, IJSRMS All Rights Reserved 1

 Int. J. Sci. Res. in Multidisciplinary Studies Vol. 4(11), Nov 2018

management in the banking system using United Bank for busiest days of the week in Zenith Bank and every other
Africa, Okpara Avenue Branch Enugu, as a case study. The Nigerian Banks.
results obtained from the study showed that the arrival
pattern follows a poisson distribution and that the service  Formulation of the Model
pattern follows an exponential distribution. The study In formulating the queuing model for this system, the
recommended that the Bank management should increase following assumptions must be put into consideration;
the number of servers to three so as to help reduce the time  The arrival of customers into the system is discrete
customers spend on queue and also reduce cost incurred from poison distribution with arrival rate.
from waiting. In addition, [5] studied the Imperatives of  There is multi service channel i.e. (M/M/S or C =
Customer Relationship Management in Nigerian Banking 2,3,…,5)
Industry. Findings from the study revealed that there is a  The service channel can only render service of
direct relationship between customer relationship infinite rate exponentially distribution with service
management and customer loyalty as well as Banks rate.
profitability. The study recommended that management of  The order of arrival and departure are the same and
banks should pursue customer relationship management we assume FIFO with infinite queue capacity.
programmes with rigor to achieve the business objectives of  Server will never remain idle if there is one or more
the bank. [6] expressed that queuing is common in banking jobs in the service node.
industry that is always characterized by customer’s  Once service is initiated, service of a job will
explosion. Waiting-line model has been used as a bail out for continue until completion.
banks from this overwhelming customers’ excessive demand The waiting area for the customers in the system is N, which
so as to improve the system performance. A queue system is is either limited or unlimited. Hence, the model can be
also applicable to commercial service. Consequently, [7] formulated appropriate by using a system for the investment
studied multiple branches of bank, provided empirical system. Kendall’s notation is introduced (V/W/X/Y/Z) as
evidence that teller waiting times affect customer introduced by [8 ].
satisfaction and retention. Their study revealed significant V- Which is the arrival distribution or pattern is poisson
heterogeneity in customer sensitivity to waiting time, some W- The service time is exponential.
of which could be explained through demographics and the X- The number of available server S in the system range
intensity of competition faced by the branch. from two to ten from the assumption above.
Y- This represents the system capacity.
III. METHODOLOGY Z- This represents the queue discipline FIFO, which is first
in first out.
The aim of this research work is to determine the Considering the above assumptions and approach, the model
productivity and efficacy of using different number of formulated is
servers by the management of Zenith Bank Plc, Abakaliki (M/M/S/N/FCFS) by Kendall’s notation
Branch and also to detect the effect of servers spending more
time in serving one customer. The source of data used in this  Single Servers Queue with M|M|1 Model
research work is a primary source of data collected at the The single server model is constructed so that queue lengths
withdrawal point of Zenith Bank Plc, Abakalik branch. The and waiting line can be predicted. Single-Channel Queuing
method applied in collecting data for this experiment is Model with Poisson Arrivals and Exponential service times
through direct observation at the Bank. The observation was (M/M/1) queue, where the first M indicates the arrival rate,
made during the working hours (8:00am – 10:00am) for a the second M indicates the service rate and the 1 indicates
period of one month at Zenith Bank Plc, Abakaliki branch. that there is only a single server.
Data were collected only on Mondays and Fridays during the
week. Monday and Friday are being considered as the
 Diagram Representing Single Server System

Arrival DEPARTURE
QUEUE SERVER

Fig 1: Diagrammatic representation of single server system

 Multiple Servers Queue with M|M|S Model depend on the state of the system). The arrival pattern being
The M|M|S system is a queuing process having Poisson stated independent, 𝜆n = 𝜆 for all n. The service times
arrival patterns, S server, with S independent, identically associated with each server are also independent, but since
distributed, exponential service times (which does not the number of servers that actually attend to customers (i.e.

© 2018, IJSRMS All Rights Reserved 2

 Int. J. Sci. Res. in Multidisciplinary Studies Vol. 4(11), Nov 2018

are not idle) does depend on the number of customers in process customer through the service time facility is state
the system, the effective time it takes the system to dependent.

 Diagram Representing Multiple Server System

ARRIVALS SERVER(C=2)
SERVER(C=2)
.
. Departure
.
SERVER(C=m)
Fig 2: A multi server queue system

Key
Customers
Queue
Fig 3: the key note of the diagram in fig 2 above.
If there are n customers in the queuing system at any c µ > l : the total number of servers must be able to serve
point in time, then the following cases may arise: customers faster than they arrive .
i. If n < c or s, (number), then there will be no queue The probability of zero customers in the system ( ) is
of customers in the system is less than the number given by:
of servers. However, (c or s - n) number of servers 1
will not be busy. The combined service rate will P0  c
then be µ n = nµ.      
ii. If n ≥ c or s, (number of customers in the system is   1 

more than or equal to the number of servers), then    1

c![1   ]
all servers will be busy and the maximum number
of customers in the queue will be (n – c or s). The
combined service rate will be µ u = cµ. c
 The parameters for multiple-server model are as … (3.1)
follows: The capacity utilization in this system is
 = Arrival rate We can use the above equation of <1
µ= Service rate
c or s = The number of servers If > 1, then the waiting line grows larger and larger i.e.
Lq = Queue Length becomes infinite if the process runs long enough.
N = System capacity The probability of n customers ( ) in the queuing system
Average number in the queue. is
Average number in the system. n
Waiting time in the queue.   

Total time lost on queue Pn   nc P0 … (3.2)
Pn (t) = probability that exactly n customers are in c!c , for n>c
queuing system at time t. n
P0 (t) = probability that there are no customers in   

queuing system in time t.
Pn   P0
ρ = Utility factor c , for n≤c … (3.3)
ℓ = Traffic intensity
When C = 1 (i.e. there is one service facility), equations
(3.3) reduces to
The formulas for the operating characteristics of the
multiple-server model are as follows.

© 2018, IJSRMS All Rights Reserved 3

 Int. J. Sci. Res. in Multidisciplinary Studies Vol. 4(11), Nov 2018

Pn    
Nq
  Tq  ... (3.6)

... (3.4) The average number of customers in the queue is
The average number of customers in the queuing system is 
c 1 Ns  Nq  ... (3.7)
   
Nq    P Total time lost at waiting per day
2 0
   Tl    8  Tq … (3.8)

c  c!1  Traffic intensity
 c 
  
ℓ= … (3.9)
… (3.5) 
The average time a customer spends in the queuing system NOTE: In the above formulas, if c = 1 i.e. only one server,
(waiting and being served) is then it can be used to calculate for single server system.
Inter-arrival time for 211 customers, T=951 Minutes
IV. RESULTS Time taken by 211 customers to be served, S= 1385
Minutes
 Parameters Estimation Arrival rate, λ= = = 0.2219
Number of customers for four days (N1, N2, N3, N4)
N1 = 54, N2 = 45, N3 = 55, N4 = 57 Service rate, μ= = = 0.1524
N = N1 + N2 + N3 + N4 = 211 Traffic intensity (ℓ) = = = 1.456
Service time: This is the time taken to serve a customer.
This implies that
i.e. service end subtract from service begin.
μ = 0.1524
Arrival time: This is the time at which each customer
λ = 0.2219
enters the system.
4.3 Table of Result
Queue parameters Two servers Three servers Four servers Five servers

P0 0.1574 0.2215 0.2312 0.2329

Nq 1.6419 0.1483 0.039 0.0074

NS 3.0979 1.6043 1.495 1.4634

Tq 7.3993 0.6683 0.1758 0.0333

Tl 13.1352hrs 1.1864hrs 0.312hrs 0.0592hrs

Ac N656.76 N59.32 N15.6 N2.96

Table 1: A table of result for adopting different number of servers

V. DISCUSSION OF RESULTS From table 1, the use of three servers shows the expected
number in the waiting line is 0.1483. The expected number
Considering the analytical solution, the capacity of the in the system is 1.6043. The expected waiting time in the
system under study is 211 customers and the arrival rate is queue is 0.6683 minutes and the expected total time lost
0.2219 while the service rate is 0.1524. The arrival rate waiting in one day is 1.1864 hours. The average cost lost per
being greater than the service rate implies that customers day from waiting is N59.32.
have to queue up, though the queue will not be long. The
probability that the servers are idle is 0.16 which shows that Column four represents the use of four servers which
the servers will be 16% idle and 84% busy. communicated that the expected number in the waiting line
The expected number in the waiting line is 1.6419. The is 0.039. The expected number in the system is 1.495. The
expected number in the system is 3.0979. The expected expected waiting time in the queue is 0.1758 minutes and the
waiting time in the queue is 7.3993 minutes and the expected expected total time lost waiting in one day is 0.312 hours.
total time lost waiting in one day is 13.1352 hours. The The average cost lost per day from waiting is N15.6
average cost lost per day from waiting is N656.76 when two The expected number in the waiting line is 0.0074. The
servers are in use. expected number in the system is 1.456. The expected

© 2018, IJSRMS All Rights Reserved 4

 Int. J. Sci. Res. in Multidisciplinary Studies Vol. 4(11), Nov 2018

waiting time in the queue is 0.033 minutes and the expected reduce the cost incurred from waiting, if other factors like
total time lost waiting in one day is 0.0592 hours. The human and machine technicalities have been taken care of.
average cost lost per day from waiting is N2.96 However, the use of single system with multiple server has
However, the average cost per day for waiting is N656.76 an edge in the sense that slow server does not affect the
and from the calculation of the comparing solutions, the movement of the queue. That is customers waiting can move
average costs per day from waiting to another server and not wait for the slow server.
The management of Zenith bank Plc, Abakaliki branch can
are N59.32, N15.6, N2.96. There had been a saving in the take extra measure in employing experienced workers
expected cost of N656.76 - N59.32 = N597.44 and N656.76 especially those expected to work at the withdrawal point
- N15.6 = N641.16 and N656.76 - N2.96 = N653.8. and take care of other faults which could be attributed to
From the time spent on serving some customers, we can see Automated Teller Machines, networks, power irregularities,
that some customers take much time to be served. The cause etc in other to avoid customers being delayed unnecessarily.
of this delay could be as a result of network problems or Moreover, more effort should be focused on the minimizing
power interruption. But in some cases, these causes could be the ideal time of the server and considering multiple server
attributed to ineffectiveness and lack of job experience. not necessarily in sequential ordering.

VI. CONCLUSION AND FUTURE SCOPE ACKNOWLEDGEMENT

From the discussion of results, it could be concluded that The authors strongly acknowledge the management of
any addition made on the number of servers will help reduce Zenith Bank Plc Abakaliki Branch for her compliance in
the time customers spend on queue and as well help to carrying out this research, for allowing us collects the
needed data for the research

REFERENCE

[1] Gross, D. and Harris, C. M. (1985) “Fundamental of Queuing [8] Kendall, (1953). Classification of
Theory”, 2ndedn. John Wiley, New York.. QueuingModel...people.revoledu.com/kardi/tutorial/Queuing/Ke
[2] Erlang A. K. (1909). The Theory of Probabilities and Telephone ndall- Notation.html.
Conversations.
[3] Ogunsakin, R. E., Babalola, B. T., and Adedara, M. T. (2013). AUTHOR’S PROFILE:
Comparison of Service Delivery by ATM in two Banks:
Application of queuing theory. Journal of Mathematics (IOSR). Mr. O. C. Asogwa currently a PhD student in University of
Vol 9, No 3, pp 52 – 56. Ibadan, Ibadan Nigeria. He is working as Lecturer in
[4] Ogbadu and Usman, (2012). Imperatives of Customer Department of Mathematics/Computer
Relationship Management in Nigeria Banking Industry. Kuwait Science/Statistics/Informatics in Federal University Ndufu
Chapter of Arabian Journal of Business and Management Review Alike Ikwo, Ebonyi State in Nigeria. He is a member of
Vol 2(1), 123-132.
[5] Odunukwe Adaora (2013). Application of Queuing Models to
Nigerian Statistical Association (NSA) and other
Customers Management in the Banking System (A Case Study of professional bodies since 2014. He has published more
United Bank for Africa, Okpara Avenue Branch Enugu): An than 10 research papers in reputed international journals
unpublished undergraduate project, Caritas University, Enugu and conferences which is available online. His main
state, Nigeria. research work focuses on statistical modelling,
[6] Ullah, A. (2014). Sub-Optimization of Bank Queuing System. experimental design and analysis, time series modelling,
Qualitative and Quantitative Analysis, Vol. 2 (2), pp. 978– 1047.
[7] Campbell, D. and Frei, F. (2010). Market Heterogeneity and etc. He has just 3 years of teaching and research
Local Capacity Decisions in Services. Manufacturing & Service experience.
Operations Management Research, Vol 5(1), 31-41

© 2018, IJSRMS All Rights Reserved 5