International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No.

11

16

WORKING CAPABILITIES OF A.T.M. WITH PARALLEL
REDUNDANT CENTRAL COMPUTER

Jitendra Singh
Astt. Prof., Dept. of Computer Sciences,
S.R.M. University, NCR Campus, Modinagar, U.P.-201204, India
Email: jitendra.jit@gmail.com

Rabins Porwal
Associate Professor, Department of Information Technology,
Institute of Technology & Science (ITS), Mohan Nagar, Ghaziabad (UP) - 201 007,INDIA
Email: rabinsp@rediffmail.com

Abstract
In this paper, the author has done his analysis to compute the working capability of an ATM
system. In this study, the author assumes single ATM. The capability of the system is affected by all
the units of the system. Using supplementary variables technique, the author has done mathematical
formulation of the system. This mathematical model has been solved by employing Laplace transform.
Probability of states (depicted in fig-1(d)), has been computed. Reliability of considered system has
been obtained. Steady-state behavior of the system and a particular case has also been obtained to
improve practical utility of the system. An important reliability parameter, viz. mean time to failure of
considered system has also been computed to improve practical utility of the model. Graphical
illustration followed by a numerical computation has been appended at the end to highlight important
results of the study.
KEY WORDS: Parallel redundancy, Markovian process, Supplementary Variables, Laplace
transforms, exponential time distribution etc.

1. Introduction:
An automated teller machine (ATM) is a computerized tele-communication
device that provides the customers of a financial institution with access to financial
transactions in a public space without the need for a human clerk or bank teller. On
most modern ATMs, the customer is identified by inserting a plastic ATM card with a
magnetic stripe or a plastic smart card with a chip that contains a unique card number
and some security information, such as an expiration date or CVC (CVV). Security is
provided by the customer entering a personal identification number (PIN). They are
sometimes referred to as ATM machines, an example of RAS syndrome. Fig-1(a)
shows an ATM mouth.
International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

17


Fig-1(a): Shows ATM mouth
Using an ATM, customers can access their bank accounts in order to make
cash withdrawals (or credit card cash advances) and check their account balances.
British Actor Reg Varney: Use the worlds first ATM in 1967 located at a branch of
Barclays Bank, Enfield. The system was developed by De La Rue.
The first mechanical cash dispenser was developed and built by Luther George
Simjian, installed in 1939 in New York city by the Bank of New York, but removed
after 6 months due to the lack of customer acceptance.
Thereafter, the history of ATMs paused for over 25 years, until De La Rue developed
the first electronic ATM, which was installed first in Enfield town in North
London United Kingdome on 27
th
June 1967 by Barclays Bank. This instance of
invention is credited to John Shepherd Barron although various other engineers
were awarded patents for related technologies at the time. Shepherd Barron was
awarded an OBE in the 2005 New Yorks Honors list.
The first person to use the machine was the British variety artist and actor Reg
Vauney. The first ATMs accepted only a single use token or voucher which was
retained by the machine. These worked on various principles including radiation and
low coercivity magnetism that was wiped by the card reader to make fraud more
difficult. The machine dispensed pre-packaged envelopes containing ten Ponds
sterling. The idea of a PIN stored on the card was developed by the British
engineer James Good Fellow in 1965. Fig-1(b) represents an ATM station with two
parallel machines
Block diagram of considered system has been shown in fig-1(c). The author has been
used two parallel redundant central computers to improve systems overall
performance. Single bank computer is available to fulfill users requirements. Fig-1(d)
represents the transition state diagram for considered system.
International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

18


Fig-1(b): Represents an ATM station with two parallel machines



Fig-1(c): System Configuration
B
1

International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

19

Fig-1(d): State-transition diagram
The following notations have been used in this model:

i

: Failure rate of subsystem ) , , ( C B A i = .
) ( j
i

:
The first order probability that
th
i failure can be repaired
in the time interval ) , ( + j j , conditioned that it was
not repaired upto the time j.
) (
0 , 0 , 0
t P
: Pr {at time t, system is all operable as subsystems A, B
and C are operable}.
. ) , (
0 , 0 ,
etc t x P
F

: Pr {at time t, system is failed due to failure of
subsystem A}. Elapsed repair time lies in the interval
) , ( + x x .
International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

20

) , (
0 , , 0
t y P
D

: Pr (at time t, system is in degraded state due to failure of
any one part of subsystem B}. Elapsed repair time lies in
the interval ) , ( + y y .
etc. ) , (
0 , ,
t x P
D F

: Pr (at time t, system is in failed state due to failure of
subsystem A while one part of subsystem B has already
failed}. Elapsed repair time for subsystem A lies in the
interval ) , ( + x x .
) ( j Si
:
{ } dj j j
i i

= ) ( . exp ) ( , j and i
) ( j D
i

:
j
j Si ) ( 1
, j and i
M.T.T.F. : Mean time to failure
L.T. : Laplace transform
s : L.T. variable

The following assumptions have been associated with this model:
1. Initially, the whole system is good and operable.
2. All failures follow exponential time distribution and are S-independent.
3. All repairs follow general time distribution and are perfect.
4. There are two central computers working in parallel redundancy.
5. On failure of any one central computer, the whole system works in degraded
state.
6. Repair facilities are always available and there is no time lap between a failure
and start of repair.

2. Literature Review:
In this section, the author has done analysis for mathematical formulation of the
model, its solution, a particular case and various results related to reliability
estimation.
2.1 Formulation of mathematical model
Probability considerations and limiting procedure yields the following set of
difference-differential equations [1]-[3], which is continuous in time and discrete in
space, governing the behavior of considered system:
( ) ( ) ( ) ( ) ( )


+ =
(

+ + +
0
0 , , 0
0
0 , 0 , 0 , 0 , 0
1
, , dy y t y P dx x t x P t P
dt
d
B D A F C B A



International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

21

( ) ( ) ( ) ( )


+ +
0
0 , , 0
0
, 0 , 0
, , dm m t m P dz z t z P
B F C F

(1)
( ) 0 , ) (
0 , 0 ,
=
(

t x P x
t x
F A

(2)
( ) 0 , ) (
0 , , 0
1
=
(

+ + + +

t y P y
t y
D B C B A

(3)
( ) 0 , ) (
, 0 , 0
=
(

t z P z
t z
F C

(4)
( ) 0 , ) (
0 , ,
=
(

t x P x
t x
D F A

(5)
( ) 0 , ) (
0 , , 0
=
(

t m P m
t m
F B

(6)
( ) 0 , ) (
, , 0
=
(

t z P z
t z
F D C

(7)
Boundary conditions are:
( ) ( ) t P t P
A F 0 , 0 , 0 0 , 0 ,
, 0 =
(8)
( ) ( ) ( ) ( ) ( ) ( )


+ + =
0
, , 0
0
0 , , 0 , 0 , 0 0 , , 0
, , , 0 dz z t z P dx x t x P t P t P
C F D A D F B D

(9)
( ) ( ) t P t P
C F 0 , 0 , 0 , 0 , 0
, 0 =
(10)
( ) ( ) t P t P
D A D F 0 , , 0 0 , ,
, 0 =
(11)
( ) ( ) t P t P
D B F 0 , , 0 0 , , 0
, 0 =
(12)
( ) ( ) t P t P
D C F D 0 , , 0 , , 0
, 0 =
(13)
Initial conditions are:
( ) 1 0
0 , 0 , 0
= P , otherwise all state probabilities are zero at 0 = t .

(14)
2.2 Solution of the model
Taking Laplace transforms of equations (1) through (13) subjected to initial
conditions (14) and on solving them [4] one by one, we obtain the following L.T. of
transition state probabilities depicted in fig-1(d):
International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

22

) (
1
) ( 0 , 0 , 0
s B
s P =

(15)
) (
) (
) ( 0 , 0 ,
s B
s D
s P
A A
F

=
(16)
) (
) (
) ( 0 , , 0
s B
s A
s P D =
(17)
) (
) (
) (
C C
, 0 , 0
s B
s D
s P F

=
(18)
) (
) ( A(s)
) (
A
0 , ,
s B
s D
s P
A
D F

=
(19)
) (
) ( A(s)
) (
B
0 , , 0
s B
s D
s P
B
F

=
(20)
) (
) ( A(s)
) (
C
, , 0
s B
s D
s P
C
F D

=
(21)
where
( )
| | ( )
C B A B
C
C
A
A
C B A B B
s D s S s S
s D
s A


+ + + +
+ + +
=
1
1
) ( ) ( 1
) (
(22)
and ) ( ) ( ) ( ) ( ) ( s S s A s S s S s s B B
B
C
C
A
A C B A
+ + + =
| | ( )
C B A
B C
C
A
A B
s S s S s A s S s A + + + + +
1
) ( ) ( ) ( ) (
(23)

2.3 Steady-state behavior of the system
Using final value theorem of L.T., viz., ( ) ( ) P s P s Lim t P Lim
s t
= =


0
(say), provided
limit on left exits, we obtain the following steady-state behavior [5] of considered
system from equations (15) through (21):
) 0 (
1
0 , 0 , 0
B
P

=
(24)
) 0 (
0 , 0 ,
B
M
P
A A
F

=


(25)
) 0 (
) 0 (
0 , , 0
B
A
P
D

=
(26)
International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

23

) 0 (
, 0 , 0
B
M
P
C C
F

=


(27)
) 0 (
) 0 (
0 , ,
B
M A
P
A A
D F

=


(28)
) 0 (
) 0 (
0 , , 0
B
M A
P
B B
F

=


(29)
) 0 (
) 0 (
, , 0
B
M A
P
C C
F D

=


(30)
where =

= ) 0 ( i
i
S M Mean time to repair i
th
failure
( )
| | ( )
C B A B C A
C B A B B
D
D
s A


+ + +
+ +
=
1
1
1
) (

and
0
) ( ) 0 (
=
(

=
s
s B
ds
d
B

2.4 Particular case
When all repairs follow exponential time distribution
In this case, setting ( )
+
= ,
) (
i
i
i
j
j S

i and j, in equations (15) through (21), we

obtain the following L.T. of various transition-states depicted in fig-1(b):
) (
1
) ( 0 , 0 , 0
s E
s P =
(31)
( )
A
A
F
s s E
s P

+
=
). (
) ( 0 , 0 ,
(32)
) (
) (
) ( 0 , , 0
s E
s F
s P D =
(33)
( )
C
C
F
s s E
s P

+
=
). (
) ( , 0 , 0
(34)
( )
A
A
D F
s s E
s F
s P

+
=
). (
) (
) ( 0 , ,
(35)
( )
B
B
F
s s E
s F
s P

+
=
). (
) (
) ( 0 , , 0
(36)
International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

24

( )
C
C
F D
s s E
s F
s P

+
=
). (
) (
) ( , , 0
(37)
where
( )
(

+
+
+
+ + + +
=
C
C C
A
A A
B C B A
B
s s
s
s F

1
) (
(38)
and
B
B B
C
C C
A
A A
C B A
s
s F
s s
s s E

+
+ + + =
) (
) (

1
1
) (
B C B A
B
C
C C
A
A A
B
s s s
s F

+ + + +
(

|
|
.
|

\
|
+
+
+
+


(39)

2.5 Reliability and M.T.T.F. evaluation
From equation (15), we have
( )
C B A
s
s R
+ + +
=
1

Taking inverse L. T., we get
( ) { } t exp ) (
C B A
t R + + = (40)
Also,

=
0
) ( . . . . dt t R F T T M

C B A
+ +
=
1
(41)

2.6 Numerical illustration
For a numerical illustration, let us consider the values
, 006 . 0 =
A
, 003 . 0 =
B
, 001 . 0 =
C
and t = 0,1,2,-----10.
Using these values in equations (40) and (41), the corresponding graphs have been
shown through figure (2) and (3) respectively. These graphs illustrate the variation
into obtained results.


International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

25


3. Results and discussion:
We have given the reliability of considered system, for various values of time t, in
fig-2. Critical examination of fig-2 reveals that reliability of the system decreases in a
constant manner approximately. It should be noted that there are no sudden jumps in
the values of reliability of considered system.
Fig-3 shows the values of M.T.T.F. for various values of failure rate of subsystem B.
In this graph, we observe that M.T.T.F. decreases catastrophically in the beginning
but thereafter it decreases in a constant manner.


Fig-2: The way reliability of considered system decreases with the increase in time.

Fig-3: The way MTTF of considered system decreases with the increase in failure rate of
subsystem B.
REFERENCES
International Journal of Computational Intelligence and Information Security, November 2011 Vol. 2, No. 11

26

[1] Cluzeau, T.; Keller, J.; Schneeweiss, W. (2008): An Efficient Algorithm for Computing the
Reliability of Consecutive-k-Out-Of-n:F Systems, IEEE TR. on Reliability, Vol.57 (1), 84-
87.
[2] Gupta P.P., Agarwal S.C. (1983): A Boolean Algebra Method for Reliability Calculations,
Microelectron. Reliability, Vol.23, 863-865.
[3] Lai C.D., Xie M., Murthy D.N.P. (2005): On Some Recent Modifications of Weibull
Distribution, IEEE TR. on Reliability, Vol.54 (4), 563-569.
[4] Tian, Z.; Yam, R. C. M.; Zuo, M. J.; Huang, H.Z.(2008): Reliability Bounds for Multi-State
k-out-of- n Systems, IEEE TR. on Reliability, Vol.57 (1), 53-58.
[5] Zhimin He., Han T.L., Eng H.O. (2005): A Probabilistic Approach to Evaluate the Reliability
of Piezoelectric Micro-Actuators, IEEE TR. on Reliability, Vol.54 (1), 44-49.