Paper 6

International Journal of Advances in Science and Technology,
Vol. 3, No.3, 2011

Stochastic Behavior of Two Parallel Unit System with
Appearance and Disappearance of Repairman and
Correlated Life Time

Mohit Kakkar
1
, Ashok Chitkara
2
, Sanjeev Kumar
3

1
Chitkara University,Himachal Pradesh,India
mohitkakkar2001@gmail.com

2
Chitkara University,Himachal Pradesh ,Iindia
ashok.chitkara@chitkara.edu.in

3
Depatment of Mathematics PIET,Haryana,India
skdagar2211@gmail.com

Abstract

This paper deals with the stochastic behavior of two dissimilar parallel units, both are
operative with different efficiency and appearance and disappearance of one repairman, and life
times are correlated. Techniques of regenerative point had been used to measure the system
effectiveness, useful to system designers and industrial managers. The graphical behaviors of
MTSF and profit function have also been studied.

Keywords: MTSF, Transition probability, Availability, Busy period

1. Introduction

Two similar parallel unit systems have widely studied in literature of reliability theory Goel
et.al.[1] analyzed a two unit priority standby system under the assumption that the operator of the
system does not need rest but in real life it is not so, a person cannot work continuously on a
particular machine and he/she needs rest after working a random amount of time .Many authors
have analyzed various systems models considering different repair policies [2-3] Gupta and Goel [4]
considered a two dissimilar unit parallel system model assuming that a delay occurs due to
dministrative action in locating and getting the repairman available to the system.Gupta et al. [5]
analyzed a two nonidentical unit stand by system model with random appearance and
disappearance of repairman.The purpose of this present study is to investigate a model assuming
that the system repairman needs rest instead of the system and also to extend the concept of random
appearance and disappearance of repairman in a two dissimilar parallel unit system with failure
and repair times of each unit as correlated random variables having their joint distribution as
bivariate exponential .
System description and Assumption are as follows:

1) System consists of two non-identical (dissimilar) units A and B, both units are initially
operating but the operation of only one unit is also sufficient for operating the system.
2) There is single repair facility which appears in and disappears from the system randomly
.once the repairman starts the repair of a failed unit he does not leave the system till all the
units are repaired that failed during his stay in the system in the system of FCFS bases.
3) The joint distribution of failure and repair times for each unit is taken to be bivariate
exponential having density function
4) Each repaired unit works as good as new
Special Issue Page 30 of 95 ISSN 2229 5216
Vol. 3, No.3, 2011

A
0
,B
0
a
A
r
,B
0
a
A
0
,B
fr
a
A
0
,B
0
v
A
F
,B
0
v
A
0
,B
F
v
A
F
,B
F
v
A
fR
,B
F
a
A
F
,B
F
a
A
F
,B
fR
a
X
1
X
2

Y
2

Y
2

Y
1

X
1
X
2

X
1

X
2

X
2

X
1

aY
1

aY
1

Y
1

S
0

S
4
S
3

S
1

S
2

S
7

S
6

S
5

S
8

S
9

TRANSITION DIAGRAM
Figure -1 TRANSITION DIAGRAM

bY
2

bY
2

Vol. 3, No.3, 2011

2. Notations

For defining the states of the system we assume the following symbols:
A
0
: unit A is in operative mode
B
0
: unit B is in operative mode
A
fr :
Unit A is in failure mode
B
fr :
Unit B is in failure mode
e
:

constant rate of repairmans availability
u
:
constant rate of repairmans not availability
A
F:
unit A in failure mode but in waiting for repairman
B
F:
unit B in failure mode but in waiting for repairman
X
i
(i=1,2): random variables representing the failure times of A
and B unit respectively for i=1,2
Y
i
(i=1,2): random variables representing the repair times of A and
B unit respectively for i=1,2
f
i
(x,y): joint pdf of (x
i
,y
i
);i=1,2

0
0 2
0
(1 ) (2 ( ); , , , 0; 0 1
( )
(2 )
( !)
i i
x y
i i i i i i i i i
j
i i i
i i i
j
r e I r xy X Y r
where
r xy
I r xy
j
o |
o | o | o |
o |
o |

=
= > s <
=

k
i
(Y/X): conditional pdf of Y
i
given X
i
=x is given by
=
0
(2 ( )
i i i
r x y
i i i i
e I r xy
o |
| o |

g
i
(.):
marginal pdf of X
i
=
(1 )
(1 )
i i
r x
i i
r e
o
o

h
i
(.):
marginal pdf of Y
i
=
(1 )
(1 )
i i
r y
i i
r e
|
|

(.), (.) :
ij ij
q Q

pdf &cdf of transition time from regenerative states pdf
&cdf of transition time from regenerative state S
i
to S
j
.
:
i
Mean sojourn time in state S
i
.
:

symbol of ordinary Convolution
( ) A t
0
( ) ( ) ( )
t
B t A t u B u du =
}

: symbol of stieltjes convolution
( ) A t
0
( ) ( ) ( )
t
B t A t u dB u =
}

Vol. 3, No.3, 2011

3. Transition Probability and Sojourn Times

The steady state transition probability can be as follows
1 1
01
(1 ) r
p
o
| u
=
+
,
2 2
02
(1 ) r
p
o
| u
=
+
,
05
p
u
| u
=
+

2 2
12.3
2 2 1 1
(1 )
(1 ) (1 )
r
p
r r
o
o |
=
+
,
1 1
10
2 2 1 1
(1 )
(1 ) (1 )
r
p
r r
|
o |
=
+

2 2
20
2 2 1 1
(1 )
(1 ) (1 )
r
p
r r
|
| o
=
+
,
1 1
21.4
2 2 1 1
(1 )
(1 ) (1 )
r
p
r r
o
| o
=
+
,
1 1
57
(1 ) r
p
o
| e
=
+
,
2 2
56
(1 ) r
p
o
| e
=
+
,
50
p
e
| e
=
+
,
1 1
62.89
1 1
(1 )
(1 )
a r
p
r
o
o e
=
+
,
1 1
61.89
1 1
(1 )
(1 )
b r
p
r
o
o e
=
+
62
1 1
(1 )
p
r
e
o e
=
+
,
71
2 2
(1 )
p
r
e
o e
=
+
,
2 2
72.89
2 2
(1 )
(1 )
a r
p
r
o
o e
=
+
,
2 2
71.89
2 2
(1 )
(1 )
b r
p
r
o
o e
=
+

(3.1-3.16)

here we get,

01 02 05
21.4 20
10 12.3
50 56 57
62 62.89 61.89
71 72.89 71.89
1
1
1
1
1
1
p p p
p p
p p
p p p
p p p
p p p
+ + =
+ =
+ =
+ + =
+ + =
+ + =

0
1
| u
=
+
,
1
1 1 2 2
1
(1 ) (1 ) r r
| o
=
+
,
2
2 2 1 1
1
(1 ) (1 ) r r
| o
=
+
,
5
1
e |
=
+
,
6
1 1
1
(1 ) r
e o
=
+
,

7
2 2
1
(1 ) r
e o
=
+
,
(3.17-3.28)

Vol. 3, No.3, 2011

4. Analysis of Characteristics

4.1 Mean Time to System Failure:

To determine the MTSF of the system, we regard the failed state of the system as absorbing
state, by probabilistic arguments, we get
0 01 1 02 2 05 5
1 10 0 13
2 20 0 24
5 50 0 57 7 56 6
6 68 2 68
7 71 1 78
( ) ( ) ( ) ( ) ( ) ( )
( ) ( )
( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( )
( ) ( )
t Q t Q t t Q t t
t Q t Q
t Q t Q
t Q t t Q t Q t t
t Q t Q
t Q t Q
| | | |
| |
| |
| | | |
| |
| |
= + +
= +
= +
= + +
= +
= +

(4.1-4.6)
Taking laplace stieltjes transforms of these relations and solving for
**
0
( ) s | ,
**
0
( )
( )
( )
N s
s
D s
| =

(4.7)

Where

0 1 01 05 71 57 2 02 05 56 62 5 05 7 05 57 6 05 56
( ) ( ) N p p p p p p p p p p p p p = + + + + + + + .

(4.8)

01 10 02 20 05 05 57 71 10 05 56 62 20
1 (1 ) (1 ) D p p p p p p p p p p p p p = + +
(4.9)

4.2 Availability Analysis:

Let ( )
i
A t be the probability that the system is in up-state at instant t given that the system entered
regenerative state i at t=0.using the arguments of the theory of a regenerative process the point wise
availability ( )
i
A t is seen to satisfy the following recursive relations
0 0 01 1 02 2 05 5
1 1 10 0 12.3 2
2 2 20 0 21.4 1
5 5 50 0 56 6 57 7
6 6 62.89
( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( )
A t M t q t A t q t A t q t A t
A t M t q t A t q t A t
A t M t q t A t q t A t
A t M t q t
= + + +
= + +
= + +
= + + +
= +
2 62 2 61.89 1
2 7 72.89 2 71 1 71.89 1
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
A t q t A t q t A t
+ +
= + + +

(4.10-4.16)
Now taking Laplace transform of these equations and solving them for
*
0
( ), A s
We get
* 1
0
1
( )
( )
( )
N s
A s
D s
=

(4.17)
The steady state availability is
* 1
0 0
0
1
( ( ))
lim
s
N
A sA s
D
= =

Where
Vol. 3, No.3, 2011

1 05 57 1 21.4 72.89 71 71.89 2 72.89 7 2 12.3 71 71.89 7 12.3 21.4
05 56 1 61.89 62 62..89 21.4 2 62 62..89 6 2 61.89 12.3 6 12.3 21.4
05 5 12.3 21.4 12.3 21.4
[ ( ) ( ) ]
[ ( ( ) ) ( ) ]
(1 ) (1 )
N p p p p p p p p p p p p
p p p p p p p p p p p p
p p p p p

= + + + + + +
+ + + + + + +
+
0 01 1 2 12.3 62 2 1 21.4
( ) ( ) p p p p + + +
(4.18)
1 10 20 10 20 0 5 05 6 56 05 7 57 05 1 57 05 20 72.89
05 56 20 20 61.89 01 02 01 20 2 57 05 10 61.89
05 57 10 10 72.89 01 02 01 10
( )( ) [( (1 )
(1 ) ( )] [( (1 )
(1 ) ( )
D p p p p p p p p p p p p p
p p p p p p p p p p p p p
p p p p p p p p p

= + + + + +
+ + + + +
+ + + +
(4.19)

4.3 Busy Period Analysis Of The Repairman:

Let B
i
(t) be the probability that the repairman is busy at instant t, given that the system entered
regenerative state I at t=0. By probabilistic arguments we have the following recursive relations for
Bi

(t)
0 01 1 02 2 05 5
1 1 10 0 12.3 2
2 2 20 0 21.4 1
5 50 0 57 7 56 6
6 62 2 62.89 2
( ) ( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( )
B t q t B t q t B t q t B t
B t W t q t B t q t B t
B t W t q t B t q t B t
B t q t B t q t B t
= + +
= + +
= + +
= + +
= +
61.89 1
6 71 1 72.89 2 71.89 1
( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( )
q t B t
+
= + +

(4.20-4.25)
Taking laplace transform of the equations of busy period analysis and solving them for
*
0
( ) B s ,we
get
* 2
0
1
( )
( )
( )
N s
B s
D s
=

(4.26)
In the steady state
* 2
0 0
0
1
( ( ))
lim
s
N
B sB s
D
= =

(4.27)

Where
2 1 01 21.4 02 61.89 05 21.4 56 05 61.89 56 05 21.4 57 72.89 05 71 57 05 72.89 57
2 12.3 01 02 05 56 62.89 62 05 56 12.3 61.89 05 57 72.89 05 12.3 57 71 71.89
[( (1 ) )]
( ( ) ( ))
N p p p p p p p p p p p p p p p p p p p p
p p p p p p p p p p p p p p p p p p p
= + + + + + +
+ + + + + + +
(4.28)
D
1
is already specified

4.4 Expected Number of Visits by the Repairman

We defined as the expected number of visits by the repairman in (0,t],given that the system
initially starts from regenerative state S
i
By probabilistic arguments we have the following recursive relations for ( )
i
V t

0 01 1 02 2 05 5
1 10 0 12.3 2
2 20 0 21.4 1
5 50 0 57 7 56 6
6 62 2 62.89 2 61
( ) ( ) (1 ( )) ( ) (1 ( )) ( ) ( )
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( )
( ) ( ) ( ) ( ) ( )
V t Q t V t Q t V t Q t V t
V t Q t V t Q t V t
V t Q t V t Q t V t
V t Q t V t Q t V t Q t V t
V t Q t V t Q t V t Q
= + + + +
= +
= +
= + +
= + +
.89 1
7 71 1 72.89 2 72.89 2
( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( )
t V t
V t Q t V t Q t V t Q t V t = + +

(4.29-4.34)
Vol. 3, No.3, 2011

Taking laplace stieltjes transform of the equations of expected number of visits
And solving them for
**
0
( ) V s , we get
** 3
0
1
( )
( )
( )
N s
V s
D s
=

(4.35)
In steady state
* 3
0 0
0
1
( ( ))
lim
s
N
V sV s
D
= =

(4.36)
Where
3 05 12.3 21.4
(1 )(1 ) N p p p = (4.37)
D
1
is already specified

4.5 Profit Analysis

The expected total profit incurred to the system in steady state is given by
0 0 1 0 2 0
P C A C B C V = (4.38)
Where
0
C =revenue/unit uptime of the system
1
C =cost/unit time for which repairman is busy
2
C =cost/visit for the repairman

5. Graphical Study of System Behavior

For a more clear view of the system characteristics w.r.t. the various parameters involved, we
plot curves for MTSF and profit function in figure-2 and figure-3 w.r.t the failure paprameter (
1
o )
of unit A for three different values of correlation coefficient (r
11
=0.25, r
12
=0.50, r
13
=0.75), between
X
1
and Y1 and two different values of repair parameter (
1
| ) of unit A while the other parameters
are kept fixed as
2 1 2
0 1 2
.005, .03, .02
500, 300, 50 C C C
o | | = = =
= = =

From the figure-2 it is observed that MTSF decreases as failure rate increases irrespective of other
parameters. the curves also indicates that for the same value of failure rate,MTSF is higher for
higher vaues of correlation coefficient(r),so here we conclude that the high value of r between
failure and repair tends to increase the expected life time of the system.
From the figure-3 it is clear that profit decreases linearly as failure rate increases. Also for the fixed
value of failure rate, the profit is higher for high correlation (r).
Following can also be interpreted from figure-3
For r
11
=0.25 the profit is depending upon whether the failure rate is < or =or> 0.0685, so in this case
the system is profitable only if the failure rate is less than 0.0685
For r
11
=0.50 the profit is depending upon whether the failure rate is < or =or> 0.0793, so in this
case the system is profitable only if the failure rate is less than 0.0793
For r
11
=0.75 the profit is depending upon whether the failure rate is < or =or> 0.0884, so in this case
the system is profitable only if the failure rate is less than 0.0884
So finally we conclude that the high correlation between failure and repair of the system yields the
better system performance.
Vol. 3, No.3, 2011

0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0
500
1000
1500
2000
2500
3000
3500
4000
4500
M
T
S
F
failure rate
MTSF vs failure rate for different values of repair rate and
correlation coefficient
r11=0.25
r12=0.50
r13=0.75

0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
-300
-200
-100
0
100
200
300
400
500
600
700
P
r
o
f
i
t
Failure rate of unit A
Profir vs Failure rate
r11=0.25
r12=0.50
r13=0.75

Figure -2

Figure -3

Vol. 3, No.3, 2011

References

[1] L.P.Goel, R.Gupta and S.k.singh, cost analysis of two-unit priority standby system with
imperfect switch and Arbitrary distribution, microelectron.reliab.25(1), (1985) 65-69.
[2] L.R.goel, P.Srivastava and R.Gupta, two unit cold standby system with correlated failures
and repairs, int.j. of system science,vol.23(3), (1992) p.379-391
[3] V.goyal and K.murari,cost anlysis of a two unit standby system with regular repairman and
patience time , microelectron.reliab.25(1), (1984) 453-459.
[4] L.P.Goel,R.Gupta,Cost benefit analysis of two unit parallel system with administrative delay
in repair,Int.j.of system science,vol.21,(1990),1369-1379
[5] R.Gupta, C.K.Goel and A. tomer. , Analysis of two unit stand by system with correlated
repair and failure with random appearance and disappearance of repairman, vol.3, (2010), 53-61
[6] R.gupta, K.Kumar, A two unit complex system with correlated failure and repair times. Pure
and applied mathematica sciences,LXVII(1-2), ,(2008) 23-34
[7] R.Gupta,V. Sharma and N.Rastogi,cost benefit analysis of a three unit redundant system with
correlated failure and repair times,int.journal of agricultural statistical sciences,vol.2 no.-
1(2006) p.-71-81
[8] A.Chaudhary,Neeraj and K.Kumar,Profit analysis of a complex system with correlation in
time to preventive maintenance and time taken in preventive maintenance,jrss,vol.3,(2010),95-
103

Paper 6

Uploaded by

Document Information

Original Description:

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

Paper 6

Uploaded by

Copyright:

Available Formats

International Journal of Advances in Science and Technology,

Vol. 3, No.3, 2011

You might also like