Case Study For Bank ATM Queuing Model

Case Study for Bank ATM Queuing Model
Abstract:
Bank ATMs would avoid losing their customers due to a long wait on the line. The bank
initially provides one ATM. However, one ATM would not serve a purpose when
customers withdraw to use ATM and try to use other bank ATM. Thus, to maintain the
customers, the service time needs to be improved. This paper shows that the queuing
theory may be used to solve this problem. We obtained the data from a bank ATM in a
city. We used Littles Theorem and M/M/1 queuing model. The arrival rate at a bank
ATM on Monday during banking time is 1 customer per minute (cpm) while the service
rate is 1.66 cpm. The average number of customers in the ATM is 1.5 and the
utilization period is 0.60. We discuss the benefits of applying queuing theory to a busy
ATM in conclusion.
Keywords: Bank ATM, Littles Theorem, M/M/1 queuing model, Queue, Waiting lines.
INTRODUCTION :
This paper uses queuing theory to study the waiting lines in Bank ATM in a city. The bank
provides one ATM in the main branch. In ATM, bank customers arrive randomly and the
service time i.e. the time customer takes to do transaction in ATM, is also random. We use
M/M/1 queuing model to derive the arrival rate, service rate, utilization rate, waiting time in
the queue and the average number of customers in the queue. On average, 500
customers are served on weekdays ( Monday to Friday ) and 300 customers are served on
weekends ( Saturday and Sunday ) monthly. Generally, on Mondays, there are more
customers coming to ATM, during 10AM to 5PM.
QUEUING THEORY :
A. Littles Theorem
Littles Theorem describes the relationship between throughput rate (i.e. arrival and service
rate), cycle time and work in process (i.e. number of customers/jobs in the system). The

theorem states that the expected number of customers () for a system in steady state can be
determined using the following equation:

Here, is the average customer arrival rate and is the average service time for a customer.
B. ATM Queuing model :
M/M/1 queuing model means that the arrival and service time are exponentially distributed
(Poisson process). For the analysis of the ATM M/M/1 queuing model, the following
variables will be investigated:
: The mean customers arrival rate
: The mean service rate
: / = utilization factor.
Probability of zero customers in the ATM :
0
= 1 (2)

n
: The probability of having n customers in the ATM:
n
=
0
n
= (1 )
n
(3)
: The average number of customers in the ATM:
= /1- = /- (4)
Lq : The average number of customers in the queue:
Lq = = 2/1- = /- (5)
Wq : The average waiting time in the queue:
W
q
= Lq/ = /- (6)

: The average time spent in the ATM, including the waiting time:
= / = 1/- (7)

Observation and Discussion :
We have collected the one month daily customer data by observation during banking time, as
shown in Table-1.

ONE MONTH DAILY CUSTOMER COUNTS

ONE MONTH TOTAL CUSTOMER COUNTS
From the above figure-1, we can say that, the number of customers on Mondays is double
the number of customers on Sundays during a month. The busiest period for the bank ATM is
on Mondays and Tuesdays during banking time (10am to 5pm). Hence, we will focus our
analysis in this time period. Also, we can observe from figure-2 that, after Monday, the
number of customers start decreasing slowly as the week progresses. On Thursdays, it is least
and on Fridays and Saturdays, it stays slightly more than Thursdays. This is because the next
day will be a holiday.
Calculations :
We have observed that, after Sunday, during first two days of a week, there are, on average
60 people coming to the ATM in one hour time period of banking time. From this we can
derive the arrival rate as:
= 60/60 = 1 customer / minute (c.p.m)

We also found out from observation that each customer spends 3/2 minutes on average in
the ATM (), the queue length is around 1 people ( ) on average and the average waiting
time is around 1/2 minutes i.e. 30 seconds. Theoretically, the average waiting time is :
W
q
= L
q
/ = 1customer / 1c.p.m = 1 minute.
From this calculation, we can see that, the observed actual waiting time does not differ by
much when it is compared with the theoretical waiting time.
Next, we will calculate the average number of people in the ATM using (1),
L = 1 c.p.m * 3/2 minutes = 3/2 = 1.5 customers.
Using (4), we can also derive the utilization rate and the service rate.
= (1+L) / L = 1(1+1.5) / 1.5 = 1.66 c.p.m
= / = 1 / 1.66 = 0.6
This is the probability that, the server, in this case ATM, is busy to serve the customers,
during banking time. So, during banking time, the probability of zero customers in the ATM
is :
P
0
= 1 = 1 0.6 = 0.4
The queuing theory provides the formula to calculate the probability of having customers
in the ATM as follows:
P
n
= (1-)
n
= (1-0.6)*0.6
n
= 0.4*0.6
n

We assume that impatient customers will start to walk when they see more than 3 people are
already queuing for the ATM. We also assume that the maximum queue length that a patient
customer can tolerate is 10 people. As the capacity of the ATM is 1 people, we can
calculate the probability of 4 people in the system (i.e in the ATM). Therefore, the
probability of customers going away = (more than 3 people in the queue) = (more than
4 people in the ATM) is :

P
5-11
=
n=5
11
Pn = 0.07558 = 7.55%
Evaluation :
The utilization is directly proportional with the mean number of customers. It means that the
mean number of customers will increase as the utilization increases. The utilization rate at
the ATM is at 0.60. However, this is the utilization rate during banking time on Mondays
and Tuesdays. On weekend, the utilization rate is almost half of it. This is because the
number of people on weekends is only half of the number of people on weekdays. In case of
the customers waiting time is lower or in other words, we waited for less than 30 seconds, the
number of customers that are able to be served per minute will increase. When the service
rate is higher the utilization will be lower, which makes the probability of the customers
going away decreases.
Benefits :
This research can help bank ATM to increase its QoS (Quality of Service), by anticipating, if
there are many customers in the queue. The result of this paper is helpful to analyse the
current system and improve the next system. Because the bank can now estimate the number
of customers waiting in the queue and the number of customers going away each day. By
estimating the number of customers coming and going in a day, the bank can set a target that,
how many ATMs are required to serve people in the main branch or any other branch of the
bank.
Conclusion :
This paper has discussed the application of queuing theory to the Bank ATM. From the
result, we have obtained that, the rate at which customers arrive in the queuing system is 1
customer per minute and the service rate is 1.66 customers per minute. The probability of
buffer flow if there are 3 or more customers in the queue is 7 out of 100 customers. The
probability of buffer overflow is the probability that, customers will run away, because may
be they are impatient to wait in the queue. This theory is also applicable for the bank, if they
want to calculate all the data daily and this can be applied to other branch ATM also. In this
way, this research can contribute to the betterment of a bank in terms of its functioning
through ATM.

Case Study For Bank ATM Queuing Model

Uploaded by

Copyright:

Available Formats

Case Study For Bank ATM Queuing Model

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Case Study For Bank ATM Queuing Model

Uploaded by

Copyright:

Available Formats

Case Study for Bank ATM Queuing Model

You might also like