Journal of Physics: Conference Series

PAPER • OPEN ACCESS

A MAP/PH/1 Production Inventory Model with Perishable Items and

Dependent Retrials
To cite this article: K P Jose and P S Reshmi 2021 J. Phys.: Conf. Ser. 1850 012020

View the article online for updates and enhancements.

This content was downloaded from IP address 213.182.196.99 on 27/07/2021 at 02:07

ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

A MAP/PH/1 Production Inventory Model with

Perishable Items and Dependent Retrials
K P Jose1 and P S Reshmi1
1
P. G. & Research Dept. of Mathematics, St.Peter’s College, Kolenchery - 682 311, Kerala,
India
E-mail: [email protected]

Abstract. This article studies a perishable inventory system with a production unit. The
production process is governed by (s, S) policy and it is exponentially distributed. The primary
arrival follows Markovian arrival process(MAP) and the service time is phase-type distributed
random variable. The inventoried items are subject to decay exponentially with a linear rate. A
newly arriving customer realizes that system is running out of stock or server busy either moves
to infinite waiting space with a pre-assigned probability or exit system with complementary
probability. Customers in the waiting space make retrials to access the free server at a linear
rate. If the system is running out of stock or the server is busy upon retrial, customers go back
to orbit with different pre-defined probabilities according to the level of inventory or exit the
system with corresponding complementary probabilities. The system is analysed using Matrix
Analytic Method(MAM) and the findings are numerically illustrated.

1. Introduction
Baek and Moon [1] analysed an inventory system attached to multiple server queue in which
items are produced internally. The Poisson process characterizes the arrival of demand and
production of items. If the system is out of stock, the unsatisfied customer wait in the system,
and all new arrivals are lost. The proposed model is analysed through the regenerative process
and product form solution is derived for probability vectors. Ravithammal et al. [2] dealt
with a production inventory system of fixed lifetime product. The deterministic demand model
considered both fixed back-orders and linear back-orders. Ioannidis et al. [3] considered a
perishable inventory system with production and impatient customers. Impatient customers
may balk upon arrival or renege while stock out period. The lifetime of items and patience
time of customers are random variables with general distributions. Also, service time and inter-
arrival time are assumed as exponential random variables. Chakraborty et al. [4] introduced a
generalised economic manufacturing quantity model with a production machine which undergoes
breakdown. Arbitrary probability distributions are assigned to time for breakdown, corrective,
and preventive repair. Krishnamoorthy and Narayanan [5] studied a system to which arrivals
according to a Poisson process and the production unit is controlled by (s, S) policy. The server
needs a random time to serve the inventory for demanding customers and which is exponentially
distributed. If the production is on, it adds single item to inventory successively and the time
required to add items is exponentially distributed. A product form solution is obtained by
assuming that the system loses customers during stock out period.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

Krishnamoorthy and Jose [6] distinguished three retrial inventory systems with a production
unit. In all the three models, the demand process follows a Poisson distribution and exponential
production time of items. The unsatisfied customer in each model join the orbit with pre-
determined probability and make retrials to achieve their demands. The inter-retrial and
service time are exponentially distributed. The unsatisfied retrial customers re-join the orbit in
accordance with a pre-assigned value or loss system with complementary probability. Jose and
Nair [7] differentiated two (s, S) production inventory systems with retrial of customers. The
production unit runs at two different rates per cycle according to the inventory level. That is,
the rate is higher during the beginning of a cycle then it is lowered when the inventory level
crosses a pre-defined level. The rate of loss of customers from the system is reduced through
that assumption. All the underlying distributions assumed are exponential. Nair and Jose [8]
proposed another work in which the variation of service rate is considered. The authors changed
the service rate into two different rates; normal rate and reduced rate. The time for producing
a single item follows an exponential distribution.
In this paper, the production inventory model proposed by Nair and Jose [8] is extended
to perishable items and inventory dependent rejoining of customers. In real life situations,
nearly all inventoried items are subject to decay and it leads us to the assumption of random
lifetime of items. The lifetime of each item is distributed as exponential with a linear rate.
Also, the model provides an orbit for retrial facility in which retrial customers rejoin the orbit
depending on the inventory level. In existing models, the lifetime of items stored is assumed as
infinite, and customers are allowed to orbit independent of the inventory level. We assume the
modelling tools as Markovian Arrival Process and Phase type distribution for arrival process and
service time distribution respectively. Thus, Matrix Analytic Method is well suited for analysing
a practical situation considering non exponential inter-arrival and service time distribution.
Also, the method reduces the problem of numerical intractability of the model. The paper is
constructed follows: section 2 model the system mathematically and stability of the system is
discussed in section 3. Section 4 and 5 deal with computation of stationary probability vector
and performance measures respectively. In section 6, a profit function is constructed and results
are numerically illustrated.

2. Mathematical model

Figure 1. Representation of the system

2
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

Consider an inventory system of perishable items governed by continuous review (s, S) policy.
If the inventory diminishes to switch on level s, the production unit is switched on and it adds one
by one item to the inventory. The manufacturing time of an item follows exponential distribution
with rate β and the production stops when the inventory level attains its maximum capacity S.
The primary arrivals follow Markovian arrival process having representation (D0 , D1 ) with m
phases. A single server is availed in the service area and service time is phase type distributed
with representation (η, T ) having n phases. If the server busy or inventory zero upon primary
arrival, customers occupy the infinite waiting space called orbit with probability γ or goes
out of the service area with probability (1 − γ). The unsatisfied primary customers wait
in the orbit and make retrials to access the server. If the inventory is zero upon retrial,
customers reoccupy the orbit with probability p0 . If the server is busy upon retrial with
inventory level j; 1 ≤ j ≤ s, then the customers reoccupy the orbit with probability pj such
that pj > pj−1 (1 ≤ j ≤ s). If the inventory level is greater than switch on level s then the
customers reoccupy the orbit with probability p. In all cases, retrial customers exit the system
with corresponding complementary probabilities. The lifetime of an item and inter-retrial times
of consecutive retrials are exponentially distributed with linear rate jω and iθ respectively.
Assume that the item does not perish when the sever is providing service on the last item left
in the inventory.
• Nt : Number of customers in the system at time t
(
0, server idle at time t
• The server status, Kt =
1, server busy at time t
(
0, production unit is off at time t
• The production status, Jt =
1, production unit is on at time t
• It : Level of inventory at time t
• H1t : arrival phase at time t
• H2t : service phase at time t
Then {Xt |t ≥ 0} is a level dependent quasi birth-death process on the state space E, where
Xt = (Nt , Kt , Jt , It , H1t , H2t ) and E = (i, 0, 0, j) ∪ (i, 0, 1, j) ∪ (i, 1, 0, j) ∪ (i, 1, 1, j) with

(i, 0, 0, j) ={(i, 0, 0, j, k) : i ≥ 0, s + 1 ≤ j ≤ S, 1 ≤ k ≤ m},

(i, 0, 1, j) ={(i, 0, 1, j, k) : i ≥ 0, 0 ≤ j ≤ S − 1, 1 ≤ k ≤ m},
(i, 1, 0, j) ={(i, 1, 0, j, k, l) : i ≥ 0, s + 1 ≤ j ≤ S, 1 ≤ k ≤ m, 1 ≤ l ≤ n},
(i, 1, 1, j) ={(i, 1, 1, j, k, l) : i ≥ 0, 1 ≤ j ≤ S − 1, 1 ≤ k ≤ m, 1 ≤ l ≤ n}.

The generator matrix has the following structure

 
B10 C
A21 B11 C 
 
Q=
 A22 B12 C 


 A23 B13 C 

.. .. ..
. . .

in which each matrix has order (2S − s)m + (2S − s − 1)mn. The entries of A2i , B1i and C are
transitions from the level i to i − 1, i and i + 1 respectively. Transitions are given in table 1,
where

3
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

Table 1. Transitions

From To Rate Description

(i, 0, 1, 0) (i + 1, 0, 1, 0) γD1
(i, 1, 0, j) (i + 1, 1, 0, j) γD1 ⊗ In j = s + 1, s + 2, . . . , S
(i, 1, 1, j) (i + 1, 1, 1, j) γD1 ⊗ In j = 1, 2, . . . , S − 1
(i, 0, 0, j) (i − 1, 1, 0, j) iθ ⊗ η j = s + 1, s + 2, . . . , S
(i, 0, 1, 0) (i − 1, 0, 1, 0) iθ(1 − p0 )Im
(i, 0, 1, j) (i − 1, 0, 1, j) iθ ⊗ η j = 1, 2, . . . , S − 1
(i, 1, 0, j) (i − 1, 1, 0, j) iθ(1 − p)Imn j = s + 1, s + 2, . . . , S
(i, 1, 1, j) (i − 1, 1, 1, j) iθ(1 − pj )Imn j = 1, 2, . . . , S − 1
(i, 0, 0, j) (i, 0, 0, j) D0 − iθIm − jωIm j = s + 1, s + 2, . . . , S
(i, 0, 1, j) (i, 0, 1, j + 1) βIm j = 0, 1, . . . , S
(i, 0, 0, s + 1) (i, 0, 1, s) (s + 1)ωIm
(i, 0, 0, j) (i, 0, 0, j − 1) jωIm j = s + 2, s + 3, . . . , S
(i, 0, 0, j) (i, 1, 0, j) D1 ⊗ η j = s + 1, s + 2, . . . , S
(i, 0, 1, S − 1) (i, 0, 0, S) βIm
(i, 0, 1, j) (i, 0, 1, j) ∆j j = 0, 1, . . . , S − 1
(i, 0, 1, j) (i, 0, 1, j − 1) jωIm j = 1, 2, . . . , S − 1
(i, 0, 1, j) (i, 1, 1, j) D1 ⊗ η j = 1, 2, . . . , S − 1
(i, 1, 0, j) (i, 0, 0, j) Im ⊗ T 0 j = s + 1, s + 2, . . . , S
(i, 1, 0, s + 1) (i, 0, 1, s) Im ⊗ T 0
(i, 1, 0, j) (i, 1, 0, j) ∇j j = s + 1, s + 2, . . . , S
(i, 1, 0, j) (i, 1, 0, j − 1) jωImn j = s + 2, s + 3, . . . , S
(i, 1, 0, s + 1) (i, 1, 1, s) (s + 1)ωImn j = s + 2, s + 3, . . . , S
(i, 1, 1, j) (i, 0, 1, j) Im ⊗ T 0 j = 1, 2, . . . , S − 1
(i, 1, 1, S − 1) (i, 1, 0, S) βImn
(i, 1, 1, j) (i, 0, 1, j) σj j = 1, 2, . . . , S − 1
(i, 1, 1, j) (i, 1, 1, j − 1) jωImn j = 2, 3, . . . , S − 1
(i, 1, 1, j) (i, 0, 1, j) βImn j = 1, 2, . . . , S − 1

(
D0 + (1 − γ)D1 − iθ(1 − p0 )Im − βIm ; j = 0
∆j =
D0 − iθIm − βIm − jωIm ; 1≤j ≤S−1

∇j = D0 + (1 − γ)D1 ⊕ −iθ(1 − pj )Imn − jωImn


0 1
D + (1 − γ)D ⊕ −iθ(1 − p1 )Imn − βImn ;
 j=1
σj = D0 + (1 − γ)D1 ⊕ −iθ(1 − pj )Imn − βImn − jωImn ; 2≤j≤s
 0
D + (1 − γ)D1 ⊕ −iθ(1 − p)Imn − βImn − jωImn ;

s+1≤j ≤S−1

3. Stability of the system

Theorem 3.1. The system under consideration is stable.

4
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

Proof. For any state r in the level i, φ(r) = i be the Lyapunov test function. Then the mean
drift of the state r in level i ≥ 1 is,


−iθ(1 − p0 ) + γ[D1 em ]k , r = (i, 0, 1, 0, k), 1 ≤ k ≤ m




−iθ, r = (i, 0, 0, j, k), s + 1 ≤ j ≤ S,




 1≤k≤m
−iθ, r = (i, 0, 1, j, k), 1 ≤ j ≤ S − 1,



dr = 1≤k≤m
1 e ⊗e ]




 −iθ(1 − p)+[γD m n (k−1)m+l , r = (i, 1, 0, j, k, l), s + 1 ≤ j ≤ S,




 1 ≤ k ≤ m, 1 ≤ l ≤ n


 1
−iθ(1 − pj )+[γD em ⊗en ](k−1)m+l , r = (i, 1, 1, j, k, l), 1 ≤ j ≤ S − 1,


1 ≤ k ≤ m, 1 ≤ l ≤ n, pj = p for j > s


Since (1 − p0 ), (1 − pj ), (1 − p) > 0, for arbitrary  > 0, there exist N 0 such that dr < − for
all r in level i ≥ N 0 . Now, the stability follows from Tweedi’s result [9].

4. Computation of stationary probability vector

Let Φ = (ϕ0 , ϕ1 , ϕ2 , ...) be the stationary probability vector of Q, where each ϕi (i ≥ 0) has
(2S − s)m + (2S − s − 1)mn elements:

ϕi = (ψi,0,0,s+1 , ..., ψi,0,0,S , ψi,0,1,0 , ..., ψi,0,1,S−1 , ψi,1,0,s+1 , ..., ψi,1,0,S , ψi,1,1,1 , ..., ψi,1,1,S−1 );
ψi,0,0,j = (ψi,0,0,j,1 , ..., ψi,0,0,j,m ), s + 1 ≤ j ≤ S
ψi,0,1,j = (ψi,0,1,j,1 , ..., ψi,0,1,j,m ), 0 ≤ j ≤ S − 1
ψi,1,0,j = (ψi,1,0,j,1,1 , ..., ψi,1,0,j,1,n , ..., ψi,1,0,j,m,1 , ..., ψi,1,0,j,m,n ), s + 1 ≤ j ≤ S;
ψi,1,1,j = (ψi,1,1,j,1,1 , ..., ψi,1,1,j,1,n , ..., ψi,1,1,j,m,1 , ..., ψi,1,1,j,m,n ), 1 ≤ j ≤ S − 1

Matrix Analytic method is used to obtain the probability vectors. According to Neuts [10], Φ
satisfies the relation
ϕN +k = ϕN −1 Rk+1 , k ≥ 0
where the matrix R is the minimal non-negative solution of the equation R2 A2 + RB1 + C = 0
with B1 = B1N , A2 = A2N ; N is obtained by Neuts -Rao truncation [11]. For 1 ≤ i ≤ N − 1;
ϕN −i = ϕN −i−1 RN −i , where RN −i = −C(B1,N −i + RN −i+1 A2,N −i+1 )−1 . Now, ϕ0 is obtained
by solving ϕ0 (B10 + R1 A21 ) = 0. The stationary
P probability vector of the truncated system is
then obtained by normalizing each ϕi by ( i ϕi )e.

5. Performance measures
a) Expected Inventory level,

∞ X
S m n ∞ S−1 m n
! !
X X X X XX X
Einv = j ψi,0,0,j,k + ψi,1,0,j,k,l + j ψi,0,1,j,k + ψi,1,1,j,k,l
i=0 j=s+1 k=1 l=1 i=0 j=1 k=1 l=1

b) Expected perishable rate, Ep = ω × Einv

c) Expected switching rate,
∞ X
m n ∞ X
m X
n
!
X X X
Esr =(s + 1)ω ψi,0,0,s+1,k + ψi,1,0,s+1,k,l + ψi,1,0,s+1,k,l Tl0
i=0 k=1 l=1 i=0 k=1 l=1

5
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

d) Expected number of orbiting customers,

∞ S m n
!
X X X X
Eorbit = i ψi,0,0,j,k + ψi,1,0,j,k
i=1 j=s+1 k=1 l=1
 
∞
X S
X m X
X n S−1
XX m X
n
+ i ψi,1,0,j,k,l + ψi,1,1,j,k,l 
i=1 j=s+1 k=1 l=1 j=1 k=1 l=1

e) Expected departures after service,

∞ X
X S m X
X n ∞ S−1
X XX m X
n
Eds = ψi,1,0,j,k,l Tl0 + ψi,1,1,j,k,l Tl0
i=0 j=s+1 k=1 l=1 i=0 j=1 k=1 l=1

f) Expected number of primary loss,

X∞ m X
X m S
X m X
X n X
m
1 1
Ela = (1 − γ) ψi,0,1,0,k Dkk1 + ψi,1,0,j,k,l Dkk1
i=0 k=1 k1 =1 j=s+1 k=1 l=1 k1 =1
S−1 m X
n X
m
!
XX
1
+ ψi,1,1,j,k,l Dkk1
j=1 k=1 l=1 k1 =1

g) Expected number of retrial customer loss,

∞
X m
X S
X m X
X n
Elr = θ i (1 − p0 )ψi,0,1,0,k + (1 − p)ψi,1,0,j,k,l +
i=1 k=1 j=s+1 k=1 l=1
s X
m X
n S−1 m X
n
!
X X X
(1 − pj )ψi,1,1,j,k,l + (1 − p)ψi,1,1,j,k,l
j=1 k=1 l=1 j=s+1 k=1 l=1

h) Overall rate of retrial, θ1∗ = θ × Eorbit

i) Successful rate of retrial,
 
X∞ X S X m ∞ S−1
X XX m
θ2∗ = θ  iψi,0,0,j,k + iψi,0,1,j,k 
i=1 j=s+1 k=1 i=1 j=1 k=1

θ2∗
j) Ratio of successful rate of retrial, θ∗ = .
θ1∗

6. Profit analysis
We define the total profit as

Eprof it = r1 Eds − c1 Esr − c2 Einv − c3 Eorbit − c4 (Ela + Elr ) − c5 Ep ,

where r1 =revenue from unit purchase per unit time, c1 =production unit running cost per unit
per unit time, c2 =inventory holding cost per unit per unit time, c3 =waiting space maintenance
cost per unit per unit time, c4 =penalty cost of customer losing per unit per unit time,
c5 =disposing cost of decayed items per unit per unit time.

6
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

6.1. Numerical illustrations

In this section, we investigate the effect of positive and negative correlations on overall rate of
retrial, successful rate of retrial and ratio of successful retrial. The variation is compared by
making the rejoin probabilities independent of the inventory level, that is, pj = p; for all j(II).
Assume that the maximum inventory level, S = 20 and the production  switch  on level,
 s = 5.
−6 3 3
Further, the remaining parameters are chosen artificially as T = ; T0 = ,η =
1 −4 3
 
0.4 0.6 and the probabilities

p0 = 0.15, p1 = 0.25, p2 = 0.3, p3 = 0.36, p4 = 0.42, p5 = 0.57, p = 0.7 (I)

Again, two types of MAP are given in following table:

D0 D1 Average arrival rate Correlation coefficient

   
−4.6 2.5 1.1 1
MAP(+) 2.52 +0.00076
3 −6 1.0 2
   
−5.49 3.49 1.50 0.5
MAP(−) 2.52 −0.00076
2.12 −5.17 1.95 1.1

If the rejoining of retrial customers to orbit is controlled by probabilities depending on the

inventory level then it will reduce the congestion in the orbit and increases the fraction of
successful retrials. As expected, table 2, 3, 4 and 5 shows that the ratio of successful rate of
retrial θ∗ is greater for (I) compared to (II) having all probabilities pj = p, for all j in both
MAP(+) and MAP(−). That is, rejoining of retrial customers to orbit in accordance with
probabilities depending on inventory level makes retrial more successful.

Table 2. Variation of θ on θ∗

MAP(+)
θ I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
1.9 1.7100 0.1635 0.0956 3.5819 0.2326 0.0649
2.0 1.7088 0.1612 0.0944 3.5856 0.2299 0.0641
2.1 1.7077 0.1590 0.0931 3.5891 0.2272 0.0633
2.2 1.7066 0.1569 0.0919 3.5926 0.2246 0.0625
2.3 1.7056 0.1548 0.0907 3.5960 0.2221 0.0618
MAP(−)
θ I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
1.9 1.6996 0.1617 0.0952 3.5671 0.2309 0.0647
2.0 1.6984 0.1595 0.0939 3.5705 0.2283 0.0639
2.1 1.6972 0.1574 0.0927 3.5737 0.2257 0.0631
2.2 1.6961 0.1553 0.0916 3.5770 0.2231 0.0624
2.3 1.6950 0.1532 0.0904 3.5801 0.2207 0.0616
β = 1.9, ω = 0.6, γ = 0.8.

7
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

Table 3. Variation of β on θ∗

MAP(+)
β I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
1.7 1.7468 0.1307 0.0748 3.8069 0.1882 0.0494
1.8 1.7262 0.1447 0.0838 3.6949 0.2076 0.0562
1.9 1.7077 0.1590 0.0931 3.5891 0.2272 0.0633
2.0 1.6913 0.1737 0.1027 3.4895 0.2470 0.0708
2.1 1.6769 0.1886 0.1124 3.3958 0.2669 0.0786
MAP(−)
β I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
1.7 1.7376 0.1294 0.0745 3.7935 0.1869 0.0493
1.8 1.7164 0.1432 0.0834 3.6805 0.2062 0.0560
1.9 1.6972 0.1574 0.0927 3.5737 0.2257 0.0631
2.0 1.6802 0.1719 0.1023 3.4731 0.2453 0.0706
2.1 1.6652 0.1866 0.1121 3.3785 0.2651 0.0785
ω = 0.6, γ = 0.8, θ = 2.1.

Table 4. Variation of ω on θ∗

MAP(+)
ω I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
0.4 1.6828 0.2083 0.1238 3.3799 0.2830 0.0837
0.5 1.6936 0.1801 0.1063 3.4932 0.2517 0.0721
0.6 1.7077 0.1590 0.0931 3.5891 0.2272 0.0633
0.7 1.7228 0.1427 0.0828 3.6721 0.2075 0.0565
0.8 1.7378 0.1296 0.0746 3.7452 0.1913 0.0511
MAP(−)
ω I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
0.4 1.6707 0.2057 0.1231 3.3638 0.2807 0.0834
0.5 1.6825 0.1781 0.1058 3.4775 0.2498 0.0718
0.6 1.6972 0.1574 0.0927 3.5737 0.2257 0.0631
0.7 1.7128 0.1413 0.0825 3.6571 0.2062 0.0564
0.8 1.7281 0.1285 0.0743 3.7305 0.1901 0.0510
β = 1.9, γ = 0.8, θ = 2.1.

8
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

Table 5. Variation of γ on θ∗

MAP(+)
γ I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
0.5 1.0108 0.1074 0.1063 1.9684 0.1601 0.0813
0.6 1.2361 0.1257 0.1017 2.4736 0.1851 0.0748
0.7 1.4685 0.1429 0.0973 3.0145 0.2074 0.0688
0.8 1.7077 0.1590 0.0931 3.5891 0.2272 0.0633
0.9 1.9534 0.1741 0.0891 4.1954 0.2445 0.0583
MAP(−)
γ I II
θ1∗ θ2∗ θ∗ θ1∗ θ2∗ θ∗
0.5 1.0036 0.1059 0.1055 1.9582 0.1583 0.0808
0.6 1.2277 0.1241 0.1011 2.4614 0.1833 0.0745
0.7 1.4590 0.1412 0.0968 3.0005 0.2058 0.0686
0.8 1.6972 0.1574 0.0927 3.5737 0.2257 0.0631
0.9 1.9420 0.1726 0.0889 4.1790 0.2431 0.0582
β = 1.9, ω = 0.6, θ = 2.1.

6.2. Optimization of total profit

Here, a numerical optimization of the total profit is carried out using a three dimensional
plot. In Figure 2, the Eprof it is plotted
 for various
 combination
 of s and S. For that,
 fix
−5.49 3.49 1.50 0.5 −6 3
the other parameters value as D0 = , D1 = ,T = , T0 =
2.12 −5.17 1.95 1.1 1 −4
 
3  
, η = 0.4 0.6 , β = 2.5, ω = 0.4, θ = 2, γ = 0.8, p = 0.8 and different costs assumed are
3
r1 = 1000, c1 = 105, c2 = 5; c3 = 10, c4 = 2, c5 = 3. Again, the probabilities pj ; (0 ≤ j ≤ s) for
various s are given below:

s and corresponding probabilities

s = 3; p0 = 0.15, p1 = 0.25, p2 = 0.30, p3 = 0.45
s = 4; p0 = 0.15, p1 = 0.25, p2 = 0.30, p3 = 0.40, p4 = 0.51
s = 5; p0 = 0.15, p1 = 0.25, p2 = 0.30, p3 = 0.40, p4 = 0.50, p5 = 0.60
s = 6; p0 = 0.12, p1 = 0.21, p2 = 0.29, p3 = 0.36, p4 = 0.45, p5 = 0.60, p6 = 0.70
s = 7; p0 = 0.13, p1 = 0.21, p2 = 0.27, p3 = 0.35, p4 = 0.44, p5 = 0.56, p6 = 0.64, p7 = 0.72
s = 8; p0 = 0.12; p1 = 0.22, p2 = 0.29, p3 = 0.35, p4 = 0.43, p5 = 0.50, p6 = 0.57, p7 = 0.64,p8 = 0.75

From figure 2, the optimum value of Eprof it = 1675.6 is obtained at s = 6, S = 21. Therefore
the optimal (s, S) pair is (6,21).

Concluding remarks
In this model, we considered a MAP/PH/1 perishable inventory system with a production unit
presided by (s, S) policy. The unsatisfied primary customers move to orbit for further retrials

9
ICMMCMSE 2020 IOP Publishing
Journal of Physics: Conference Series 1850 (2021) 012020 doi:10.1088/1742-6596/1850/1/012020

Figure 2. Variation of Eprof it on (s, S)

with pre-determined probability. The retrial customers join back to the orbit with different
probabilities in accordance with items present in the inventory. The lifetime of inventoried items
and inter-retrial times of customers in the orbit is characterized by independent exponential
distributions in linear rate. A profit function was constructed and the results were numerically
studied. One can extend this work by considering the Batch Markovian Arrival Process(BMAP)
instead of MAP. Further, another way of extension is possible if the random lifetime of the item
is replaced by a common lifetime.

References
[1] Baek J W and Moon S K 2016 Journal of the Korean Statistical Society 45 14–24
[2] Ravithammal M, Uthayakumar R and Ganesh S 2014 International Journal of Mathematical Analysis 8
1549–1559
[3] Ioannidis S, Jouini O, Economopoulos A A and Kouikoglou V S 2013 Annals of Operations Research 209
115–138
[4] Chakraborty T, Giri B C and Chaudhuri K S 2008 European Journal of Operational Research 185 606–618
[5] Krishnamoorthy A and Viswanath N C 2013 European Journal of Operational Research 228 358–366
[6] Krishnamoorthy A and Jose K P 2008 Int. J. Inf. Manag. Sci 19 367–389
[7] Jose K P and Nair S S 2017 Journal of Industrial Engineering International 13 369–380
[8] Nair S S and Jose K P Palestine Journal of Mathematics 8 328–347
[9] Tweedie R L 1975 Sufficient conditions for regularity, recurrence and ergodicity of markov processes
Mathematical Proceedings of the Cambridge Philosophical Society vol 78 (Cambridge Univ Press) pp 125–
136
[10] Neuts M F 1981 Johns Hopkins University, Baltimore
[11] Neuts M F and Rao B M 1990 Queueing systems 7 169–189

10