International Journal of Mathematics and Statistics Invention (IJMSI)

E-ISSN: 2321 4767 P-ISSN: 2321 - 4759

www.ijmsi.org Volume 1 Issue 2 December. 2013 PP 01-15
www.ijmsi.org 1 | P a g e
Perishable Inventory System with a Finite Population and
Repeated Attempts

K. Jeganathan And N. Anbazhagan
Department of Mathematics, Alagappa University, Karaikudi-630 004, Tamil Nadu, India.


ABSTRACT : In this article, we consider a two commodity continuous review perishable inventory system
with a finite number of homogeneous sources of demands. The maximum storage capacity of
i
S units for the
th
i commodity 1,2) = (i . The life time of items of each commodity is assumed to be exponentially distributed
with parameter
i
1,2) = (i . The time points of primary demand occurrences form independent quasi random
distributions each with parameter
i
1,2). = ( i A joint reordering policy is adopted with a random lead time
for orders with exponential distribution. When the inventory position of both commodities are zero, any arriving
primary demand enters into an orbit. The demands in the orbit send out signal to compete for their demand
which is distributed as exponential. We assume that the two commodities are both way substitutable. The joint
probability distribution for both commodities and number of demands in the orbit is obtained for the steady
state case. Various system performance measures are derived and the results are illustrated with numerical
examples.

KEYWORDS: Retrial Demand, Positive Lead-Time, Finite Population, Perishable Inventory, Substitutable,
Markov Process, Continuous Review.

I. INTRODUCTION
The analysis of perishable inventory systems has been the theme of many articles due to its potential
applications in sectors like food industries, drug industries, chemical industries, photographic materials,
pharmaceuticals, blood bank management and even electronic items such as memory chips. The often quoted
review articles ([21], [23]) and the recent review articles ([24], [16]) provide excellent summaries of many of
these modelling efforts. Most of these models deal with either the periodic review systems with fixed life times
or continuous review systems with instantaneous supply of reorders. One of the factors that contribute the
complexity of the present day inventory system is the multitude of items stocked and this necessitated the multi-
commodity systems. In dealing with such systems, in the earlier days models were proposed with independently
established reorder points. But in situations were several product compete for limited storage space or share the
same transport facility or are produced on (procured from) the same equipment (supplier) the above strategy
overlooks the potential savings associated with joint ordering and, hence, will not be optimal. Thus, the
coordinated, or what is known as joint replenishment, reduces the ordering and setup costs and allows the user
to take advantage of quantity discounts [17].

Inventory system with multiple items have been subject matter for many investigators in the past. Such
studies vary from simple extensions of EOQ analysis to sophisticated stochastic models. References may be
found in ([7], [17], [20], [22], [25], [28]) and the references therein. Multi commodity inventory system has
received more attention on the researchers on the last five decades. In continuous review inventory systems,
Ballintfy [11] and Silver [25] have considered a coordinated reordering policy which is represented by the triplet
) , , ( s c S , where the three parameters
i
S ,
i
c and
i
s are specified for each item i with
i i i
S c s s s , under the
unit sized Poisson demand and constant lead time. In this policy, if the level of i th commodity at any time is
below
i
s , an order is placed for
i i
s S items and at the same time, any other item j ) ( i = with available
inventory at or below its can-order level
j
c , an order is placed so as to bring its level back to its maximum
capacity
j
S . Subsequently many articles have appeared with models involving the above policy and another
article of interest is due to Federgruen et al. [14], which deals with the general case of compound Poisson
demands and non-zero lead times. A review of inventory models under joint replenishment is provided by Goyal
and Satir [15].
Perishable Inventory System With A Finite..
www.ijmsi.org 2 | P a g e
Kalpakam and Arivarignan [17] have introduced ) , ( S s policy with a single reorder level s defined
in terms of the total number of items in the stock. This policy avoids separate ordering for each commodity and
hence a single processing of orders for both commodities has some advantages in situation where in
procurement is made from the same supplies, items are produced on the same machine, or items have to be
supplied by the same transport facility. Krishnamoorthy and Varghese [18] have considered a two commodity
inventory problem without lead time and with Markov shift in demand for the type of commodity namely
commodity-1, commodity-2 or both commodity, using the direct Markov renewal theoretical results.
Anbazhagan and Arivarignan ([3], [4], [5], [6]) have analyzed two commodity inventory system under various
ordering policies. Yadavalli et al. [29] have analyzed a model with joint ordering policy and varying order
quantities. Yadavalli et al. [30] have considered a two-commodity substitutable inventory system with Poisson
demands and arbitrarily distributed lead time. All the models considered in the two-commodity inventory
system assumed lost sales of demands during stock out periods.

Traditionally the inventory models incorporate the stream of customers (either at fixed time intervals
or random intervals of time) whose demands are satisfied by the items from the stock and those demands which
cannot be satisfied are either backlogged or lost. But in recent times due to the changes in business
environments in terms of technology such as Internet, the customer may retry for his requirements at random
time points. The concept of retrial demands in inventory was introduced in [9] and only few papers ([2], [26],
[27], [31] ) have appeared in this area. Moreover product such as bath soaps, body spray, etc., may have
different flavours and the customer would be willing to settle for one only when the other is not available. These
aspects provided the motivation for writing this paper. We will focus on the case in which the population under
study is finite so each individual customer generates his own flow of primary demands. For the analysis of finite
source retrial queue in continuous time, the interested reader is referred to Falin and Templeton [12], Artalejo
and Lopez-Herrero [10], Falin and Artalejo [13], Almasi et al. [1] and Artalejo [8] and references therein.
The rest of the paper is organized as follows. In the next section, we describe the mathematical model. The
steady state analysis of the model is presented in section 3 and some key system performance measures are
derived in section 4. In section 5, we calculate the total expected cost rate in the steady state. Several
numerical results that illustrate the influence of the system parameters on the system performance are discussed
in section 6. The last section is meant for conclusion.

II. MATHEMATICAL MODEL
We consider a continuous review perishable inventory system with a maximum stock of
i
S units for
the i th commodity 1,2) = (i and the demands originated from a finite population of sources N . Each source
is either free or in the orbit at any time. The primary demand for
th
i commodity is of unit size and the time
points of primary demand occurrences form independent Quasi-random distributions each with parameter
i

1,2) = (i . The items are perishable in nature and the life time of items of each commodity is assumed to be
exponentially distributed with parameter
i
1,2) = (i . The reorder level for the
th
i commodity is fixed as
i
s
) (1
i i
S s s s and an order is placed for both commodities when both the inventory levels are less than or equal
to their respective reorder levels. The ordering quantity for the
th
i commodity is
1,2) = 1, > (= i s s S Q
i i i i
+ items. The requirement 1 > +
i i i
s s S , ensures that after a replenishment
the inventory level will always be above the respective reorder levels; otherwise it may not be possible to place
reorder (according to this policy) which leads to perpetual shortage. The lead time is assumed to be
exponentially distributed with parameter 0 > | . Both the commodities are assumed to be both way
substitutable in the sense that at the time of zero stock of any one commodity, the other one is used to meet the
demand. If the inventory position of both the commodities are zero thereafter any arriving primary demand
enters into the orbit. These orbiting demands send out signal to compete for their demand which is distributed as
exponential with parameter 0) (> u . In this article, the classical retrial policy is followed, that is, the probability
of a repeated attempt is depend on the number of demands in the orbit. The retrial demand may accept an item
of commodity- i with probability
i
p 1,2) = (i , where 1 =
2 1
p p + . We also assume that the inter demand
times between the primary demands, lead times, life time of each items and retrial demand times are mutually
independent random variables.

Perishable Inventory System With A Finite..
www.ijmsi.org 3 | P a g e
2.1 Notations:
ones all containing dimension e appropriat of tor column vec a : e
matrix Zero : 0
| | A matrix a of position ) , ( at entry :
th
ij
j i A
j i i k V k
j
i
1, , = : + e

ij ij
o o 1 :

otherwise 0
= if 1
:
i j
ij
o

>
. 0,
0, 1,
: ) (
otherwise
x if
x H
} , {0,1,2, :
1 1
S E
} , {0,1,2, :
2 2
S E
} , {0,1,2, :
3
N E

3 2 1
: E E E E

III. Analysis
Let ) (
1
t L , ) (
2
t L and ) ( t X denote the inventory position of commodity-I, the inventory position of
commodity-II and the number of demands in the orbit at time , t respectively. From the assumptions made on
the input and output processes it can be shown that the triplet 0} )), ( ), ( ), ( {(
2 1
> t t X t L t L is a continuous
time Markov chain with state space given by . E To determine the infinitesimal generator
)) , , ( ), , , (( =
3 2 1 3 2 1
j j j i i i a O , E j j j i i i e ) , , ( ), , , (
3 2 1 3 2 1
, of this process.

Theorem 1:
The infinitesimal generator of this Markov process is given by,


= ) ) , , , ( ), , , ( (
3 2 1 3 2 1
j j j i i i a
Perishable Inventory System With A Finite..
www.ijmsi.org 4 | P a g e

e e +
+ + +
e e e
+ +
e e e
+ +
e e e
+ +
e e e

e e e

e e

e e

e
+ +

,
, , 0, = ), ) (
, = , = , = ) )( ((
, , ,
, = 1, = , = , ) )( (
, , ,
, = , = 1, = , ) )( (
, , ,
, = , = , =
, , ,
1, = 1, = , = ,
, , ,
1, = , = 1, = ,
, 0, = ,
1, = , = 1, =
or
, , 0, =
1, = 1, = , = ,
, 0, = 0, =
1, = , 0 = , 0 = ), )( (
0 3
2
0 2 1 2 2 3 0
2
3 3 2 2 1 1 2 2 2 1 3
0 3
2
1 2
1
0 1
3 3 2 2 1 1 2 2 2 1 3
0 3
2
0 2
1
1 1
3 3 2 2 1 1 1 1 2 1 3
0 3
2
0 2
1
0 1
3 3 2 2 2 1 1 1
1 3
2
1 2
1
1 1
3 3 2 2 1 1 2 3
1 3
2
1 2
1
1 1
3 3 2 2 1 1 1 3
1 3 2
1
1 1
3 3 2 2 1 1
1 3
2
1 2 1
3 3 2 2 1 1 3
1
0 3 2 1
3 3 2 1 2 1 3
N
S
i
N
S S
N
S S
N
s s
N
S S
N
S S
N
S
N
S
N
V i V i i i s H i
i j i j i j i i N
V i V i V i
i j i j i j i i N
V i V i V i
i j i j i j i i N
V i V i V i
i j Q i j Q i j
V i V i V i
i j i j i j p i
V i V i V i
i j i j i j p i
V i i V i
i j i j i j
V i V i i
i j i j i j i
V i i i
i j j j i N
| u o



|
u
u
u

e e e +
+ + + +
otherwise
V i V i V i i s H i s H i
i j i j i j i i i N
N
S S
0
, , , )), ( ) (
, = , = , = ) )( ((
0 3
2
0 2
1
1 1 2 2 1 1 3
3 3 2 2 1 1 2 2 1 1 2 1 3
| u


Proof:
The infinitesimal generator )) , , ( ), , , ((
3 2 1 3 2 1
j j j i i i a of this process can be obtained using the
following arguments:

1: Let 0 0, > 0, >
3 2 1
> i i i .
A primary demand from any one of the ) (
3
i N sources or due to perishability takes the inventory level
) , , (
3 2 1
i i i to ) , 1, (
3 2 1
i i i with intensity
1 1 1 3
) ( i i N + for I-commodity or ) , , (
3 2 1
i i i to ) 1, , (
3 2 1
i i i
with intensity
2 2 2 3
) ( i i N + for II-commodity.

The level ) , (0,
3 2
i i , and ) ,0, (
3 1
i i , respectively, is taken to ) 1, (0,
3 2
i i , with intensity
2 2 2 1 3
) )( ( i i N + + , and ) 1,0, (
3 1
i i with intensity
1 1 2 1 3
) )( ( i i N + + .

2: If the inventory position of both the commodities are zero then any arriving primary demand enters
into the orbit. Hence a transition takes place from ) (0,0,
3
i to 1) (0,0,
3
+ i with intensity ) )( (
2 1 3
+ i N ,
Perishable Inventory System With A Finite..
www.ijmsi.org 5 | P a g e
1 0
3
s s N i .

3: Let 1 0, > 0, >
3 2 1
> i i i .

A demand from orbit takes the inventory level ) , , (
3 2 1
i i i to 1) , 1, (
3 2 1
i i i with intensity
1 3
p i u for I-
commodity or ) , , (
3 2 1
i i i to 1) 1, , (
3 2 1
i i i with intensity
2 3
p i u for II-commodity.

The level ) , (0,
3 2
i i , and ) ,0, (
3 1
i i , respectively, is taken to 1) 1, (0,
3 2
i i and 1) 1,0, (
3 1
i i with
intensity u
3
i .

4: From a state ) , , (
3 2 1
i i i with ) , ( ) , (
2 1 2 1
s s i i s , 0
3
> i a replenishment by the delivery of orders
for both commodities takes the inventory level to ) , , (
3 2 2 1 1
i Q i Q i + + ,
1 1 1
= s S Q ,
2 2 2
= s S Q , with
intensity of this transition | .

We observe that no transition other than the above is possible.

Finally the value of ) ) , , ( ), , , ( (
3 2 1 3 2 1
j j j i i i a is obtained by
)) , , ( ), , , (( = )) , , ( ), , , ((
3 2 1 3 2 1
3 2 1
)
3
,
2
,
1
( )
3
,
2
,
1
(
3 2 1 3 2 1
j j j i i i a i i i i i i a
j j j
j j j i i i

=


Hence we get the infinitesimal generator ). ) , , ( ), , , ( (
3 2 1 3 2 1
j j j i i i a

In order to write down the infinitesimal generator O in a matrix form, we arrange the states in
lexicographic order and group 1) 1)( (
2
+ + N S states as:

,0), , ( , ), ,1, ( , ,1,1), ( ,1,0), ( ), ,0, ( , ,0,1), ( ,0,0), (( >= <
2
S q N q q q N q q q q
. , 0,1, = )) , , ( , ,1), , (
1 2 2
S q for N S q S q

Then the rate matrix O has the block partitioned form with the following sub matrix [
1 1
]
j i
O at the
1
i -
the row and
1
j -th column position.

e +
e
e
O
. ,
, = ,
1, = ,
, = ,
= ] [
1
0 1 1 1 1
1
1 1 1 1
1
1
0 1 1 1
1
1 1
otherwise
V i Q i j C
V i i j B
V i i j A
s
S
i
S
i
j i
0


where

e +
. 0,
, =
= ] [
2
0 2 2 2 2
2 2
otherwise
V i Q i j W
C
s
j i

e + +
e + +

. 0,
, , = ), ) )( ((
, 1, = ), )( (
= ] [
0 3 3 3 2 1 3
1
0 3 3 3 2 1 3
3 3
otherwise
V i i j i N
V i i j i N
H
N
N
j i
|


Perishable Inventory System With A Finite..
www.ijmsi.org 6 | P a g e

e
e
. 0,
, , =
, 1, =
0, = 0, =
= ] [
2
1 2 2 2
2
2
1 2 2 2
2
2 2
2 2
0
otherwise
V i i j H
V i i j F
i j H
A
S
i
S
i
j i



1 1
, 1, = S i For

e
. 0,
, , =
0, = , =
= ] [
2
1 2 2 2
1
2 2 2
1
2 2 1
otherwise
V i i j M
i i j V
B
S
i
i
j i i



1 1
, 1,2, = S i For

e
e
. 0,
, , =
0, = , =
, 1, =
= ] [
2
1 2 2 2
2 1
2 2 2
1
2
1 2 2 2
2
2 2 1
otherwise
V i i j L
i i j J
V i i j G
A
S
i i
i
S
i
j i i



1 1
, 1,2, = S i For

e + +
e
. 0,
, , = , ) )( (
, 1, = ,
= ] [
0 3 3 3 1 1 2 1 3
1 3 3 3 3
3 3 1
otherwise
V i i j i i N
V i i j i
V
N
N
j i i

u



1 1
, 1,2, = S i For

e +
e
. 0,
, , = , ) (
, 1, = ,
= ] [
0 3 3 3 1 1 1 3
1 3 3 3 1
3 3 1
otherwise
V i i j i i N
V i i j p
M
N
N
j i i

u



2 2
, 1,2, = S i For

e + +
e
. 0,
, , = , ) )( (
, 1, = ,
= ] [
0 3 3 3 2 2 2 1 3
1 3 3 3 3
3 3 2
otherwise
V i i j i i N
V i i j i
F
N
N
j i i

u



2 2
, 1,2, = S i For

e +
+ + +
. 0,
, , = ), ) (
) )( ((
= ] [
0 3 3 3 2 2 3
2 2 2 1 3
3 3 2
otherwise
V i i j i s H i
i i N
H
N
j i i
| u




2 2
, 1,2, = S i For
Perishable Inventory System With A Finite..
www.ijmsi.org 7 | P a g e

e +
e
. 0,
, , = , ) (
, 1, = ,
= ] [
0 3 3 3 2 2 2 3
1 3 3 3 2 3
3 3 2
otherwise
V i i j i i N
V i i j p i
G
N
N
j i i

u



1 1
, 1,2, = S i For

e
+ + + +
. 0,
, , = )) (
) )( ((
= ] [
0 3 3 3 1 1
3 1 1 2 1 3
3 3 1
otherwise
V i i j i s H
i i i N
J
N
j i i
|
u


, , 1,2, = ; , 1,2, =
2 2 1 1
S i S i For

e +
+ + + +
. 0,
, , = )), ( ) (
) )( ((
= ] [
0 3 3 3 2 2 1 1 3
2 2 1 1 2 1 3
3 3 2 1
otherwise
V i i j i s H i s H i
i i i N
L
N
j i i i
| u




1
=
+ N
I W |

It may be noted that the matrices
1
i
A , ,
1
i
B
1 1
, 1,2, = S i ,
0
A and C are square matrices of order
1) 1)( (
2
+ + N S . The sub matrices
1
i
V ,
1
i
M ,
1
i
J , ,
2 1
i i
L
1 1
, 1,2, = S i , , , 1,2, =
2 2
S i W , H ,
2
i
F ,
2
i
H ,
2
i
G ,
2 2
, 1,2, = S i , are square matrices of order 1) ( + N .

It can be seen from the structure of O that the homogeneous Markov process
0} : )) ( ), ( ), ( {(
2 1
> t t X t L t L on the finite space E is irreducible, aperiodic and persistent non-null.

Hence the limiting distribution

(0)], (0), (0), | = ) ( , = ) ( , = ) ( [
lim
=
2 1 3 2 2 1 1
)
3
,
2
,
1
(
X L L i t X i t L i t L Pr
t
i i i

t exists.

Let ), , , , ( =
)
1
(
(1) (0)
S
H H H

partitioning the vector,
)
1
( i
H into as follows:


1 1
)
2
,
1
( ,2)
1
( ,1)
1
( ,0)
1
( )
1
(
, 0,1,2, = ), , , , , ( = S i
S i i i i i
H H H H H

which is partitioned as follows:

. , 0,1,2, = ; , 0,1,2, = ), , , ( =
2 2 1 1
) ,
2
,
1
( ,0)
2
,
1
( )
2
,
1
(
S i S i
N i i i i i i
t t H

Then the vector of limiting probabilities satisfies

1. = = e and 0 O (1)




Perishable Inventory System With A Finite..
www.ijmsi.org 8 | P a g e
Theorem 2:

The limiting distribution is given by,

, , 0,1, = , =
1 1
1
)
1
( )
1
(
S i
i
Q i
O H H
where

( ) |
( )|

, , 1, =
,
1) (
, = ,
1, , 0,1, = , 1) (
=
1 1 1
1
1
1
1
1
1 1
1
1
1 1 1 1
1
1 1
1
1 1 1
1
1
1
1
1
1
1 1
1 1
0 =
1
1
1 1
2
1 1
1 1
1
1
1
1
1
1
1
1
1 1
1 1
1
S Q i
A B B A B
CA A B B A B
Q i I
Q i A B B A B
i i j S j S j S
j S j s j s Q Q Q
i S
j
i Q
i i Q Q Q
i Q
i

(2)

The value of
)
1
( Q
H can be obtained from the relation

( ) |

1
1 1
1
1 1 1
1
1
1
1
1
1
1 1
1
1
0 =
1
1
)
1
(
1) (
j S j s j s Q Q Q
s
j
Q Q
CA A B B A B
( )|
1
1
1
1
1
1
2
1
1
1 1
1
1
1 1 1 1
Q Q Q Q j S j S j S
A B A B B A B +
+

+ +

(3)
| 0, = 1) (
1
0 1 1
1
1
1
1 1
1
C A B B A B
Q Q Q
Q

+
and
( )

+ H

+

I A B B A B
i i Q Q Q
i Q
Q
i
Q
1
1
1
1
1
1
1
1
1 1
1 1
1
1
0 =
1
)
1
(
1) (
( ) |

\
|
+

+
+

1
1 1
1
1 1 1
1
1
1
1
1
1
1 1
1 1
0 =
1
1
1 1
2
1
1
1
=
1
1) (
j S j s j s Q Q Q
i S
j
i Q
S
Q i
CA A B B A B (4)
( )|)| 1. =
1
1
1
1
1
1 1
1
1
1 1 1 1
e A B B A B
i i j S j S j S

+

Proof:

The first equation of (1) yields the following set of equations :
1, , 0,1, = 0, =
1 1
1
)
1
(
1
1
1)
1
(
H + H
+
+
Q i A B
i
i
i
i
(5)
, = 0, =
1 1
)
1 1
(
1
)
1
(
1
1
1)
1
(
Q i C A B
Q i
i
i
i
i
+
+
H + H + H (6)
1, , 1, = 0, =
1 1 1
)
1 1
(
1
)
1
(
1
1
1)
1
(
+ H + H + H

+
+
S Q i C A B
Q i
i
i
i
i
(7)
. = 0, =
1 1
)
1 1
(
1
)
1
(
S i C A
Q i
i
i
H + H (8)

Solving the above set of equations we get the required solution.

Perishable Inventory System With A Finite..
www.ijmsi.org 9 | P a g e

IV. SYSTEM PERFORMANCE MEASURES
In this section some performance measures of the system under consideration in the steady state are
derived.

4.1 Expected inventory level
Let
1
i
q and
2
i
q denote the average inventory level for the first commodity and the second commodity
respectively in the steady state. Then

)
3
,
2
,
1
(
1
0 =
3
2
0 =
2
1
1 =
1
1
=
i i i
N
i
S
i
S
i
i
i t q

(9)
and

)
3
,
2
,
1
(
2
0 =
3
2
1 =
2
1
0 =
1
2
=
i i i
N
i
S
i
S
i
i
i t q

(10)

4.2 Expected reorder rate
Let
r
q denote the mean reorder rate in the steady state. Then

,0)
2
1,
1
(
1 1 2 0
2
1
2
0 =
2
) 1) ( ( =
i s
i
s
i
r
s N N
+
+ + +

t o q

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

s i
i
s
i
s N N t o

)
3
,
2
1,
1
(
1 3 2 3 2 3 0
2
1 1 1 3
1 =
3
2
0 =
2
) ) ) (( 1) ( ) ((
i i s
i
N
i
s
i
p i p i i N s i N
+
+ + + + + +

t u u o
(11)

)
3
1,
2
,
1
(
2 3 1 3 1 3 0
1
2 2 2 3
1 =
3
1
0 =
1
) ) ) (( 1) ( ) ((
i s i
i
N
i
s
i
p i p i i N s i N
+
+ + + + + +

t u u o

4.3 Expected perishable rate
Let
1
p
q and
2
p
q denote the expected perishable rates for the first commodity and the second
commodity respectively in the steady state. Then

)
3
,
2
,
1
(
1 1
0 =
3
2
0 =
2
1
1 =
1
1
=
i i i
N
i
S
i
S
i
p
i t q

(12)
and

)
3
,
2
,
1
(
2 2
0 =
3
2
1 =
2
1
0 =
1
2
=
i i i
N
i
S
i
S
i
p
i t q

(13)


4.4 Expected number of demands in the orbit
Let
o
q denote the expected number of demands in the orbit. Then

)
3
,
2
,
1
(
3
1 =
3
2
0 =
2
1
0 =
1
=
i i i
N
i
S
i
S
i
o
i t q

(14)




Perishable Inventory System With A Finite..
www.ijmsi.org 10 | P a g e
4.5 Expected an arriving demand enters into the orbit
The expected an arriving primary demand enters into the orbit is given by

)
3
(0,0,
2 1 3
1
0 =
3
) )( ( =
i
N
i
a
i N t q +

(15)
4.6 The overall rate of retrials
The overall rate of retrials for the orbit customers in the steady state. Then

)
3
,
2
,
1
(
3
1 =
3
2
0 =
2
1
0 =
1
=
i i i
N
i
S
i
S
i
or
i ut q

(16)

4.7 The successful rate of retrials
The successful rate of retrials for the orbit customers in the steady state. Then

)
3
,
2
,
1
(
3
1 =
3
2
1 =
2
1
1 =
1
)
3
,0,
1
(
3
1 =
3
1
1 =
1
)
3
,
2
(0,
3
1 =
3
2
1 =
2
=
i i i
N
i
S
i
S
i
i i
N
i
S
i
i i
N
i
S
i
sr
i i i ut ut ut q

+ + (17)

4.8 Fraction of successful rate of retrials
Let
fr
q denote the fraction of successful rate of retrials is given by

or
sr
fr
q
q
q = (18)

V. COST ANALYSIS
To compute the total expected cost per unit time (total expected cost rate),
the following costs, are considered.

1 h
c : The inventory holding cost per unit item per unit time for I-commodity.
2 h
c : The inventory holding cost per unit item per unit time for II-commodity.
s
c : The setup cost per order.
1 p
c : Perishable cost of the I - commodity per unit item per unit time.
2 p
c : Perishable cost of the II- commodity per unit item per unit time.
w
c : Waiting cost of an orbiting demand per unit time.

The long run total expected cost rate is given by

. = ) , , , , (
2 2 1 1 2 2 1 1
2 1 2 1 o w p p p p r s i h i h
c c c c c c N s s S S TC q q q q q q + + + + +
(19)

Substituting the values of q s we get ) , , , , (
2 1 2 1
N s s S S TC


(
(

+
(
(

)
3
,
2
,
1
(
2
0 =
3
2
1 =
2
1
0 =
1
2
)
3
,
2
,
1
(
1
0 =
3
2
0 =
2
1
1 =
1
1
=
i i i
N
i
S
i
S
i
h
i i i
N
i
S
i
S
i
h
i c i c t t

(
(

+
(
(

+

)
3
,
2
,
1
(
1 1
0 =
3
2
0 =
2
1
1 =
1
1
)
3
,
2
,
1
(
3
1 =
3
2
0 =
2
1
0 =
1
i i i
N
i
S
i
S
i
p
i i i
N
i
S
i
S
i
w
i c i c t t
+
(
(

+

)
3
,
2
,
1
(
2 2
0 =
3
2
1 =
2
1
0 =
1
2
i i i
N
i
S
i
S
i
p
i c t

|
|
.
|

\
|
+ + +
+

,0)
2
1,
1
(
1 1 2 0
2
1
2
0 =
2
) 1) ( (
i s
i
s
i
s
s N N c t o
Perishable Inventory System With A Finite..
www.ijmsi.org 11 | P a g e

|
|
.
|

\
|
+ + + +
+

1,0)
2
,
1
(
2 2 1 0
1
2
1
0 =
1
1) ( (
s i
i
s
i
s N N t o (20)

|
|
.
|

\
|
+ + + + + +

)) ) (( 1) ( ) ((
2 3 2 3 0
2
1 1 1 3 1 3
1 =
3
2
0 =
2
p i i N s p i i N
i
N
i
s
i
u o u

(
(

(
|
|
.
|

\
|
+ + + + + +
+

)
3
1,
2
,
1
(
1 3 1 3 0
1
2 2 2 3 2 3
1 =
3
1
0 =
1
)) ) (( 1) ( ) ((
i s i
i
N
i
s
i
p i i N s p i i N t u o u

Due to the complex form of the limiting distribution, it is difficult to discuss the properties of the cost
function analytically. Hence, a detailed computational study of the cost function is carried out in the next
section.

VI. NUMERICAL ILLUSTRATIONS
In this section we discuss some interesting numerical examples that qualitatively describe the
performance of this inventory model under study. Our experience with considerable numerical examples
indicates that the function ), , (
2 1
S S TC to be convex. Appropriate numerical search procedures are used to
obtain the optimal values of , TC
1
S and
2
S (say ,
*
TC
*
1
S and
*
2
S ). The effect of varying the system
parameters and costs on the optimal values have been studied and the results agreed with what one would
expect. A typical three dimensional plot of the total expected cost function is given in Figure 1 .In Table 1 gives
the total expected cost rate as a function of
*
1
S and
*
2
S by fixing the parameters and the cost values:
2, =
1
s 3, =
2
s 10, = N 0.01, =
1
0.02, =
2
0.01, = u 0.2, =
1
0.1, =
2
0.02, = | 12, =
s
c
0.01, =
1 h
c 0.04, =
2 h
c 0.4, =
1 p
c 0.5, =
2 p
c 6, =
w
c 0.4 =
1
p and 0.6 =
2
p .
From the Table 1 the total expected cost rate is more sensitive to the changes in
*
2
S than that of in
*
1
S .
Some of the results are presented in Tables 2 through 6 where the lower entry in each cell gives the total
expected cost rate and the upper entries the corresponding
*
1
S and
*
2
S .


Perishable Inventory System With A Finite..
www.ijmsi.org 12 | P a g e
2, =
1
s 3, =
2
s 10, = N 0.01, =
1
0.02, =
2
0.01, = u 0.2, =
1
0.1, =
2
0.02, = |
12, =
s
c
0.01, =
1 h
c 0.04, =
2 h
c 0.4, =
1 p
c 0.5, =
2 p
c 6, =
w
c 0.4, =
1
p 0.6 =
2
p .
Figure 1: A three dimensional plot of the cost function ) , (
2 1
S S TC

Table 1: Total expected cost rate as a function of
1
S and
2
S


6.1 Example 1
In the first example, we look at the impact of
1
,
2
,
1
, and
2
on the optimal values ) , (
*
2
*
1
S S
and the corresponding total expected cost rate
*
TC . For this, first by fixing the parameters and cost values as
2, =
1
s 3, =
2
s 10, = N 0.02 = | , 0.01 = u , 0.4 =
1
p , 0.6 =
2
p , 0.01 =
1 h
c , 0.04 =
2 h
c , 12 =
s
c ,
6 =
w
c , 0.4 =
1 p
c and 0.5 =
2 p
c . Observe the following from Tables 2 and 3 :

1. From the Table 2 , it is observed that the
*
TC ,
*
1
S and
*
2
S increase when
1
and
2
increase. The result
is obvious as
1
and
2
increase it has impact on higher re-ordering and the cost of carrying to orbit
customers. Hence arrival rates are vital to this system. Also the
*
TC is more sensitive to changes in
1
than
that of in
2
.
2. From the Table 3 , it is observed that if
1
and
2
increase then
*
1
S and
*
2
S decrease, and the
*
TC
increases, in a significant amount. This results is obvious as
1
and
2
increase, more items will be perished
that finally incurred a substantial amount of costs to the system. From the observation it seems that the
*
TC is
very sensitive to changes in
2
than that of in
1
.

6.2 Example 2
In this example, we study the impact of
s
c ,
1 h
c ,
2 h
c ,
1 p
c ,
2 p
c and
w
c on the optimal values
) , (
*
2
*
1
S S and the corresponding
*
TC . Towards this end, first by fixing the parameter values as
2, =
1
s 3, =
2
s 10, = N 0.01 =
1
, 0.02 =
2
, 0.02 = | , 0.01 = u , 0.2 =
1
, 0.1 =
2
, 0.4 =
1
p
and 0.6 =
2
p .
Observe the following from Tables 6 4 :

1. The total expected cost rate increases when
1 h
c ,
2 h
c ,
s
c ,
w
c ,
1 p
c and
2 p
c increase monotonically.

2. As
1 h
c and
2 h
c increase, the optimal values
*
1
S and
*
2
S decrease monotonically. This is to be expected
since
1 h
c and
2 h
c increase, we resort to maintain low stock in the inventory.

3. Similarly, when
w
c increases, the values of
*
1
S and
*
2
S increase monotonically. This is because if
w
c
increases then we have to maintain high inventory to reduce the number of waiting customers in the orbit.

2
S
1
S
29 30 31 32 33
88 52.866093 52.863418 52.862517 52.863258 52.865527
89 52.866088 52.863289 52.862247 52.862850 52.864984
90 52.866167 52.863315 52.862134 52.862599 52.864598
91 52.866586 52.863494 52.862227 52.862502 52.864368
92 52.867052 52.8638820 52.862361 52.862555 52.864288
Perishable Inventory System With A Finite..
www.ijmsi.org 13 | P a g e
4. As
*
1
S and
*
2
S increase monotonically,
s
c increases. This is a common decision that we have to maintain
more stock to avoid frequent ordering.

5. If
1 p
c and
2 p
c increase monotonically then
*
1
S and
*
2
S decrease and
*
TC increases. We also note that the
total expected cost rate is more sensitive to changes in
1 p
c than that of in
2 p
c .

Table 2: Sensitivity of
1
and
2
on the optimal values


Table 3: Variation in optimal values for different values of
1
and
2




Table 4: Effect of varying
1 h
c and
2 h
c on the optimal values


2

1

0.010 0.015 0.020 0.025 0.030
0.005 104 28 104 30 104 30 105 33 105 34
52.166935 52.404997 52.616266 52.805787 52.977922
0.010 104 28 104 30 104 31 105 33 105 34
52.471973 52.677218 52.862134 53.030387 53.184492
0.015 105 28 105 30 105 31 105 33 106 34
52.735666 52.916048 53.080348 53.231281 53.370828
0.020 106 28 106 30 106 31 106 34 106 34
52.967782 53.128612 53.276383 53.413216 53.540578
0.025 107 28 107 30 107 32 107 34 107 35
53.174930 53.319961 53.454181 53.579277 53.696383
2

1

0.06 0.08 0.10 0.12 0.14
0.20

105 31 105 31 104 31 102 30 100 26
50.734492 51.983923 52.862134 53.512248 53.989839
0.25

99 31 95 31 90 31 87 30 86 26
51.777332 53.098322 54.023159 54.718720 55.247406
0.30

67 27 65 27 62 27 60 26 60 23
52.403399 53.768192 54.715274 55.438513 56.014391
0.35

49 26 47 24 44 24 42 24 41 23
52.802492 54.198838 55.153269 55.880546 56.468607
0.40 37 25 35 22 32 21 31 21 29 20
53.067357 54.498210 55.446462 56.169518 56.753754
1 h
c
2 h
c
0.005 0.010 0.015 0.020 0.025
0.02

98 36 93 36 87 35 83 35 79 35
52.576308 52.655217 52.730014 52.800966 52.868637
0.03

98 34 91 33 86 33 82 33 77 32
52.683493 52.761980 52.836322 52.907085 52.974514
0.04

97 32 90 31 85 31 81 31 76 30
52.784094 52.862134 52.936195 53.006699 53.073884
0.05

95 30 89 29 84 29 80 29 76 29
52.878746 52.956645 53.030349 53.100529 53.1673640
0.06 94 28 89 28 84 28 79 27 75 27
52.968191 53.045806 53.119568 53.189435 53.256085
Perishable Inventory System With A Finite..
www.ijmsi.org 14 | P a g e
Table 5: Influence of
w
c and
s
c on the optimal values
















Table 6: Variation in optimal values for different values of
1 p
c and
2 p
c
















VII. CONCLUSIONS
In this paper we consider a finite source two commodity perishable inventory system with
substitutable and retrial demands. This model is most suitable to two different items which are substitutable. The
joint probability distribution for both commodities and number of demands in the orbit is obtained in the steady
state case. Finally, we give numerical examples to illustrate the effect of the parameters on several performance
characteristics.
ACKNOWLEDGMENT
N. Anabzhagans research was supported by the National Board for Higher Mathematics (DAE), Government
of India through research project 2/48(11)/2011/R&D II/1141. K. Jeganathans research was supported by
University Grants Commission of India under Rajiv Gandhi National Fellowship F.16-1574/2010(SA-III).

REFERENCES
[1] Almsi, B., Roszik, J., Sztrik, J., (2005). Homogeneous finite source retrial queues with server subject to breakdowns and
repairs. Mathematical and Computer Modelling, 42, 673 - 682.
[2] Anbazhagan, N., Jinting, W., Gomathi, D., (2013). Base stock policy with retrial demands. Applied Mathematical Modelling,
37, 4464 - 4473.
[3] Anbazhagan, N., Arivarignan, G., (2000). Two-commodity continuous review inventory system with coordinated reorder
policy. International Journal of Information and Management Sciences, 11(3),19 30.
[4] Anbazhagan, N., Arivarignan, G., (2001). Analysis of two-commodity Markovian inventory system with lead time. The Korean
Journal of Computational and Applied Mathematics, 8(2), 427 - 438.
[5] Anbazhagan, N., Arivarignan, G. (2003). Two-commodity inventory system with individual and joint ordering policies.
International Journal of Management and Systems, 19(2), 129 - 144.
[6] Anbazhagan, N., Arivarignan, G., Irle, A., (2012). A two-commodity continuous review inventory system with substitutable
items. Stochastic Analysis and Applications, 30, 129 - 144.
[7] Agarwal, V., (1984). Coordinated order cycles under joint replenishment multi-item inventories. Naval Logistic Research
Quarterly, 131 - 136.
[8] Artalejo, J. R. (1998). Retrial queues with a finite number of sources. Journal of the Korean Mathematical Society, 35, 503 -

w
c
s
c
2 4 6 8 10
8

23 13 56 23 91 30 127 36 164 41
18.666668 35.866303 52.753058 69.441440 85.985889
10

23 17 56 24 90 30 127 36 164 42
18.732696 35.926358 52.808218 69.493279 86.035232
12

22 19 56 25 90 31 127 37 164 42
18.792123 35.984528 52.862134 69.544235 86.084255
14

20 20 55 26 90 32 127 38 164 43
18.845248 36.040683 52.915075 69.594647 86.132520
16 19 21 55 27 90 33 126 38 163 44
18.892830 36.094711 52.966959 69.644419 86.180477

2 p
c
1 p
c
0.2 0.5 0.8 1.1 1.4
2

175 43 167 35 162 29 159 24 159 20
51.737733 52.071328 52.346578 52.577109 52.769276
4

94 39 90 31 87 26 86 23 85 20
52.545497 52.862134 53.126890 53.353545 53.549672
6

63 36 61 29 59 25 58 22 57 19
53.058156 53.368609 53.628758 53.853509 54.049595
8

47 35 45 28 45 27 44 21 43 18
53.431616 53.739166 53.829995 54.221504 54.418675
1.0 37 34 35 27 35 23 34 20 34 18
53.723445 54.029803 54.287583 54.511543 54.700904
Perishable Inventory System With A Finite..
www.ijmsi.org 15 | P a g e
525.
[9] Artalejo, J. R., Krishnamoorthy, A., Lpez-Herrero, M. J., (2006). Numerical analysis of (s, S) inventory systems with repeated
attempts. Annals of Operations Research, 141, 67 - 83.
[10] Artalejo, J. R., Lpez-Herrero, M. J., (2007). A simulation study of a discrete-time multiserver retrial queue with finite
population. Journal of Statistical Planning and Inference, 137, 2536 - 2542.
[11] Ballintify, J.L., (1964). On a basic class of inventory problems. Management Science, 10, 287 297. Falin, G.I., Templeton,
J.G.C., (1997). Retrial Queues. Chapman Hall, New York.
[12] Falin, G.I., Artalejo, J.R., (1998). A finite source retrial queue. European Journal of Operational Research, 108, 409 - 424.
[13] Federgruen, A., Groenvelt, H., Tijms, H.C., (1984). Coordinated replenishment in a multi-item inventory system with compound
Poisson demands. Management Science, 30, 344 - 357.
[14] Goyal, S.K., Satir, T., (1989). Joint replenishment inventory control: deterministic and stochastic models. European Journal of
Operational Research, 38, 2 - 13.
[15] Goyal, S. K., Giri, B. C., (2001). Recent trends in modeling of deteriorating inventory. European Journal of Operational
Research, 34(1), 1 - 16.
[16] Kalpakam, S., Arivarignan, G., (1993). A coordinated multi-commodity (s, S) inventory system. Mathematical and Computer
Modelling, 18, 69 - 73.
[17] Krishnamoorthy, A., Varghese, T.V., (1994). A two commodity inventory problem. International of Information and
Management Sciences, 3, 55 - 70.
[18] Kulkarni, V. G., (1995). Modelling and Analysis of Stochastic Systems. Chapman & Hall.Miller, B.L., (1971). A multi-item
inventory model with joint probability back-order criterian. Operations Research, 1467 - 1476.
[19] Nahmias, S., (1982). Perishable inventory theory : A review. Operations Research, 30, 680 - 708.2Ohno, K., Ishigaki, T., (2001).
A multi-item continuous review inventory system with compound Poisson demands. Mathematical Methods of Operations
Research, 53, 147 - 165.
[20] Raafat, F., (1991). A survey of literature on continuously deteriorating Inventory models. Journal of Operational Research
Society, 42, 27 - 37.
[21] Shah, N. H., Shah, Y. K., (2000). Literature survey on inventory models for deteriorating items.Ekonomski Anali, 44, 221- 237.
[22] Silver, E.A., (1974). A control system for coordinated inventory replenishment. European Journal of Operational Research, 12,
647 - 671.
[23] Sivakumar, B., (2008). Two commodity inventory system with retrial demand. European Journal of Operational Research, 187,
70 - 83.
[24] Sivakumar, B., (2009). A perishable inventory system with retrial demands and a finite population. Journal of Computational and
Applied Mathematics, 224, 29 - 38.
[25] Thomstone, T.M., Silver, E.A., (1975). A coordinated inventory control system. International Journal of Production Research,
581 - 602.
[26] Yadavalli, V.S.S., Anbazhagan, N., Arivarignan, G., (2004). A two-commodity continuous review inventory system with lost
sales. Stochastic Analysis and Applications, 22, 479 - 497.
[27] Yadavalli, V.S.S., De, C., Van Schoor, W., Udayabaskaran, S., (2006). A substitutable two product inventory inventory system
with joint ordering policy and common demand. Applied Mathematics and Computation, 172(2), 1257 - 1271.
[28] Ushakumari P.V., (2006). On (s, S) inventory system with random lead time and repeated demands. Journal of Applied
Mathematics and Stochastic Analysis, 1 - 22.